DNA2.2.2 - Training Plan - CERN...INFSO-RI-261611 2010-2011 © Members of EMI collaboration PUBLIC 1 / 54 EUROPEAN MIDDLEWARE INITIATIVE. DNA2.2.2 - T. RAINING . P. LAN. EU DELIVERABLE:

INFSO-RI-261611 2010-2011 © Members of EMI collaboration PUBLIC 1 / 54

EUROPEAN MIDDLEWARE

INITIATIVE

DNA2.2.2 - TRAINING PLAN

EU DELIVERABLE: D2.2.2

Document identifier: EMI-DNA2.2.2-1277577-Training_Plan-v1.0.doc

Date: 31/10/2011

Activity: NA2

Lead Partner: TCD

Document status: Final

Document link: http://cdsweb.cern.ch/record/1277577

Abstract: This document presents the EMI training plan including the types of training, the training targets, the organization of training events, and the expected collaboration with other projects.

DNA2.2.2 - TRAINING PLAN Doc. Identifier: EMI-DNA2.2.2-1277577-Training_Plan-v1.0.doc Date: 31/10/2011

EMI RI-261611 © Members of the EMI collaboration PUBLIC 2 / 54

Copyright notice:

Copyright (c) Members of the EMI Collaboration. 2010-2011.

See http://www.eu-emi.eu/about/Partners/ for details on the copyright holders.

EMI (“European Middleware Initiative”) is a project partially funded by the European Commission. For more information on the project, its partners and contributors please see http://www.eu-emi.eu.

This document is released under the Open Access license. You are permitted to copy and distribute verbatim copies of this document containing this copyright notice, but modifying this document is not allowed. You are permitted to copy this document in whole or in part into other documents if you attach the following reference to the copied elements: "Copyright (C) 2010-2011. Members of the EMI Collaboration. http://www.eu-emi.eu ".

The information contained in this document represents the views of EMI as of the date they are published. EMI does not guarantee that any information contained herein is error-free, or up to date.

EMI MAKES NO WARRANTIES, EXPRESS, IMPLIED, OR STATUTORY, BY PUBLISHING THIS DOCUMENT.



Delivery Slip

Name Partner / Activity Date Signature

From Kathryn Cassidy TCD/NA2 22/10/2011

Reviewed by Florida Estrella Shiraz Memon

CERN/NA1 JUELICH/JRA1 05/12/2011

Approved by PEB 05/12/2011

Document Log

Issue Date Comment Author / Partner 1 28/09/2011 Table of Contents Kathryn Cassidy/TCD

2 22/10/2011 v0.5 for review Kathryn Cassidy/TCD

3 05/12/2011 v0.8 final version Kathryn Cassidy/TCD

4 19/12/2011 v1.0 PEB approved version PEB

Document Change Record

Issue Item Reason for Change 1

2

3





TABLE OF CONTENTS1. INTRODUCTION ......................................................................................................................................... 7

1.1. PURPOSE .................................................................................................................................................. 7 1.2. DOCUMENT ORGANIZATION ............................................................................................................... 7 1.3. REFERENCES ............................................................................................................................................ 7 1.4. DOCUMENT AMENDMENT PROCEDURE ..................................................................................................... 8 1.5. TERMINOLOGY ......................................................................................................................................... 8

2. EXECUTIVE SUMMARY ......................................................................................................................... 10 3. TRAINING STRATEGY ............................................................................................................................ 11

3.1. USERS .................................................................................................................................................... 11 3.2. GOALS ................................................................................................................................................... 12

4. STATUS REPORT ON TRAINING PLAN .............................................................................................. 13 4.1. IN-REACH TRAINING .............................................................................................................................. 13

4.1.1 Second In-reach Training Session ............................................................................................. 13 4.2. RESOURCE ADMINISTRATOR AND END-USER TRAINING ........................................................................ 14

4.2.1 GridKa School 2010 .................................................................................................................. 14 4.2.2 EMI Technical Conference 2011 ............................................................................................... 14 4.2.3 EU Joint DCI Summer School 2011 .......................................................................................... 15 4.2.4 GridKa School 2011 .................................................................................................................. 16 4.2.5 EGI Technical Forum 2011 ....................................................................................................... 16 4.2.6 Summary of Feedback Results ................................................................................................... 17

4.3. ONLINE DOCUMENTATION AND TRAINING MATERIALS ......................................................................... 17 4.3.1 Documentation ........................................................................................................................... 17 4.3.2 Online Tutorials ......................................................................................................................... 18

4.4. TRAINING COLLABORATIONS ................................................................................................................. 18 4.4.1 European Grid Infrastructure .................................................................................................... 18 4.4.2 GridKa School ........................................................................................................................... 19 4.4.3 DCI Summer School ................................................................................................................... 19

5. MEASURING SUCCESS AND METRICS .............................................................................................. 20 5.1.1 Measuring effectiveness of training activities ............................................................................ 20 5.1.2 Use of existing data .................................................................................................................... 21 5.1.3 The holistic “EMI” view, as opposed to several middleware stacks ......................................... 21

5.2. SPECIFIC METRICS ................................................................................................................................. 21 6. COVERAGE OF EMI COMPONENTS IN TRAINING ........................................................................ 22 7. FUTURE PLANS ......................................................................................................................................... 24

7.1. ONLINE DOCUMENTATION AND TRAINING MATERIALS ......................................................................... 24 7.2. IN-REACH TRAINING .............................................................................................................................. 24 7.3. USER TRAINING ..................................................................................................................................... 24

7.3.1 Follow-up survey ....................................................................................................................... 26 7.4. TRAINING COLLABORATIONS AND DISSEMINATION OF MATERIALS ...................................................... 26 7.5. TRAINING INFRASTRUCTURE .................................................................................................................. 26 7.6. LIST OF SPECIFIC PLANNED ACTIVITIES ................................................................................................. 27

8. CONCLUSIONS .......................................................................................................................................... 29 ANNEX A: USER FEEDBACK SURVEY ........................................................................................................ 30 ANNEX B: DETAILED IN-REACH TRAINING FEEDBACK ..................................................................... 31

FEEDBACK ON GUIDELINES FOR USAGE OF TOOLS USED BY THE EMI PROJECT (E.G. ETICS) ......................... 31



FEEDBACK ON USING THE EMI TESTBED ........................................................................................................ 33 FEEDBACK ON CERTIFICATION, CHANGE MANAGEMENT AND RELEASE MANAGEMENT POLICIES WITHIN THE

EMI PROJECT ......................................................................................................................................... 34 ANNEX C: DETAILED USER TRAINING FEEDBACK .............................................................................. 37

FEEDBACK ON EMI TUTORIAL FOR RESOURCE ADMINISTRATORS 1 .............................................................. 37 FEEDBACK ON EMI TUTORIAL FOR RESOURCE ADMINISTRATORS 2 .............................................................. 38 FEEDBACK ON EMI TUTORIAL FOR RESOURCE ADMINISTRATORS 3 .............................................................. 39 FEEDBACK ON EMI TUTORIAL FOR RESOURCE ADMINISTRATORS 4 .............................................................. 40 FEEDBACK ON EMI TUTORIAL FOR USERS ..................................................................................................... 41 FEEDBACK ON DCISS EMI TRAINING DAY ................................................................................................... 43 FEEDBACK ON EGI TECHNICAL FORUM 2011 ................................................................................................ 46

ANNEX D: OFFICIAL DCI SUMMER SCHOOL REPORT ......................................................................... 48 THE DISTRIBUTION OF THE STUDENTS/ EACH PROJECT .................................................................................... 51 OVERALL EVALUATION .................................................................................................................................. 51 EVALUATION PER DAY: .................................................................................................................................. 53 FURTHER COMMENTS: .................................................................................................................................... 53



1. INTRODUCTION

1.1. PURPOSE This document presents the first year update on the training activities for the European Middleware Initiative with respect to the initial DNA2.2.1 Training Plan (M6) [R1], and includes an updated plan for the coming year.

1.2. DOCUMENT ORGANIZATION Chapter Two provides an executive summary of this deliverable. Chapter Three outlines the overall training strategy of EMI, while Chapter Four reviews progress to date, with respect to the initial DNA2.2.1 Training Plan (M6) [R1]. Chapter Five analyses the success of the training activities along with the target and achieved metrics for the activity. Chapter six looks at the coverage of EMI components by the training to-date. Finally, Chapter 7 presents some conclusions. Annex A contains the training feedback form, while the results of the user feedback are presented in Annexes B and C. Annex D gives the report of the DCI Summer School organizers.

1.3. REFERENCES

R1 DNA2.2.1 Training Plan (M6) http://cdsweb.cern.ch/record/1277575/files/EMI-DNA2.2.1-1277575-Training_Plan-v1.0.pdf

R3 EMI in-reach training http://www.eu-emi.eu/in-reach-training

R4 Second EMI in-reach event http://indico.cern.ch/conferenceDisplay.py?confId=127278

R5 GridKa School http://gridka-school.scc.kit.edu/

R6 GridKa School 2010 http://gridka-school.scc.kit.edu/2010/

R7 EGI User Forum 2011 http://uf2011.egi.eu/

R8 EMI User Tutorial (DCI Summer School 2011) https://twiki.cern.ch/twiki/bin/view/EMI/DCISSUserTutorial

R9 EMI Integrating practical exercise https://twiki.cern.ch/twiki/bin/view/EMI/DCISSIntegratingPractical

R10 GridKa School 2011 http://gridka-school.scc.kit.edu/2011/

R11 EMI User Tutorial (EGI Technical Forum 2011) https://twiki.cern.ch/twiki/bin/view/EMI/EGITF2011UserTutorial

R12 EMI User Tutorial http://www.eu-emi.eu/user-tutorial

R13 EMI Documentation http://www.eu-emi.eu/documentation

http://cdsweb.cern.ch/record/1277575/files/EMI-DNA2.2.1-1277575-Training_Plan-v1.0.pdf

http://www.eu-emi.eu/in-reach-training

http://www.eu-emi.eu/in-reach-training

http://indico.cern.ch/conferenceDisplay.py?confId=127278

http://gridka-school.scc.kit.edu/

http://gridka-school.scc.kit.edu/2010/

http://uf2011.egi.eu/

https://twiki.cern.ch/twiki/bin/view/EMI/DCISSUserTutorial

https://twiki.cern.ch/twiki/bin/view/EMI/DCISSIntegratingPractical

http://gridka-school.scc.kit.edu/2011/

https://twiki.cern.ch/twiki/bin/view/EMI/EGITF2011UserTutorial

http://www.eu-emi.eu/user-tutorial

http://www.eu-emi.eu/documentation



R14 SA2 Policies https://twiki.cern.ch/twiki/bin/view/EMI/EMISa2DocumentationPolicy

R15 EMI Integrating practical exercise http://www.eu-emi.eu/integrating-practical

R16 NeSC http://www.nesc.ac.uk/

R17 ISSGC http://www.issgc.org/

R18 NA2 Training Topics https://twiki.cern.ch/twiki/bin/view/EMI/EmiNa2T3TrainingTopics

R19 EGI Training Working Group https://wiki.egi.eu/wiki/Training_Working_Group

R20 EGI Training Market Place http://www.egi.eu/user-support/training_marketplace/

R21 2nd EMI Technical Conference/ EGI Community Forum 2012 http://cf2012.egi.eu/

R22 FutureGrid http://www.futuregrid.org

R23 DCI Summer School 2011 http://www.lpds.sztaki.hu/eudciss2011/

R24 EGI Technical Forum 2011 http://tf2011.egi.eu/

1.4. DOCUMENT AMENDMENT PROCEDURE This document can be amended by the authors further to any feedback from other teams or people. Minor changes, such as spelling corrections, content formatting or minor text re-organisation not affecting the content and meaning of the document can be applied by the authors without peer review. Other changes must be submitted to peer review and to the EMI PEB for approval.

When the document is modified for any reason, its version number shall be incremented accordingly. The document version number shall follow the standard EMI conventions for document versioning. The document shall be maintained in the CERN CDS repository and be made accessible through the OpenAIRE portal.

1.5. TERMINOLOGY CDS CERN Document Server

CERN European Organisation for Nuclear Research

EGI European Grid Infrastructure

EMI European Middleware Initiative

EMI 1 EMI Middleware release version 1

ESFRI European Strategy Forum on Research Infrastructures

https://twiki.cern.ch/twiki/bin/view/EMI/EMISa2DocumentationPolicy

http://www.eu-emi.eu/integrating-practical

http://www.nesc.ac.uk/

http://www.issgc.org/

https://twiki.cern.ch/twiki/bin/view/EMI/EmiNa2T3TrainingTopics

https://wiki.egi.eu/wiki/Training_Working_Group

http://www.egi.eu/user-support/training_marketplace/

http://cf2012.egi.eu/

http://www.futuregrid.org/

http://www.lpds.sztaki.hu/eudciss2011/

http://tf2011.egi.eu/



ETICS e-Infrastructure for Testing, Integration and Configuration of Software

NA2 Networking Activity 2: Outreach and Collaborations

NeSC National e-Science Centre

NGI National Grid Initiative

OGF Open Grid Forum

PEB Project Executive Board

QA Quality Assurance

SA2 Service Activity 2: Quality Assurance

VRC Virtual Research Community (within EGI)



2. EXECUTIVE SUMMARY

NA2 Outreach and Collaborations has conducted a range of training-related activities in line with the initial training plan presented in DNA2.2.1. An online in-reach training event was held in February - March 2011 to ensure that project members had the necessary skills and knowledge about the tools and processes of the project to prepare for the EMI 1 Kebnekaise release. A number of user-focused training events were also held covering training for end-users and for resource administrators. These include a total of ten hands-on training sessions, as well as over six training presentations, delivered at five separate events.

Feedback was collected, where possible, from participants in training events and this has been used to improve subsequent events. The overall feedback scores have been very positive with every event meeting or exceeding the target feedback score of 4.

Documentation forms the starting point for many new EMI users and also for the creation of more elaborate online training materials. A new searchable database of documentation has been put in place containing documentation reviewed by NA2 and SA2 which complies with the EMI documentation requirements. This is available on the EMI website allowing users to easily find the documentation they need.

Two user tutorials, one covering the majority of the EMI user-interface commands, and another more complex real-life exercise, have also been created, and these along with recordings of the on-line in-reach training sessions are available on the EMI website. These materials are being constantly improved and additional tutorials are in development.

NA2 pursues collaborations with several external projects and entities in order to increase the reach of the EMI training activities. A number of external training events have been supported over the past year.

While these activities have all been very successful, NA2 is attempting to improve future training activities by surveying members on in-reach training requirements, performing follow-up surveys for end-user training, making better use of the data collected in feedback surveys, and gradually moving from training on separate middleware products to a more holistic view of a single EMI middleware.

Training has so far been delivered on a range of EMI components, and the components covered will increase in the coming period.

A number of future training events are being planned including an additional in-reach training event in December 2011. Further user training events are also planned including two summer schools run in collaboration with other projects, and workshops at events such as the 2nd EMI Technical Conference/EGI Community Forum 2012, OGF34, the International Supercomputing Conference 2012, and others. Training topics will include the basic user and admin training which has been provided to date but will also be expanded to include more developer-specific training, as well as training on new features of the Matterhorn and Monte Bianco releases, or of update releases where appropriate.

Plans also exist for and continuous improvement of the online documentation and training materials.A training infrastructure is in planning which will ensure that training events can be better supported in future without fear of technical issues which can occur when using the existing testbed. A full list of planned training-related activities is given in Table 5 at the end of this document.



3. TRAINING STRATEGY

As outlined in the description of work, NA2 will coordinate a comprehensive training program organized in three main levels:

• training and knowledge transfer activities within the EMI project itself or “in-reach”

• user training on EMI technologies for a wider audience

• on-line training courses and self-paced training activities

The EMI project aims to coordinate and integrate several different middleware products to produce a unified European middleware distribution. This makes training and knowledge transfer potentially more complicated than might be the case in other projects focused on a single product. There are a total of 29 distinct product teams, managing roughly 100 components in the project. Each team works on their own software, but also has a need to understand the interactions with a range of other products developed in other product teams.

The Technical Knowledge Management task in EMI NA2 must, therefore, not only focus on training end-users of the software, but a significant effort must be put into in-reach, or knowledge transfer within the project, in order to ensure that the developers understand how their software will interact with the applications from other product teams and other middleware solutions.

User training in EMI aims both to increase the user base of the EMI product suite and to train existing users, introducing new features and increasing the depth of their knowledge of the products. It therefore incorporates induction level courses and materials aimed at new users as well as more in-depth training. All materials produced must be kept up-to-date to ensure that users can update their skills when new versions of software are made available. Training sessions will include hands-on workshops as well as lectures and presentations.

It should be noted at the outset that the training effort within the EMI project is not particularly large by comparison with projects such as EGEE in the past. It is not envisaged, therefore, that EMI will directly run a significant amount of user training.

Instead, NA2 will coordinate training activities, and facilitate technical areas in organizing their own training activities. NA2 will also edit and publish training materials to the wiki and website based on materials supplied by the technical areas.

All project partners should aim to have an involvement in training, whether that takes the form of participating in in-reach discussions, giving formal talks and training workshops to project members or end-users, or production of manuals and training materials which can be adapted by the NA2 work package to form part of the repository of self-paced on-line training materials. Apart from this, good documentation of the products itself are expected to be provided by the product teams of EMI.

At the same time, NA2 will pursue collaborations with other projects and leverage the training resources of projects such as EGI (and through them the National Grid Infrastructure (NGI) of each partner country), Schools of Computing, User Communities and others to run training on the EMI middleware. To facilitate this, NA2 will publish training materials to public repositories and encourage other projects to make use of these.

3.1. USERS User training for EMI needs to consider two main categories of users



• End users – researchers and members of the scientific communities who will use resources running the EMI middleware to do their jobs

• Technical users – technical experts within the user communities, site managers and system administrators who manage infrastructures running EMI software, and application developers writing software which relies on EMI middleware components

The main focus will be on this second category of users. This decision has been taken due to practical constraints (it is simply impossible for EMI to undertake to train all user communities due to the limited training resources available) and because it is the application developers, system administrators and site managers who are seen as the main customers of EMI, due to its nature as a middleware provider for distributed computing infrastructure.

It is envisaged that the wider user community will be reached through the project's collaborations with other partners. For example EGI has a role in coordinating training across the NGIs and has developed a program of accredited Virtual Research Communities (VRCs) who are the scientific communities using the EGI infrastructure, and EMI will work with EGI to ensure that the training which is provided to these users meets our aims.

3.2. GOALS The main goal of EMI training activities is to enlarge the base of expert developers and engineers knowledgeable in EMI technologies through comprehensive training programs and materials. It is translated to the following tangible objectives:

1. Produce training materials and resources for the EMI product suite based on the documentation produced by the product teams, at the same time changing the granularity of training from “middleware training” to “product training”

2. Facilitate knowledge transfer within the EMI project by coordinating in-reach events and producing in-reach training materials

3. Run a limited number of user-training events, focused mainly on communicating changes in the middleware to the application developer, system administrator and site manager communities

4. Disseminate EMI training materials to trainers within the user communities such as EGI and the VRCs, and collaborate with external projects to have the EMI products included in their training activities



4. STATUS REPORT ON TRAINING PLAN

4.1. IN-REACH TRAINING In-reach is the term used to indicate training activities aimed at the members of the EMI project. Two sessions have been held so far, with an additional one per project year planned. The first in-reach training session took place at the EGI Technical Forum on the 16th September 2010. This event was described in the initial version of this deliverable DNA2.2.1.

4.1.1 Second In-reach Training Session Deliverable DNA2.2.1 outlined plans to hold an online in-reach training event. In January 2011 NA2 produced a survey which was sent to all EMI members to determine the in-reach training needs of the project.

The proposed topics were as follows:

1. Obtaining, Installing and Using the EMI middleware / products

2. Guidelines for usage of tools used by the EMI project (e.g. ETICS)

3. Using the EMI Infrastructure

4. Certification, change management and release management guidelines

5. QA Process

Respondents indicated relatively high levels of interest in all topics.

Other questions asked in the survey related to the mode of training delivery, with respondents indicating that a half-day on-line training format would be optimal.

EMI NA2 and SA2 held a joint “in-reach” training event at the end of February and beginning of March 2011. This event was aimed at developers within the EMI project and covered a range of topics relevant for the EMI 1 release.

The event took place from 24th February to 3rd March 2011, with training consisting of three two-hour sessions spread across these days:

1. Thursday Feb 24th: Guidelines for usage of tools used by the EMI project

2. Tuesday March 1st: Using the EMI testbed

3. Thursday March 3rd: EMI Release, Change, Testing, Documentation and Certification Policies

The agenda for the session is available on Indico [R4]. Recordings of the sessions were made and are available online via the project website [R3].

The sessions were well attended with approximately 30 participants per session. A feedback survey was circulated for each session. Participants were asked to rate various aspects of the course, as well as the advertising and registration process for the course (which was done via Indico) and the learning environment used (Adobe Connect). An example of the type of survey form used is given in Annex A.

The feedback on the training was very positive with the participants indicating that it was useful to them and would help them in their work. The average rating given to the sessions was very high, with all sessions scoring above the target metric of 4 out of 6. The average feedback ratings for each course are given in Table 1.



Training Session Event Location Date Feedback rating

Guidelines for usage of tools used by the EMI project SA2 Training Online Feb 2011 4.81

Using the EMI testbed SA2 Training Online March 2011 4.03

EMI Release, Change, Testing, Documentation and Certification Policies

SA2 Training Online March 2011 5.03

Overall Average Feedback Score 4.62 Table 1: In-reach Training Average Feedback Scores

Participants were overall very satisfied with the content, duration and method of delivery of the training. Comments indicated only a few minor problems with the online presentation mode and the learning environment, but did not suggest any preference for face-to-face sessions in the future.

4.2. RESOURCE ADMINISTRATOR AND END-USER TRAINING NA2 has organized and supported a number of training events including those focused at both resource administrators and end-users.

Where possible participants were asked to complete a short survey in order to measure their satisfaction with the training and to solicit comments and feedback which might help improve future training events.

4.2.1 GridKa School 2010 The GridKa School [R5] is hosted annually by Steinbuch Centre for Computing (SCC) of Karlsruhe Institute of Technology (KIT). It is organized by KIT, the Swiss National Grid Initiative (SwiNG) and the Helmholtz Alliance 'Physics at the Terascale'. The school offers a broad range of topics, including training on several EMI components, including ARC (covered for the first time in 2010), dCache, gLite and UNICORE. The agenda for the 2010 school is available on the GridKa school website [R6].

In 2010 NA2 made contact with the GridKa School organizers to collaborate in order to ensure that the school reflected the needs of the EMI project. As part of this collaboration many of the training materials used in the school were reviewed by NA2 and the EMI technical training contacts. EMI were also asked to provide a speaker on the topic of security, and several EMI project members attended the school as tutors for the hands-on parts of the curriculum.

As this event was not organized directly by EMI it was not possible to collect participant feedback using the standard Feedback form. However the comments received from the school organizers were very favorable and EMI was again involved in the GridKa School 2011 as described later in this document.

4.2.2 EMI Technical Conference 2011 A number of training events were organized at the EMI Technical Conference / EGI User Forum 2011 [R7]. These included four training workshops aimed at resource administrators, and one aimed at end-users. This training coincided with the release of EMI 1 Kebnekaise and introduced the new features of the release, with particular emphasis on the changes in the installation and configuration processes, as well as introducing the full range of EMI user client tools to participants who may



already have some experience with one of the component middleware stacks. The sessions were as follows:

1. EMI Tutorial for Resource Administrators 1

o What changes with EMI 1?

o Installation and configuration of CREAM


o Installation and Configuration of UNICORE


o Installation and Configuration of dCache

o Installation and Configuration of StoRM


o Installation and Configuration of A-REX

5. EMI Tutorial for Users

o Overview of user-client usage: ARC, gLite, UNICORE and dCache

The hands-on training took place on the EMI testbed resources accessed via a UI machine hosted by INFN in Catania. Users were issued with short-lived GILDA certificates for authentication and authorization.

The sessions were very well attended, and user feedback was collected for each session via an end of course survey. The feedback received was again very positive with participants indicating that the material was useful and interesting, the practical exercises were appreciated and the support from the tutors was helpful. The main negative comments were about the network connectivity and the time available for the exercises, and resultant lack of depth of some sections. Some useful feedback was also received on specific sections of the tutorial which has been used to improve those sections for future tutorials.

The full feedback is presented in Annex C.

4.2.3 EU Joint DCI Summer School 2011 In March 2011 members of the EGI Training Workgroup agreed to run an EU DCI Summer School [R23] which would be organized jointly by a number of DCI projects including EMI, IGE, EDGI, DEGISCO, StratusLab, EGI-InSPIRE, as well as local organizers MTA SZTAKI. The school was held from 11th to 16th July 2011 in Budapest, Hungary, and the first day of the school was devoted to the EMI middleware stack.

Three EMI members participated on the school's Programme Committee and EMI was involved in influencing the direction and curriculum of the school as a whole, as well as the details of the EMI training day.

The programme of the EMI day was as follows:

• Overview of EMI middleware

• Advanced EMI topics

• Hands-on: Compute & Data

• Hands-on: Advanced Workflow (Integrating Practical)



• Wrap-up and look ahead

The EMI middleware was introduced, some advanced topics including component installation were then covered in the “Advanced EMI topics” presentation, and in the afternoon two hands-on sessions were held which first introduced the user client commands and then asked participants to complete a realistic scientific data analysis task using the EMI tools.

The material for the first hands-on session built upon on material presented in the EMI Tutorial for Users at the EMI Technical Conference in Vilnius. This material was updated and edited for clarity and to address deficiencies highlighted by feedback from the previous session. The instructions used for this first hands-on session are available on the project Wiki at [R8].

The second hands-on session presented a modified version of an exercise created by the National eScience Centre in Edinburgh as part of the International Summer School of Grid Computing (ISSGC) programme. This involved using a Java application to search for patterns in a data file. The entire search space was too large to analyze in a single job and thus learners had to submit a parameter-sweep-style job using their chosen EMI client tools. Hints to narrow down the search were stored on a Storage Element and had to be retrieved using the gLite data commands. The results had then to be uploaded to another SE using dCache to be viewed by the class via the dCache WebDAV interface. The exercise thus integrated use of several different EMI technologies and showed learners how each complemented the others. The instructions for the advanced practical exercise can be found on the project Wiki at [R9].

The hands-on training took place on the EMI testbed resources accessed via a UI machine hosted by INFN in Catania. Users were issued with short-lived GILDA certificates for authentication and authorization.

The event was attended by 36 students from 11 countries and 17 separate institutes. Feedback on the EMI training was very positive, with an average satisfaction rating of 4.56. Comments were also favorable, with participants indicating that they found the exercises interesting and enjoyable as well as informative. Some participants found that there was not enough time to complete the exercises, and a few suggested that there was too great a jump in difficulty between the first and second tutorial sessions. The full feedback is presented in Annex C.

The school as a whole was a success with general agreement that it should be run again next year if possible. EMI has already committed to be involved in next year's school. The official report on the school produced by the local organizers SZTAKI is attached in Annex D.

4.2.4 GridKa School 2011 The GridKa School 2011 was held from 5th to 9th September 2011. The school again covered EMI topics including ARC, dCache, gLite and UNICORE. EMI was asked to provide a tutor to present the hands-on gLite Administration Workshop and other EMI project members were present to present the other components. The agenda for the 2011 school is available on the GridKa school website [R10].

As this event was not organized directly by EMI it was not possible to collect participant feedback using the standard Feedback form, however, the organizers did collect feedback and the EMI-led training scored quite well. For example the gLite installation session scored 8.0 out of 10 on the quality of the training, and 8.5 out of 10 on the relevance of the tutorial.

4.2.5 EGI Technical Forum 2011 EMI organized a User Tutorial at the EGI Technical Forum in Lyon, France in September 2011 [R24]. This was based on material presented at the EMI Technical Conference / EGI User Forum and at the DCI Summer School; however, based on the feedback received from these sessions the content was slightly reduced to allow users to complete the exercises in the time available. The materials used are



available on the project Wiki at [R11]. The tutorial focused mainly on the compute technologies and learners who desired more information on EMI data technologies were referred to the EMI User Tutorial on the public EMI website [R12].

The hands-on training took place on the new EMI SA2 dedicated training testbed resources accessed via a UI machine hosted by INFN in Catania. Users were issued with short-lived GILDA certificates for authentication and authorization.

Feedback was very positive with an average satisfaction rating of 4.9 out of 6. The reduced content meant that more of the participants were able to complete the entire exercise, although the comments received in the feedback survey did include one to the effect that there was not enough time to complete the exercise. The complete feedback is presented in Annex C.

4.2.6 Summary of Feedback Results Table 2 presents a short summary of the average feedback rating for the training sessions. As you can see the overall average figure of 4.55 is above the target metric of 4 out of 6.

Training Session Event Location Date Feedback rating

EMI Tutorial for Resource Administrators 1

EMI Technical Conference / EGI User Forum

Vilnius, Lithuania April 2011 4.33






Vilnius, Lithuania April 2011 4




EMI Tutorial for Users EMI Technical Conference / EGI User Forum


EMI Training Day EU Join DCI Summer School Budapest, Hungary July 2011 4.56

User Tutorial EGI Technical Forum Lyon, France

September 2011 4.9

Overall Average Feedback Score 4.55 Table 2: User Training Average Feedback Scores

4.3. ONLINE DOCUMENTATION AND TRAINING MATERIALS While NA2 has organized or supported several training events, a major part of the EMI training strategy involves providing online materials which can be shared and reused. As outlined in the initial Training Plan Deliverable DNA2.2.1 [R1], this is necessary to broaden the reach of the project's training activities.

4.3.1 Documentation A documentation library has been created on the public website [R13]. This is a LifeRay portlet with a database back-end which allows searching the documentation library by component name, version, intended audience, document type, etc. The documentation listed in the library is based on the reviewed documentation list produced by the Documentation Team and is regularly updated.



This new documentation library supersedes the Wiki pages which were created at the start of the project to hold a list of links to documentation, as much of the material in the older list was not produced by EMI, but by its predecessor projects.

The documentation review is an NA2 task carried out in collaboration with SA2 which reviews documentation produced by the product teams to ensure that it complies with the minimum documentation requirements as specified in the EMI Documentation Policy [R14].

4.3.2 Online Tutorials A number of online tutorials have been produced covering the various EMI client tools for compute and data. These are consolidated in the public EMI Tutorial page [R12]. Whenever a face-to-face training event takes place, a Wiki page is created containing the practical instructions, and this is used by participants during the training event. The material to be covered may be all or a subset of the material on the EMI Tutorial page. After the event, the Wiki page is left unchanged as a record of the training session, but the public EMI Tutorial page can be updated to reflect the feedback received by participants in the training session.

An advanced Tutorial is also available which presents learners with a realistic scientific exercise which must be completed using a variety of EMI technologies. This is called the Integrating Practical as it integrates several EMI components and technologies into one exercise.

The Integrating Practical is a modified version of an exercise created by the National eScience Centre in Edinburgh NeSC [R16] as part of the International Summer School of Grid Computing (ISSGC) [R17] programme. The exercise involves using a Java application to search for patterns in a data file. The entire search space is too large to analyze in a single job and thus learners must submit a parameter-sweep-style job using their chosen EMI client tools. Learners are asked to use both compute and data clients in order to complete the exercise. A version of the Integrating Practical is available on the EMI public website at [R15].

An online UI installation tutorial is currently in development and will be available on the EMI public website in Q4 2011.

4.4. TRAINING COLLABORATIONS Leveraging collaborations to increase the reach and effectiveness of EMI training resources is a key part of the NA2 training strategy. Several successful collaborations are on-going.

4.4.1 European Grid Infrastructure EGI is a partnership between National Grid Initiatives (NGIs) and the coordinating body in Europe; they are thus a major customer of EMI and the contact point for the NGIs. Furthermore through the EGI accredited Virtual Research Community (VRC) programme, EGI is a contact point for a wide range of end-users in various scientific domains, European Strategy Forum on Research Infrastructures (ESFRI) projects, national research projects and other potential end-users.

EMI are members of the EGI Training Working Group [R19], the purpose of which is to develop and monitor a training coordination strategy for the community of actual and potential users of Distributed Computing Infrastructure across Europe. The group holds occasional telephone meetings and communicates via an email list.

EGI has created a Training Marketplace [R20] which will allow training providers to advertise their events and share training materials. The Marketplace was launched in July 2011 and supersedes the previous EGI digital library. Public training materials from the EMI website are available via the EGI Training Marketplace, and EMI has also provided feedback to EGI to improve the Training Marketplace interface.



4.4.2 GridKa School EMI has participated in two GridKa schools providing speakers on various topics and review of materials. The school caters for up to 100 students each year and covers a wide range of EMI technologies. In 2010 and 2011 EMI has collaborated with the organizers to provide speakers and assist with producing training materials. In the past the school has worked with individual middleware technologies and the legacy of this is still apparent. It is hoped that in 2012 EMI can increase involvement in the school in order to present a more holistic view of the EMI middleware, rather than having ARC, dCache, gLite and UNICORE presented separately.

4.4.3 DCI Summer School The first DCI Summer School was held in July 2011 and involved EMI, IGE, EDGI, DEGISCO, StratusLab, EGI-InSPIRE, as well as local organizers MTA SZTAKI. This was the first event run as a collaborative training event by the European DCI projects and it was very successful. It is likely that the school will go ahead again in 2012 and EMI have committed to be involved. The event has huge scope as a platform for further collaboration on training with the participating projects.



5. MEASURING SUCCESS AND METRICS The Description of Work identifies the number and quality of training events organized or attended, the average number of trained users and the impact of EMI training programs on other projects' or initiatives' training programs as the important metrics or KPIs for the Technical Knowledge Management activity.

DNA2.2.1 detailed the process for measuring the quality of the courses and training activities via surveys and course-feedback forms asking users to rate their training experience. A form for gathering user feedback was developed by NA2 and first used during the EMI Technical Conference 2011. An example of the feedback form used is given in Annex A.

The training activities of EMI have been very successful in terms of the number of training events organized, the number of people trained and the participant feedback. To-date, training activities have mainly been targeted towards the EGI user community and the EGI VRCs with several training workshops taking place at EGI events. In addition, a number of broader-focused events such as the DCI Summer School and the GridKa Summer School have been organized or supported. These events have extended the targeted user communities beyond the EGI user base with training being delivered to potential and actual Grid users from over 12 countries. These learners have been introduced to the new features of EMI Kebnekaise, and have been familiarized with the installation and configuration process, allowing them to install EMI products on their home infrastructures.

The collaboration with other DCI projects in the DCI Summer School has paved the way for closer collaboration with these projects on training in future. The DCI Summer School is in planning to be held again in 2012 with the potential for further collaboration and integration. Training materials have been made available on-line and disseminated to EGI and through them to the NGIs and VRCs. This will be continued with other communities targeted in future, including directly working with NGIs as well as new user communities such as PRACE.

The activities performed so far have highlighted some issues which are now being addressed. These include the lack of a dedicated training infrastructure, as well as the need for ongoing update and improvement to the training materials and resources. In addition, problems with the process of measuring success of training activities have been identified, and these are discussed in more detail in section 5.1.1 below.

5.1.1 Measuring effectiveness of training activities It has been difficult to ascertain the effectiveness of the training activities, in terms of skills transfer to real life situations, or number of people trained who have gone on to become new users of EMI software. This is partly due to the fact that EMI does not run a Grid infrastructure and thus has no way to see if trained users have gone on to request accounts or port their applications to a Grid infrastructure running EMI software.

One way to address this would be to perform follow-up surveys where a number of participants in past training events are contacted after a certain period of time and given a follow-up survey to determine whether they believe that the training they received has helped them to perform their jobs, whether they have gone on to use EMI software, which EMI components they have used, etc.

It may be possible to perform such a follow-up survey with some of the participants in the training sessions which have already been conducted, however, a more systematic approach which plans for this type of information gathering will be adopted for future events. For example, the post-course survey should collect user contact details and ask permission to contact participants in future in order to comply with data-protection and privacy laws and best-practices. The feedback form is being updated in line with these requirements. In the meantime a small-scale follow-up survey will be conducted.



5.1.2 Use of existing data In addition to gathering more data, more use could be made of the data which already exists. The comments collected when relevant, via the feedback surveys may contain suggestions which could be of use to the product teams as well as to NA2 and the training teams. A mechanism needs to be put in place to share the feedback with members of the EMI project beyond the NA2 group.

5.1.3 The holistic “EMI” view, as opposed to several middleware stacks There is a need to move towards a more holistic approach where training focuses less on particular middleware offerings and more on components of a single EMI middleware. This was difficult to achieve in the earlier stages of the EMI project as the middleware was only beginning to be consolidated. However, as time goes on the training events and materials are being modified to reflect the increased integration between the previously separate middleware stacks.

5.2. SPECIFIC METRICS The metrics for measuring success of the training activities are outlined in table 3 with achieved and target numbers.

Description of metric Year 1 Year 2 Year 3

Achieved Target Achieved (So far) Target Target

Number of training events organized directly by EMI 5 2 1 2 2

Number of training events organized jointly or collaborations in the training events of other projects

1 3 2 3 3

Number of in-reach events 2 2 0 1 1

Number of people trained on EMI middleware 1001 50 >40 50 50

Learner feedback (from 1 to 6) 4.48 >4 4.73 > year 1 score

> year 2 score

Table 3: Metrics

Most metrics were met or exceeded for the first year. The project is also on-course to meet the targets for year two. The number of jointly-organized training events was lower than expected in the first year of the project, as the project was new and the relationships with external projects had yet to be built. NA2 is now more proactively seeking out these relationships to ensure that there is more participation in external events over the remaining lifetime of the project.

1 This figure is based on the number of participants in the training sessions, although some did not complete

the hands-on exercises



6. COVERAGE OF EMI COMPONENTS IN TRAINING Training so far has included Compute and Data clients as well as installation and configuration of selected components. Table 4 indicates what types of training have been delivered for each component. While many components have no user-facing interface and are not appropriate targets for user training, there are several products which have not yet had any training delivered on them, but which would be suitable subjects for such training. This is particularly the case for site administrator training, which can include using certain EMI tools as an administrator, as well as installing and configuring the whole range of EMI products.

A gap analysis will be performed by NA2 in the coming months in order to identify priorities for site administrator training to be delivered in the coming period.

Product User Training Resource Administrator Training

AMGA No No

APEL Parsers No No

APEL Publisher No No

ARC CE N/A Yes

ARC Clients Yes No

ARC Core N/A No

ARC gridftp server N/A No

ARC InfoSys Yes No

ARGUS N/A No

ARGUS-EES No No

BDII core N/A No

BDII site N/A Yes

BDII top N/A No

BLAH N/A No

CEMon N/A Yes

CREAM Yes Yes

CREAM LSF module N/A No

CREAM TORQUE module N/A Yes

dCache Yes Yes

Delegation Java N/A No

DGAS-sensors N/A No

DPM Yes Yes

EMI UI Yes No

EMI WN N/A No

FTS N/A No

GFAL/lcg_util Yes (lcg_util) No

gLExec N/A No

gLite CLUSTER N/A Yes



gLite MPI No No

gLite-gSoap/gss N/A No

glite-proxyrenewal N/A No

glite-yaim-core N/A Yes

GridSite N/A No

L&B Yes No

LCAS N/A No

lcg-info-clients Yes No

LCMAPS N/A No

LCMAPS-plugins-c-pep N/A No

LFC Yes No

RAL-SAGA-SD No No

StoRM-SE Yes Yes

TORQUE server config N/A Yes

TORQUE WN config N/A Yes

Trustmanager N/A No

UNICORE Client Yes Yes

UNICORE Gateway N/A Yes

UNICORE HILA N/A No

UNICORE Registry N/A No

UNICORE Services Environment N/A Yes

UNICORE TSI N/A No

UNICORE WS N/A No

UNICORE XUUDB N/A No

UVOS N/A No

VOMS Yes No

VOMS Admin N/A No

WMS Yes No Table 4: Training coverage by product



7. FUTURE PLANS The main goal of the training activity is to enlarge the base of expert developers and engineers knowledgeable in EMI. This section outlines the plans to meet this goal.

7.1. ONLINE DOCUMENTATION AND TRAINING MATERIALS In-line with the first objective of producing training materials and resources for the EMI product suite, NA2 will continue to review the documentation produced by the product teams and to publish training materials based on this documentation.

The documentation library described in section 4.3.1 will continue to be maintained and updated based on the NA2/SA2 documentation reviews as new versions of documentation become available. This will involve a large effort around each major and minor software release, and ongoing work throughout the project lifetime to ensure that the library remains up-to-date.

The training material on the website is also under continual development. The material is updated after each face-to-face event based on participant feedback, and additional online tutorials covering additional topics such as installation and maintenance are planned. A basic client installation tutorial is in development and due to be uploaded to the public website in Q4 2011.

7.2. IN-REACH TRAINING The second objective of the Technical Knowledge Management activity within NA2 is to facilitate knowledge transfer within the EMI project. To this end NA2 has committed to deliver a minimum of four in-reach training sessions over the course of the project. Two events were delivered in the first year, and further events will be organized covering updated procedures, or other topics of interest to the project members.

The next planned in-reach event will take place on 2nd December, organized jointly by NA2 and SA1. The training will cover Debian Packaging and will take place online as for the second in-reach event described in section 4.1.1.

NA2 will identify other appropriate topics for in-reach training based on the technical plans, Matterhorn roadmap, etc. Possible topics include updated QA procedures, use of new tools such as the Verification Dashboard and others. In addition, project members will be surveyed by email in early 2012 to priorities future training topics.

Members can also request training on specific topics on the Training Topics Wiki page maintained by NA2 [R18] or by sending an email to [email protected]. Use of both of these mechanisms has been limited so far.

7.3. USER TRAINING The third objective of the Technical Knowledge Management task is to run user training events covering the EMI product suite. Users in the context of EMI include resource administrators, end users and application developers, and the training strategy aims to target each of these user groups to different extents. Training to date has focused largely on resource administrators and end-users, but training for application developers is now also being planned.

EMI has directly organized two training workshops to date, at the EMI Technical Conference / EGI User Forum 2011 and EGI Technical Forum 2011. The project has also collaborated with other projects such as the GridKa School and the DCI Summer School to deliver EMI training at these events.

mailto:[email protected]



The GridKa School will be held again next year and EMI hopes to have a stronger involvement in order to influence the curriculum which is still very much focused on separate middleware stacks. By next year's school EMI 2 “Matterhorn” will be available, and it is hoped that this will provide impetus for change in the School's curriculum toward a more integrated view of the EMI middleware stack.

The organizers of the DCI Summer School have suggested running another event next year, perhaps earlier in the calendar year. While no firm plans are in place yet, EMI has indicated its commitment in any future school. Summer Schools and longer workshops of this type provide a unique opportunity to provide in-depth training to users, as opposed to the shorter one to two hour workshops organized at conferences and technical meetings. This produces more expert EMI users as a result who can go on to skillfully administer EMI installations, become advocates for EMI products within their organizations, and potentially train other users.

Training events will be organized at the 2nd EMI Technical Conference/EGI Community Forum in 2012 [R21], and the EGI Technical Forum 2012. Other events will be targeted as appropriate. Possible targets which were previously investigated include OGF and the IEEE/ACM International Conference on Grid Computing, both of which were co-hosted with the EGI Technical Forum in Lyon in September, at which a training event was held. Tutorial proposals will be submitted to future events in these series, as well as other relevant conferences such as the International Supercomputing Conference, the IEEE International Conference on e-Science, and others.

A Tutorial proposal for the International Supercomputing Conference 2011 in Hamburg, Germany was not accepted because the tutorial schedule was already full at the time of submission, but a tutorial will be submitted to ISC'12 which takes place from 17th to 21st June 2012. EMI will also be present at ISC'12 with a booth and thus the training will form part of a broad dissemination strategy aimed at a HPC-oriented audience, who are largely not current EMI users.

NA2 will be more proactive in identifying possible target events for tutorials, and also in pursuing collaborations with other projects and entities. This will include discussions directly with the NGIs as well as via EGI. In line with the objective to increase the EMI user base, the Technical Knowledge Management task will collaborate with the dissemination task of NA2 in targeting new user communities such as PRACE. Training will be offered to these user communities to provide them with the skills required to use the EMI product suite. Training workshops are planned for events such as OGF34 and ISC in 2012 which have a broader focus than the current EMI user community. This will also enable new potential users to be targeted and provided with training in EMI products.

Training for end users and system administrators will be organized to coincide with the EMI 2 Matterhorn release, focusing on the new features of the release, as well as more general training.

Training so far has focused on end-users and resource administrators, but there has been limited training provided for application developers. Training aimed at those developing applications which use the EMI APIs will be provided in future. The first developer-focused training workshop will be proposed for the EGI Community Forum / EMI Technical Conference in March 2012. Based on experiences at this event other similar workshops will be planned. Possible venues include the e-Science 2012 and ISC 2012.

A train the trainers workshop is also being proposed for the 2nd EMI Technical Conference/EGI Community Forum 2012. This event will attract many participants from within the NGIs and is thus a very suitable forum to reach members of the National Grid projects who are responsible for training users within their countries. Training the trainers events, by introducing the EMI products and features to the local NGI members responsible for training, have the potential to multiply the effect of EMI training efforts and thus greatly increase the number of trained EMI users.

A list of events at which user training is planned is given in section 5.5 of this document.



7.3.1 Follow-up survey A follow-up survey will be conducted by contacting selected participants from training sessions held to-date. The aim of this survey will be to determine the effectiveness and impact of EMI training activities.

The survey will attempt to find answers to the following questions:

• Did participants learn anything in the training that they could transfer to their work practice?

• If participants had not used EMI components (as a user or administrator) before the training, have they gone on to do so afterwards?

• If not, why?

• Do participants think that more training is required?

• What topics would participants suggest for additional training?

This information will help us to assess the effectiveness of our training activities and also to plan future events. The first such survey will attempt to contact a small subset of participants in previous training events where contact details are available. In future this will be done more systematically and participants in every training course will be invited to opt-in for future follow-up surveys.

7.4. TRAINING COLLABORATIONS AND DISSEMINATION OF MATERIALS The fourth objective of the NA2 Technical Knowledge Management task involves dissemination of training materials within potential user communities. A large part of this work will involve continued participation in the EGI Training Workgroup and publishing materials to the EGI Training Marketplace. Other new user communities will also be targeted in this fashion; in particular the PRACE user community will be targeted in collaboration with the Dissemination task of NA2. Some new training materials may be developed specifically for these new communities, and those already in existence will be disseminated more widely within the communities.

The geographic scope of the training activities is also being extended. The proposed new Future Grid training infrastructure (described below in section 7.5) will allow US and other non-EU participants to try out the EMI product suite as well as giving EMI a permanent training infrastructure to use in all training activities.

7.5. TRAINING INFRASTRUCTURE Until DCISS’11, held in July 2011, the infrastructure used for training was the Certification and Integration Testbed deployed by SA2 for the continuous integration testing process of EMI software products. The primary purpose of this infrastructure of continuous integration implies that some products might be not mature enough, potentially resulting in a negative perception of EMI middleware from students. For the EGI Technical Forum 2011 SA2 deployed a new testbed expressly dedicated for training and demonstration needs of that event. This test bed was, however, a one-off installation, and is not permanently available.

Ideally a permanent dedicated training infrastructure would be available. This would ensure that learners are not forced to use a rapidly changing and potentially unstable testing infrastructure, and similarly that the testing activities of EMI are not impacted by the training schedule.

In addition, a dedicated training infrastructure has the potential to provide access to remote and self-paced learners, who cannot be supported on an ad-hoc basis on an internal EMI testing resource.

The project is investigating the opportunities offered by Future Grid project [R22] to provide a permanent training infrastructure and extend the number of services available for training. Future



Grid has agreed to provide a permanent limited training testbed, which can be used for EMI training events and also potentially by self-paced learners. In addition, this testbed will be available for users and training activities in the US, thus geographically expanding the EMI training impact.

7.6. LIST OF SPECIFIC PLANNED ACTIVITIES A list of proposed training activities in the coming project year is given in Table 5.

Event/Title Details Audience Location Date Status

Third in-reach training event

Training for EMI developers on Debian packaging. EMI developers Online December 2011 In Planning

In-reach training topics survey

Members of the EMI project will be surveyed by email to identify and priorities topics for future in-reach training events.

EMI developers Online Q1 2012 In Planning

Client Installation Tutorial on Website

A short client installation tutorial aimed site administrators and end users. The tutorial expands the installation section of the existing User Tutorial.

Resource admins and end-users

Online Q4 2011 In progress

First training follow-up survey

A subset of attendees at past user-training events will be contacted for a follow-op survey to determine the long-term usefulness of the EMI user training.

Resource admins and end users

Online Q4 2011 In Planning

EGI Community Forum / EMI Technical Conference

Training workshops for resource administrators, trainers in the NGIs and application developers are planned.

Resource administrators, trainers and end users

Munich, Germany March 2012 In Planning

OGF 34

The OGF is the standards development body for Grid computing. The large number of technical participants would make it suitable for application developer training. End-users might also be targeted in order to introduce the EMI product suite to a wider non-European audience.

Application developer, end user.

Oxford, UK March 2012 TBC

DCISS

Training for resource administrators, application developers and end-users. Aimed at participants of the DCI Summer School who come from a range of different backgrounds. Details of the exact training to be provided will be decided when more information on the focus of

Resource admins, application developers and end-users

TBC Q3 2012 TBC



the school is available.

GridKa 2012

Training for resource administrators on installation and configuration of EMI components.

Resource admins

Karlsruhe, Germany September 2012 In Planning

Fourth in-reach training event

In-reach training event aimed at developers within the EMI project. Topics to be determined based on the in-reach topics survey mentioned above.

EMI developers Online 2012 TBC

ISC

The ISC event gives access to the super-computing user community who are not traditional EMI users. This will be a good chance to introduce the EMI middleware to this user base. Training will probably focus mainly towards end-users in order to introduce the capabilities of the middleware to this new community.

End-users Hamburg, Germany June 2012 TBC

Grid2012

A technical conference which attracts many Grid developers and administrators. Training sessions for resource administrators, application developers and perhaps also end-users will be proposed by NA2.

Resource admins, application developers and end-users

TBC 2012 TBC

E-Science 2012 This conference may be a good target for application developer training.

Application developer TBC 2012 TBC

Table 3: Specific Training Activities for the Coming Period



8. CONCLUSIONS A range of training activities have been organized in this period, in line with the plans laid out in Deliverable DNA2.2.1. In addition, online materials have been made available, and these are being shared through our training collaborations.

Just as documentation and training materials are being constantly updated and improved, NA2 is endeavoring to conduct an analysis of the training activities in order to improve their effectiveness and reach. An analysis of the feedback received to date on the training activities has already suggested some areas which require change or further attention, and this feedback is being used to continually improve the content of the EMI training. In addition, follow-up surveys to measure long-term effectiveness and impact of training activities are planned.

In order to better plan future events, a further internal survey is planned to identify suitable in-reach training topics, and a gap analysis on the user training coverage of EMI products will be carried out. A number of proposed future training events, including online and face-to-face sessions have been proposed, although many are still to be confirmed.



ANNEX A: USER FEEDBACK SURVEY Joint European DCI Summer School, Budapest, Hungary

Date: Monday 11th July 2011 09:00 - 18:00Please note that all data is collected anonymously and there is no link to your identity, organisation or location.

1. Please rate the following aspects of the course

Poor Fair Average Good Very Good Excellent Overview of EMI middleware

Advanced EMI topics Hands-on: Compute & Data

Hands-on: Advanced Workflow (Integrating Practical)

Wrap-up and look ahead

Overall Evaluation Advertising & Registration

Facilities 2. Please provide any additional feedback on the course aspects rated above

Overview of EMI middleware Advanced EMI topics Hands-on: Compute & Data Hands-on: Advanced Workflow (Integrating Practical)

Wrap-up and look ahead Overall Evaluation Advertising & Registration

Facilities 1. What did you most like about this training? 2. What did you least like about this training? 3. Which of the training subjects were the most useful to you? 4. Which of the training subjects did you find the least useful? 5. Is there anything else you would like the training to have covered? 6. Further Comments

ANNEX B: DETAILED IN-REACH TRAINING FEEDBACK FEEDBACK ON GUIDELINES FOR USAGE OF TOOLS USED BY THE EMI PROJECT

(E.G. ETICS) Users were asked to rate the following aspects of the course on a six-point scale from Poor to

Excellent. • EMI Tools overview • EMI Build and Integration Policy • EMI Packaging Policy • The new ETICS client 1.5.0 • QA tools overview • Overall Evaluation • Advertising & Registration • Learning Environment (Adobe Connect) They could also provide an optional free text comment on each of the above aspects. Page two of the survey asked the following questions with answers in the form of free-text comments: 1. What did you most like about this training? 2. What did you least like about this training? 3. Which of the training subjects were the most useful to you? 4. Which of the training subjects did you find the least useful? 5. Is there anything else you would like the training to have covered? 6. Further Comments Converting the rating the rating to a numeric scale from 1 to 6, the average satisfaction score is 4.81 out of 6. The full rating is given in Table 1 while a graph of the rating responses is in Figure 1.

Poor Fair Average Good Very Good Excellent Response Count

EMI Tools overview 0.0% (0) 0.0% (0) 0.0% (0) 21.4% (3) 57.1% (8) 21.4% (3) 14

EMI Build and Integration Policy

0.0% (0) 0.0% (0) 0.0% (0) 28.6% (4) 42.9% (6) 28.6% (4) 14

EMI Packaging Policy

0.0% (0) 0.0% (0) 0.0% (0) 28.6% (4) 50.0% (7) 21.4% (3) 14

The new ETICS client 1.5.0

0.0% (0) 0.0% (0) 0.0% (0) 42.9% (6) 50.0% (7) 7.1% (1) 14

QA tools overview 0.0% (0) 0.0% (0) 7.1% (1) 28.6% (4) 50.0% (7) 14.3% (2) 14

Overall Evaluation 0.0% (0) 0.0% (0) 0.0% (0) 14.3% (2) 64.3% (9) 21.4% (3) 14

Advertising & Registration 7.7% (1) 15.4% (2) 0.0% (0) 23.1% (3) 30.8% (4) 23.1% (3) 13

Learning Environment (Adobe Connect)

0.0% (0) 0.0% (0) 0.0% (0) 35.7% (5) 42.9% (6) 21.4% (3) 14

Table 1: Rating for EMI Tools in-reach training

Figure 1: Graph of rating for the EMI Tools in-reach training Comments

The new ETICS client 1.5.0 : • “Actually, this topic wasn't exposed explicitly” Overall Evaluation : • “Quite clear and useful.” Advertising & Registration : • “Good management.” Learning Environment (Adobe Connect) : • “Prefer to write instead of speaking not to interrupt the discussion flow; I would suggest to collect

written questions all the time so one can reason a bit about them.” • “Some mistakes with buttons” • “Quite good, only a few troubles with "Web links" window.” What did you most like about this training? • “The speaker was quite clear.” • “Focused on the needed things but open to discuss the unclear ones” • “Content, speaker, duration” What did you least like about this training? • “Audio from other participants :(” • “I was already acquainted with the general usage of the ETICS web portal. That part was a bit

boring for me.” • “With Adobe Connect you don't really go raising you hand, you prefer writing -- but I think this

has his good side, you keep track of the discussion” Which of the training subjects were the most useful to you? • “Packaging guidelines.” • “Learning some of the details of the Build and Release policy. Getting a final statement about how

to use dependency, which have been used in different manners so far (use them/don't use them).”

• “Packaging guidelines as I was working on them”

• “All” Which of the training subjects did you find the least useful? • “All useful.” • “ETICS web portal, see above.” • “None” Is there anything else you would like the training to have covered? • “No” FEEDBACK ON USING THE EMI TESTBED Users were asked to rate the following aspects of the course on a six-point scale from Poor to

Excellent. • Introduction • gLite testbed • dCache data management • Unicore testbed • Arc testbed • Overall Evaluation • Advertising & Registration • Learning Environment (Adobe Connect) They could also provide an optional free text comment on each of the above aspects. Page two of the survey asked the following questions with answers in the form of free-text comments: 1. What did you most like about this training? 2. What did you least like about this training? 3. Which of the training subjects were the most useful to you? 4. Which of the training subjects did you find the least useful? 5. Is there anything else you would like the training to have covered? 6. Further Comments Converting the rating the rating to a numeric scale from 1 to 6, the average satisfaction score is 4.03 out of 6. The full rating is given in Table 2 while a graph of the rating responses is in Figure 2.


Introduction 9.1% (1) 0.0% (0) 18.2% (2) 45.5% (5) 27.3% (3) 0.0% (0) 11

gLite testbed 9.1% (1) 0.0% (0) 27.3% (3) 54.5% (6) 9.1% (1) 0.0% (0) 11

dCache data management 9.1% (1) 0.0% (0) 36.4% (4) 36.4% (4) 18.2% (2) 0.0% (0) 11

Unicore testbed 9.1% (1) 0.0% (0) 18.2% (2) 45.5% (5) 27.3% (3) 0.0% (0) 11

Arc testbed 9.1% (1) 9.1% (1) 27.3% (3) 27.3% (3) 27.3% (3) 0.0% (0) 11




0.0% (0) 0.0% (0) 0.0% (0) 50.0% (1) 50.0% (1) 0.0% (0) 2

Table 2: Rating for the EMI Testbed in-reach training Comments

Introduction • “In general chosen presentation tool gives good voice but much less satisfactory image. No way to

resize it, difficult to read and follow.” gLite testbed • “Less client usage. More service discovery and interfaces.” dCache data management • “Less client usage. More service discovery and interfaces.” Unicore testbed • “Less client usage. More service discovery and interfaces.” Arc testbed • “Less client usage. More service discovery and interfaces.” • “screen sharing” What did you most like about this training? • “Hands-on approach” Further Comments • “the screensharing was not always easy to follow. The slides/text was too small, difficult to read.”

Figure 2: Graph of rating for EMI Testbed in-reach training FEEDBACK ON CERTIFICATION, CHANGE MANAGEMENT AND RELEASE MANAGEMENT POLICIES WITHIN THE EMI PROJECT Users were asked to rate the following aspects of the course on a six-point scale from Poor to Excellent. • Overview of EMI component release checklist • Practical Examples • Overall Evaluation • Advertising & Registration • Learning Environment (Adobe Connect) They could also provide an optional free text comment on each of the above aspects.

Page two of the survey asked the following questions with answers in the form of free-text comments: 1. What did you most like about this training? 2. What did you least like about this training? 3. Which of the training subjects were the most useful to you? 4. Which of the training subjects did you find the least useful? 5. Is there anything else you would like the training to have covered? 6. Further Comments Converting the rating the rating to a numeric scale from 1 to 6, the average satisfaction score is 5.03 out of 6. The full rating is given in Table 3 while a graph of the rating responses is in Figure 3.


Component release checklist

0.0% (0) 0.0% (0) 0.0% (0) 0.0% (0) 71.4% (5) 28.6% (2) 7

Practical Examples 0.0% (0) 0.0% (0) 0.0% (0) 14.3% (1) 71.4% (5) 14.3% (1) 7




0.0% (0) 0.0% (0) 14.3% (1) 28.6% (2) 42.9% (3) 14.3% (1) 7

Table 3: Rating for the EMI Release Management in-reach training Comments

Overview of EMI component release checklist • “Before the training, I didn't even know it existed.” Overall Evaluation • “Really useful!! Maybe a bit earlier would have been better...” Learning Environment (Adobe Connect) • “Working very well. Good choice!” • “not fully supported for linux” • “Good for audience on all systems, for presenters only on Windows :(“ What did you most like about this training? • “Very structured, step by step and with lots of examples.” • “practical examples” • “Ability to ask questions” What did you least like about this training? • “That it was so late” Which of the training subjects were the most useful to you? • “To be honest all the procedures were very useful to have explained.” • “emi clients usage on test bed” • “Not many as I anyway had to figure out the answers earlier.” • “Finding the checklist.” Which of the training subjects did you find the least useful? • “none” Is there anything else you would like the training to have covered? • “client usage of dCache and Argus on the testbeds”

• “Yes - a subsequent training (or telco) AFTER EMI-1 rc1 deadline would be useful. Mostly to discuss problems with policies which were found.”

Further Comments • “Really good. Thanks!”

Figure 3: Graph of rating for EMI Release Management in-reach training

ANNEX C: DETAILED USER TRAINING FEEDBACK FEEDBACK ON EMI TUTORIAL FOR RESOURCE ADMINISTRATORS 1 Users were asked to rate the following aspects of the course on a six-point scale from Poor to Excellent. • What changes with EMI 1? • Installation and configuration of CREAM • Hands-on Exercises • Overall Evaluation • Advertising and Registration • Facilities They could also provide an optional free-text comment on each of the above aspects. Page two of the survey asked the following questions with answers in the form of free-text comments: 1. What did you most like about this training? 2. What did you least like about this training? 3. Which of the training subjects were the most useful to you? 4. Which of the training subjects did you find the least useful? 5. Is there anything else you would like the training to have covered? 6. Further Comments Converting the rating to a numeric scale from 1 to 6, the average satisfaction score is 4.33 out of 6. The full rating is given in Table 4.


What changes with EMI 1? 0.0% (0) 0.0% (0) 18.2% (2) 54.5% (6) 18.2% (2) 9.1% (1) 11

Installation and configuration of CREAM

0.0% (0) 0.0% (0) 23.1% (3) 46.2% (6) 7.7% (1) 23.1% (3) 13

Hands-on Exercises 0.0% (0) 0.0% (0) 23.1% (3) 38.5% (5) 23.1% (3) 15.4% (2) 13



Facilities 0.0% (0) 0.0% (0) 0.0% (0) 30.8% (4) 53.8% (7) 15.4% (2) 13

Table 4: Rating for EMI Tutorial for Resource Administrators 1 Comments

What changes in EMI 1? • “I missed this part” • “I was counting on something different by reading the title.” • “N/A column in survey to indicate if not all parts of talk were attended” Installation and configuration of CREAM • “It was a good experience to see how to configure the CREAM and see the YAIM in work.” Hands-on Exercises • “good tutorial, too bad we don't have time to better understand what can be configured in the

configuration files” • “I'd expect at least checking functionality of the service to be included in the tutorial.” • “Having a WN to test would have been lovely” • “I'm not sure that it really works.”

Facilities • “the set up” • “Network is critical on such an interactive tutorial.” What did you most like about this training? • “Hands on” • “I was seen a CREAM configuration in practice.” • “the exercise” • “well organized, easy to follow” • “The hands-on approach” What did you least like about this training? • “I can't try it out.” • “to fast, but due to time constraints, was not possible more then it was done” Which of the training subjects were the most useful to you? • “Seeing the configuration tool.” • “we will see in staged rollout, if this goes as well as here” Which of the training subjects were the least useful to you? • “Seeing how to install the host credentials.” Is there anything else you would like the training to have covered? • “Cluster testing.” FEEDBACK ON EMI TUTORIAL FOR RESOURCE ADMINISTRATORS 2 Users were asked to rate the following aspects of the course on a six-point scale from Poor to Excellent • Installation and Configuration of UNICORE • Hands-on Exercises • Overall Evaluation • Advertising & Registration • Facilities They could also provide an optional free-text comment on each of the above aspects. Page two of the survey asked the following questions with answers in the form of free-text comments 1. What did you most like about this training? 2. What did you least like about this training? 3. Which of the training subjects were the most useful to you? 4. Which of the training subjects did you find the least useful? 5. Is there anything else you would like the training to have covered? 6. Further Comments Converting the rating to a numeric scale from 1 to 6, the average satisfaction score is 4.9 out of 6. The full rating is given in Table 5.

Poor Fair Average Good Very Good Excellent Response

Count

Installation and Configuration of UNICORE

0.0% (0) 0.0% (0) 0.0% (0) 25.0% (1) 25.0% (1) 50.0% (2) 4




Facilities 0.0% (0) 0.0% (0) 0.0% (0) 50.0% (2) 25.0% (1) 25.0% (1) 4 Table 5: Rating for EMI Tutorial for Resource Administrators 2

Comments

Overall Evaluation • “Really enjoyed the tutorial” Facilities • “The network or virtual machines are slow.” What did you most like about this training? • “Well prepared, techers frindly and helpful” • “it was fun” • “The felling that I configured a grid for myself.” What did you least like about this training? • “I had to wait a bit for some URLs, and I had to kill the yum autoupdate” Which of the training subjects were the most useful to you? • “General introduction” • “interesting concept of Applications” • “How to configure a basic resource, and how to create an own application.” Which of the training subjects did you find the least useful? • “I don't know” Is there anything else you would like the training to have covered? • “how to extract the key and cert pem from the keystore” FEEDBACK ON EMI TUTORIAL FOR RESOURCE ADMINISTRATORS 3 Users were asked to rate the following aspects of the course on a six-point scale from Poor to Excellent • Installation and Configuration of dCache • Installation and Configuration of StoRM • Hands-on Exercises • Overall Evaluation • Advertising & Registration • Facilities They could also provide an optional free-text comment on each of the above aspects. Page two of the survey asked the following questions with answers in the form of free-text comments 1. What did you most like about this training? 2. What did you least like about this training? 3. Which of the training subjects were the most useful to you? 4. Which of the training subjects did you find the least useful? 5. Is there anything else you would like the training to have covered? 6. Further Comments Converting the rating to a numeric scale from 1 to 6, the average satisfaction score is 4 out of 6. The full rating is given in Table 6. Note that only one participant completed the feedback survey for this session, and no comments were provided.


Count

Installation and configuration of dCache

0.0% (0) 0.0% (0) 0.0% (0) 0.0% (0) 100.0% (1) 0.0% (0) 1

Installation and 0.0% (0) 0.0% (0) 100.0% (1) 0.0% (0) 0.0% (0) 0.0% (0) 1

configuration of StoRM




Facilities 0.0% (0) 0.0% (0) 0.0% (0) 100.0% (1) 0.0% (0) 0.0% (0) 1

Table 6: Rating for EMI Tutorial for Resource Administrators 3 FEEDBACK ON EMI TUTORIAL FOR RESOURCE ADMINISTRATORS 4 Users were asked to rate the following aspects of the course on a six-point scale from Poor to Excellent • Installation and Configuration of A-REX • Hands-on Exercises • Overall Evaluation • Advertising & Registration • Facilities They could also provide an optional free-text comment on each of the above aspects. Page two of the survey asked the following questions with answers in the form of free-text comments 1. What did you most like about this training? 2. What did you least like about this training? 3. Which of the training subjects were the most useful to you? 4. Which of the training subjects did you find the least useful? 5. Is there anything else you would like the training to have covered? 6. Further Comments Converting the rating to a numeric scale from 1 to 6, the average satisfaction score is 4.57 out of 6. The full rating is given in Table 7.


Count

Installation and Configuration of A-REX

0.0% (0) 0.0% (0) 0.0% (0) 0.0% (0) 83.3% (5) 16.7% (1) 6




Facilities 0.0% (0) 0.0% (0) 33.3% (2) 33.3% (2) 33.3% (2) 0.0% (0) 6

Table 7: Rating for EMI Tutorial for Resource Administrators 4 Comments

Hands-on Exercises

• “Very "Copy-pasteable", make things easy.” Advertising & Registration • “Was there any ?” Facilities • “Small tables would have been nicer for my back.” What did you most like about this training? • “it worked!” • “Clear written step by step instructions on the web page, easy to find via the bit.ly link. VMs and

certificate infrastructure worked well. Good that VMs had a ready and working SGE install.” • “The simple copy paste instructions” • “That it worked.” • “nice tutors, good materials, professional support” What did you least like about this training? • “vm a bit slow (but not terrible)” • “The lack of depht, although it is understandable, there's not much time.” • “The missing table” • “-” Which of the training subjects were the most useful to you? • “all in general quite useful” • “the new arc client commands :)” • “installation” • “answers to my individual questions” Which of the training subjects did you find the least useful? • “none specifically” • “--” • “hard to say” Is there anything else you would like the training to have covered? • “Voms proxy user mapping, but I have a lot of experience with ARC, so it's not suitable for this

level for the course” • “yes. The differences between middlewares, underlying unique features etc.” Further comments • “well-organised, clear instructions” • “Thanks!” FEEDBACK ON EMI TUTORIAL FOR USERS Users were asked to rate the following aspects of the course on a six-point scale from Poor to Excellent • Overview of ARC user clients usage • Overview of gLite user clients usage • Overview of UNICORE user clients usage • Overview of dCache user clients usage • Hands-on Exercises • Overall Evaluation • Advertising & Registration • Facilities They could also provide an optional free-text comment on each of the above aspects. Page two of the survey asked the following questions with answers in the form of free-text comments 1. What did you most like about this training? 2. What did you least like about this training? 3. Which of the training subjects were the most useful to you? 4. Which of the training subjects did you find the least useful? 5. Is there anything else you would like the training to have covered?

6. Further Comments

Converting the rating to a numeric scale from 1 to 6, the average satisfaction score is 4.58 out of 6. The full rating is given in Table 8.


Count

Overview of ARC user clients usage

0.0% (0) 0.0% (0) 0.0% (0) 0.0% (0) 33.3% (1) 66.7% (2) 3

Overview of gLite user clients usage

0.0% (0) 0.0% (0) 0.0% (0) 33.3% (1) 0.0% (0) 66.7% (2) 3

Overview of UNICORE user clients usage

0.0% (0) 0.0% (0) 33.3% (1) 33.3% (1) 0.0% (0) 33.3% (1) 3

Overview of dCache user clients usage

0.0% (0) 0.0% (0) 33.3% (1) 33.3% (1) 0.0% (0) 33.3% (1) 3




Facilities 0.0% (0) 33.3% (1) 0.0% (0) 0.0% (0) 66.7% (2) 0.0% (0) 3

Table8: Rating for EMI Tutorial for Users Comments

Overview of ARC user clients usage • “Missed ARC data clients” Overview of gLite user clients usage • “Couldn't find file uploaded with lcg-cr” Overview of UNICORE user clients usage • “Didn't understand data management information” Overview of dCache user clients usage • “Missed information on the different commands” Hands-on Exercises • “Ok” Overall Evaluation • “Ok” Advertising & Registration • “Ok” Facilities • “Too small room” What did you most like about this training? • “Everything actually worked, except ARC submission to CREAM.” • “Got hands-on experience of all tools” What did you least like about this training? • “It is quite hard to actually participate hands-on for 1.5 hours and concentrate so long.”

• “The room was too crowded” Which of the training subjects were the most useful to you? • “dCache ability to serve several protocols at the same time SRM, FTP, WebDAV)” • “All equally useful” Which of the training subjects did you find the least useful? • “Everything was useful.” • “All equally useful” Is there anything else you would like the training to have covered? • “Best practices in submitting bulk jobs/parameter sweeps, possibilities of scripting or integrating

submission tools with user applications, so that jobs can be submitted from/within user application.”

• “More info on data management” FEEDBACK ON DCISS EMI TRAINING DAY Users were asked to rate the following aspects of the course on a six-point scale from Poor to Excellent • Overview of EMI middleware • Advanced EMI topics • Hands-on: Compute & Data • Hands-on: Advanced Workflow (Integrating Practical) • Wrap-up and look ahead • Overall Evaluation • Advertising & Registration • Facilities They could also provide an optional free-text comment on each of the above aspects. Page two of the survey asked the following questions with answers in the form of free-text comments 1. What did you most like about this training? 2. What did you least like about this training? 3. Which of the training subjects were the most useful to you? 4. Which of the training subjects did you find the least useful? 5. Is there anything else you would like the training to have covered? 6. Further Comments Converting the rating to a numeric scale from 1 to 6, the average satisfaction score is 4.56 out of 6. The full rating is given in Table 9. Poor Fair Average Good Very Good Excellent Response

Count

Overview of EMI middleware

0.0% (0) 0.0% (0) 0.0% (0) 35.7% (5) 57.1% (8) 7.1% (1) 14

Advanced EMI topics 0.0% (0) 0.0% (0) 7.1% (1) 71.4% (10)

21.4% (3) 0.0% (0) 14

Hands-on: Compute & Data

0.0% (0) 0.0% (0) 0.0% (0) 33.3% (5) 46.7% (7) 20.0% (3) 15

Hands-on: Advanced Workflow (Integrating Practical)

0.0% (0) 6.7% (1) 0.0% (0) 33.3% (5) 33.3% (5) 26.7% (4) 15

Wrap-up and look ahead

0.0% (0) 0.0% (0) 9.1% (1) 36.4% (4) 45.5% (5) 9.1% (1) 11


Advertising & Registration

0.0% (0) 0.0% (0) 0.0% (0) 42.9% (6) 50.0% (7) 7.1% (1) 14

Facilities 0.0% (0) 7.1% (1) 0.0% (0) 50.0% (7) 35.7% (5) 7.1% (1) 14

Table9: Rating for DCISS EMI Training Day Comments

Overview of EMI Middleware • “I did not heard a lot about EMI until today, and this knoweledge is very useful” • “Nothing” • “Good focus on the main aspects.” Advanced EMI Topics • “It was quite practical topic but we couldn't try it.” • “The demo was a little fast.” Hands-on: Compute and Data • “Nothing” • “Very good style to let me find out about the tools in this written form - so I had the chance to

complete the task without waiting for others.” • “Very useful tutorial.” Hands-on: Advanced Workflow (Integrating Practical) • “I learned how to use, that I've heard during the day” • “The task was quite interesting.” • “Nice idea!” Wrap-up and look ahead • “Nothing” Overall Evaluation • “Nothing” Advertising & Registration • “Nothing” • “I do not know what is it about” Facilities • “It was not too hard, but I've been learnt a lot” • “Computers were quite slow and we had to wait lot of times to download pages (more than 1

min)” • “The food was awesome” • “I do not know what is it about” What did you most like about this training? • “The practice was very useful” • “Organization” • “The training contained meaningful information, focused on the most important aspects and not on

useless details.” • “The pillar finding game :)” • “The overall view” • “Hands-ons” • “Very good written tutorial” • “practical” • “training” • “The general presentations and the integrating practical hands-on session.” • “introduction of EMI” • “Advanced EMI Topic” • “The materials and the information”

What did you least like about this training? • “It's not easy to do the practice right after the presentations.” • “Some points go fast Relation between thoery & practice” • “The time was too short to explore all the functionalists brought by EMI.” • “nothing” • “-” • “The broken wifi” • “some point was fast” • “find pillar making job to arc” • “Then hands on guides were interesting, but I think I wasn't prepared enough for the first one, so it

was just a copy/paste exercise for me. It would have been helpful to have more information about each step and not just the step itself.”

Which of the training subjects were the most useful to you? • “the overview of EMI, because I have not used EMI yet, so I have heard a lots of useful

information.” • “Installation of some middle-ware” • “Both trainings were useful: the first one in presenting the functionalists of the EMI using different

middlewares and the second one for actually applying these functionalists.” • “the hands-ons were usefuls.” • “training example” • “Advanced Workflow: Becuase we could try Grid system on real problem” • “View over different middlewares” • “glite handout” • “glite” • “General presentations and hands-on session.” • “hands-on serveral middleware such as arc, glite, unicore, dcash, etc” • “The examples from ARC, Unicore, dCache” Which of the training subjects did you find the least useful? • “Advenced EMI topics were hard to understand for me, because the reasons I have mentioned

above.” • “nothing” • “-” • “No one. All interesting.” • “finding pillar that require some program ablility” Is there anything else you would like the training to have covered? • “Requirments before comming to the school” • “Execute different processes using EMI, maybe examples brought by the users.” • “more time for training” • “-” • “no” • “No” • “hand-on easier example” Further Comments • “if the material is able to be studied on line before participating. as well after” • “-” • “thank you” • “The advanced EMI topic should provide some general view of each presented subject, just for the

beginners to have on overview of the entire system. It was supposed that the audience has some basic knowledge abot the gLite, ARC etc.”

• “nothing else” Some general comments were also collected in the discussion sesssion at the end of the EMI day:

1. “Some of topics are new to students, would be useful to put reference on line in advance, to assess needed skills.”

2. “It's better to write a small guide on how to install UI or configure elements” 3. “Put same initial context of the grid, not everyone know what is 'submitting a job'.” 4. “Hands on : jump too big between first and second part of tutorial : propose problems to solve

even in the first part. however more time would be needed.” 5. “It would be good to see the difference among the middleware.” 6. “Advanced topic talk in the morning was a bit hard to understand for who has not experience” 7. “the format of tutorial is good (web page) is good because it allows to go on your own speed” FEEDBACK ON EGI TECHNICAL FORUM 2011 Users were asked to rate the following aspects of the course on a six-point scale from Poor to Excellent • Hands-on: Compute • Hands-on: Data • Overall Evaluation • Advertising & Registration • Facilities Page two of the survey asked the following questions with answers in the form of free-text comments 1. What did you most like about this training? 2. What did you least like about this training? 3. Which of the training subjects were the most useful to you? 4. Which of the training subjects did you find the least useful? 5. Is there anything else you would like the training to have covered? 6. Further Comments Converting the rating to a numeric scale from 1 to 6, the average satisfaction score is 4.9 out of 6. The full rating is given in Table 10.


Count

Hands-on: Compute 0.0% (0) 0.0% (0) 20.0% (1) 0.0% (0) 40.0% (2) 40.0% (2) 5

Hands-on: Data 0.0% (0) 0.0% (0) 20.0% (1) 0.0% (0) 40.0% (2) 40.0% (2) 5



Facilities 0.0% (0) 0.0% (0) 20.0% (1) 0.0% (0) 40.0% (2) 40.0% (2) 5

Table 10: Rating for EGI-TF 2011 User Tutorial Comments

What did you most like about this training? • “The availability of the teachers.” • “discover arc and unicore” • “Comparison of the three m/wares on one place.” What did you least like about this training? • “This was a lot to do in one hour and an half.” • “not enough electric plugs”

Which of the training subjects were the most useful to you? • “The ARC part.” Which of the training subjects did you find the least useful? • “The voms-proxy part.” Is there anything else you would like the training to have covered? • “I can't say as I haven't finished it.”

ANNEX D: OFFICIAL DCI SUMMER SCHOOL REPORT


Training Event Summary Report

Title of Training Event: Joint European DCI Summer School 2011

Date: 11/07/2011- 16/07/2011

Expected Learning Outcome: To give insights to the technologies provided by the EMI, IGE, EDGI, , DEGISCO and StratusLab, EGI-InSpire projects.

Language(s) of talks: English

Trainer names: 6 different projects/lecturers EMI: Morris, Riedel, Emidio Giorgio, Kathryn Cassidy. IGE: Florian Zerner, Adrian Colesa, Ioan Lucian Muntean

EDGI: Peter Kacsuk, Ad Emmen, Robert Lovas, Zoltan Farkas, Jozsef Kovacs. DEGISCO: Ad Emmen, Sandor Acs, Robert Lovas, Kitti Varga StratusLab: Mohammed Airaj, Evangelos Angelou

EGI-InSpire: Szabolcs Hernath, Edina Varga, Agnes Szeberenyi

Location: SZTAKI Laboratory of Parallel and Distributed Systems, Budapest, Hungary

URL (AgendaMaker): http://indico.lpds.sztaki.hu/indico/conferenceDisplay.py?confId=10

Length of event: 6 days

Local URL (pictures of events, etc): http://www.lpds.sztaki.hu/eudciss2011/


The following table lists the organisations the participants came from, the country of each organisation, and how many participants came from each organisation.

Organisation, Country Number of Participants

IRZ, Germany 1

Technical University of Cluj-Napoca, Romania 4

Korea Institute of Science and Technology Information, South Korea 2

LRI/CNRS, France 2

GRNET, Greece 1

University of Westminster, UK 1

HIAST, Syria 3

Eötvös Lóránd University Budapest, Hungary 1

Budapest University of Technology and Economics, Hungary 2

SZTAKI Laboratory of Parallel and Distributed Systems, Hungary 11

Juelich Supercomputing Centre, Germany 1

Grid-Ireland, EMI, Ireland 1

AlmereGrid, the Netherlands 1

NIIF, Hungary 2

INFN, Italy 1

CongressTeam 2000 Ltd., Hungary 1

KFKI Research Institute for Particle and Nuclear Physics, Hungary 1

Total number of participants: 36 Total number of countries: 11 Total number of institutes :17

This section shows the answers to the feedback question “male or female?” which is required under EU guidelines.

Male: 28 Female: 8 Didn't answer: 0


THE DISTRIBUTION OF THE STUDENTS/ EACH PROJECT

The table below gives you an overview on how many students were sent by each projects. Student Participants/represented Projects.

Name of the project Number of Students

EMI ?

IGE 4

EDGI 5

DEGISCO 4

StratusLab 1

EGI-InSpire 3

no related projects 1

sum 18

OVERALL EVALUATION

The following table and charts show an analysis of the feedback scores. The n=xxxx refers to the number of responses received.

Course Component Average value (1: poor - 6:excellent)

Standard Deviation (n=xxx)

Goal 1: give insight of the EMI, IGE, EDGI,

DEGISCO, Stratuslab, EGI-InSpire projects

5 20

Overall 4,8 20

Presentations 4,7 19

Lab exercises 4,6 20

Advertising & Registration 4,8 19

Facilities 5,1 19



EVALUATION PER DAY:

The charts below illustrate the ratings given each day by the participants (20 feedback forms have been completed).

FURTHER COMMENTS:

1. The need of a social event/trip at the beginning of the school (like on Thursday).

2. It is very tiring to focus all day long.

3. For the EMI hands-on: a more detailed description is needed.

4. StratusLab hands-on: problems with the infrastructure, only one tutor helping everyone, that made it very slow.

5. No differentiation between beginners and advanced users. Maybe optional tasks would have been useful for the advanced users.


6. Some of the long presentations were repeating concepts.

7. Some hands-on tasks were too easy, some were too confusing.

8. For the DEGISCO day’s hands-on there were no materials, only the website, which is not good enough.

9. No big picture given at the end of the DEGISCO and StratusLab day.

DNA2.2.2 - Training Plan - CERN...INFSO-RI-261611 2010-2011 © Members of EMI collaboration PUBLIC 1 / 54 EUROPEAN MIDDLEWARE INITIATIVE. DNA2.2.2 - T. RAINING . P. LAN. EU DELIVERABLE:

Documents