Http:// FOSS4G: 52°North WPS Behind the buzz of Cloud Computing - 52°North Open Source Geoprocessing Software in the Clouds FOSS4G 2009.

http://www.52north.org/wpsFOSS4G: 52°North WPS

Behind the buzz of Cloud Computing-

52°North Open Source Geoprocessing Software in the Clouds

FOSS4G 2009

Bastian Schäffer, 52°North



Agenda

• Cloud Computing 101

• WPS 101

• WPS in the Clouds


Cloud Computing 101


Definition

• Emerging trend in mainstream IT

• Cloud as a metaphor to represent large networks and infrastructures

• Store data, run applications and compute on distributed (3rd party) facilities

• Paradigm behind buzzword!?!


Characteristics

• Efficiency Resource Pooling

• Outtasking

• On-demand

• Scalability

• Cost


Mass market


Mass market


Mass market


Legally Binding Performance

• INSPIRE– Search Queries

• Maximum 3 sec response time*• Handle 30 parallel requests per second

– Image Download• Maximum 5 sec response time*• Handle 20 parallel requests per second

– Overall 99% service availability (362 days/year uptime)

* During normal traffic (90% of uptime).


Characteristics

• Efficiency Resource Pooling

• Outtasking

• On-demand

• Scalability

• Cost



IaaS

PaaS

dSaaSSaaS

Architecture


Technology

• Hardware Cluster

• Virtualization

• Grid Computing

• Application Framework (Hadoop)

• etc.


Cloud Types


Providers

- Amazon Web Services

- Google Apps Engine

- ….


Amazon Web Services• Small Instance

– The small instance (default) equates to "a system with 1.7 GB of memory, 1 EC2 Compute Unit (1 virtual core with 1 EC2 Compute Unit), 160 GB of instance storage, 32-bit platform"

• Large Instance– The large instance represents "a system with 7.5 GB of memory, 4

EC2 Compute Units (2 virtual cores with 2 EC2 Compute Units each), 850 GB of instance storage, 64-bit platform".

• Extra Large Instance– The extra large instance offers the "equivalent of a system with 15

GB of memory, 8 EC2 Compute Units (4 virtual cores with 2 EC2 Compute Units each), 1690 GB of instance storage, 64-bit platform."

• High-CPU Instance– Instances of this family have proportionally more CPU resources

than memory (RAM) and address compute-intensive applications.




• Cloud Watch• Auto Scaling

– CPUUtilization– NetworkIn– NetworkOut– DiskWriteOps– DiskReadBytes– DiskReadOps– DiskWriteBytes– RequestCount– HealthyHostCount– UnHealthyHostCount

- Elastic Cloud Balancing


Costs


Google Apps Engine


- Easy to manage

- Java or Python

- Several restrictions

- BigTable vs. RDBMS

Google Apps Engine


Costs


Costs


WPS 101 – What is WPS about?


OGC Web Services

Introduction

SOS


OGC Web Services

Introduction

SOS

WMS


OGC Web Services

Introduction

SOS

WMS

WFS


OGC Web Services

Introduction

SOS

WMS

WCS

WFS


OGC Web Services

Introduction

SOS

WMS

WCS

WFS

CSW


SDI Integration

• Processing


SDI Integration

• Processing


SDI Integration

• Processing


SDI Integration

• Processing


OGC Web Services

Introduction

SOS

WMS

WCS

WFS

CSW


SDI Integration

• Computational Power

• Network Bandwidth


SDI Integration

• Computational Power

• Network Bandwidth

serverside

processing


SDI Integration

• Processing


SDI Integration

• Processing Service


SDI Integration

• Processing ServiceStandardized Interface

unified accessDeploy-once-use-everywhere


SDI Building Blocks


OGC Web Services

Introduction

SOS

WMS

WCS

WFS

CSW


OGC Web Services

Introduction

SOS

WMS

WCS

WFS

CSW

WPS


Introduction

WPS -Introduction


WPS 101

What is a WPS?

Web Based Processing WPS

WPS


WPS

WPSWPS

Process 1

Process 2

Process n


WPS

WPS

GetCapabilities

WPS

Process 1

Process 2

Process n


WPS

WPS

GetCapabilities

DescribeProcess

WPS

Process 1

Process 2

Process n


WPS

WPS

GetCapabilities

DescribeProcess

Execute

WPS

Process 1

Process 2

Process n


WPS

Example

???

WPS

???

???

???


WPS

Example

???

WPS

Buffer

Process 2

Process n


???

WPS

Buffer

Process 2

Process n

WPS

Example

20


???

WPS

Buffer

Process 2

Process n

WPS

Example

20


Introduction

Technical Features


WPS additional features

Execution

Introduction


WPS additional features

ExecutionSynchronousAsynchronous

BindingGETPOSTSOAP

Introduction


Execution synchronous

Object1 Object2

Introduction


Execution asynchronous

Push-modelObject1 Object2

Introduction


Execution asynchronous

Pull-modelObject1 Object2

50%

reference

result

Introduction


Introduction

Clients


http://giv-bandog.uni-muenster.de:8081/OL/SpanishRoads.html


uDig

• CLIENTS


Jump

• CLIENTS


Performance Tests








Demo




Conclusion

• WPS use possible in cloud environments• Scaleability provided• Google Cloud

easy to use good for light-weight applicationGood integration with other google apps

• Amazon cloudMore complicated to useFull-scale solutionFlexible

• Cloud Interoperability present from client perspective (external)not from provider perspective (internal)


Conclusion

Future:

Impact on GIS Business Models– Selling of full GIS packages SaaS

– Other impacts on SDIs• Role model• Etc.


publishpublishfind

bind

bindGeoresource owner

Georesource owner

Georesource provider

Georesource provider


Conclusion

• THURSDAY:

13:30 Parkside G04

„Geoprocessing in the clouds“

introduction to geospatial layer in the cloud


Thank You

[email protected]

Http:// FOSS4G: 52°North WPS Behind the buzz of Cloud Computing - 52°North Open Source Geoprocessing Software in the Clouds FOSS4G 2009.

Documents

north wps slide

north slide

north wps costs

north wps mass market

north wps cloud computing

north wps cloud types

clouds foss4g

demand scalability cost