M2M Platform-as-a-Service for Sustainability Governance Hong-Linh Truong and Schahram Dustdar Distributed Systems Group Vienna University of Technology [email protected] http://pc3l.infosys.tuwien.ac.at 1 SOCA 2012, 18 Dec 2012, Taipei, Taiwan
Apr 22, 2015
M2M Platform-as-a-Service for
Sustainability Governance
Hong-Linh Truong and Schahram Dustdar
Distributed Systems Group
Vienna University of Technology
[email protected] http://pc3l.infosys.tuwien.ac.at
1 SOCA 2012, 18 Dec 2012, Taipei,
Taiwan
Outline
Context, motivation, and approach
Linking M2M data
Platform as a service
Prototype
Conclusions and future work
SOCA 2012, 18 Dec 2012,
Taipei, Taiwan
2
The context – sustainability governance
SOCA 2012, 18 Dec 2012, Taipei,
Taiwan 3
Cities, e.g. including:
10000+ buildings
1000000+ sensors
Near realtime analytics
Predictive data
analytics
Visual Analytics
Enterprise
Resource
Planning
Emergency
Management
Internet/public cloud
boundary
Organization-specific
boundary
Tracking/Log
istics
Infrastructure
Monitoring
Infrastructure/Internet of Things
...
Motivation (1)
SOCA 2012, 18 Dec 2012,
Taipei, Taiwan
4
Multiple phases, different data
gathering processes, different types
of data
Big and near-realtime data
Different types of analytics
Not a single programmig
language/model
Covering simple to complex
applications
Hong Linh Truong, Schahram Dustdar: A survey on cloud-based sustainability governance systems. IJWIS 8(3): 278-295 (2012)
Motivation (2)
SOCA 2012, 18 Dec 2012, Taipei,
Taiwan 5
Low-level (big sensor-based) cloud-based data infrastructures and
analytics platforms for single type of stakeholders are not enough
Only a few cloud-based infrastructures are investigated for managing
low-level data for sustainability governance
(Open) e-science data or sensor Web platforms mainly support one type
of stakeholders
A small example
Approach – Platform as a Service
Link near-realtime monitoring data with facility
monitored objects
Using linked data models and leveraging data services
for monitoring data and for monitored object information
Manual/automatic processes to establish the links
Develop data-as-a-service and platform-as-a-
service concepts for sustainabiltiy governance
Support near-realtime and predictive analytics
Different application models and bot-as-a-service
SOCA 2012, 18 Dec 2012,
Taipei, Taiwan
6
Linking cloud-based M2M data
SOCA 2012, 18 Dec 2012, Taipei,
Taiwan 7
Different situations in realistic systems:
Monitored object descriptions are/are not well-defined
Monitored object information might or might not
available
Sensor data can/cannot be annotated
DaaS for sustainability governance
Monitoring data Data-as-a-Service
Facility information Data-as-a-Service
SOCA 2012, 18 Dec 2012,
Taipei, Taiwan
8
Platform-as-a-Service for
sustainability governance
Different analytics application models, such as
batch, workflow and stream applications and
intelligent bots
different programming models and languages
offline predictive analytics of large-scale data but also
near-realtime analytics and bot-as-a-service
SOCA 2012, 18 Dec 2012,
Taipei, Taiwan
9
Platform-as-a-Service and Bots
SOCA 2012, 18 Dec 2012, Taipei,
Taiwan 10
Hong Linh Truong, Phu H. Phung, Schahram Dustdar: Governing Bot-as-a-Service in Sustainability Platforms - Issues
and Approaches. Procedia CS 10: 561-568 (2012)
Cloud-based sustainability
governance analysis framework
SOCA 2012, 18 Dec 2012, Taipei,
Taiwan 11
Prototype
Near-realtime monitoring data are obtained from
Niagara AX gateways, part of the Pacific
Controls Galaxy Platform
http://www.pacificcontrols.net/products/galaxy.html
An RDF-based data service for buiding
concepts and links
SusGov Apps profiles are in RDF
Using Allergro Graph
(http://www.franz.com/agraph/allegrograph)
Java-based PaaS with RESTful APIs
SOCA 2012, 18 Dec 2012,
Taipei, Taiwan
12
Linking M2M Cloud data - example
SOCA 2012, 18 Dec 2012, Taipei,
Taiwan 13
Cloud-based sustainability
governance analysis framework
SOCA 2012, 18 Dec 2012, Taipei,
Taiwan 14
Data dependencies
Application discovery
Local execution environment
Results
Conclusions and Future Work
We present
Techniques to link monitoring data and monitored
objects in cloud-based M2M systems
Platform-as-a-Service and data services for different
types of data analytics required by different
stakeholders
Future plan
Large-scale tests
Dynamic near-realtime analytics by combining bots
and cloud predictive data analytics
SOCA 2012, 18 Dec 2012,
Taipei, Taiwan
15
16
Thanks for your attention
Hong-Linh Truong
Distributed Systems Group
Vienna University of Technology
http://www.infosys.tuwien.ac.at/staff/truong
SOCA 2012, 18 Dec 2012, Taipei,
Taiwan