Case Study Overview

www.appistry.comThe Fabric of Business

Simplifying Development of High-Volume

Geospatial Intelligence Applications

Bob Lozano

Rocket City Geospatial Conference 2007

1


Case Study Overview

• Data: Raw satellite image is retrieved from SAN and broken up into tiles

• Process: Tiles are processed using proprietary GeoEye algorithms for sharpening, geocorrection, etc.

• Result: Tiles are reassembled and stored back in SAN

Challenges:

• Multi-core / SMP development complexity• Risk, cost and agility of traditional platforms• Meeting customer SLAs

Results:

• Imaging applications now able to process in excess of 5 TB of satellite imagery per day

• Developers able to focus on core competencies• Capital savings greater than $1.2 million• Easily meet customer requirements for

maximum processing time

Appistry EAF at GeoEye

Data Process ResultThe leading provider of satellite imagery for government and commercial applications, GeoEye is building its next-generation image processing applications on Appistry EAF.

“ By relying on the application fabric to provide scalability, reliability and manageability, we can leave our infrastructure concerns behind and focus on providing maximum value to our customers. ” – Ray Helmering,VP Photogrammetric Engineering at GeoEye

2


VP of Engineering: My ChallengesEnormous increases in data flow …

• New satellites (data

sources)

• Sophisticated analytics

• Internet distribution

• New business models

• System to System (SOA)

• Customer Expectations

•Mobile Access

• Constant change

DEPARTMENTALAPPLICATIONS

ENTERPRISEAPPLICATIONS

DA

TA

INT

EN

SE

(VO

LU

ME

)

CPU INTENSIVE(# OF SERVERS)3


Appistry Enterprise Application Fabric A grid-based application platform that dramatically simplifies the development

and deployment of agile applications for high-volume data/transaction processing

Well, take a look at an “application fabric”

4


“Googlization” Simplified

• Effortless scale, on-demand• Decreased time-to-market• Agile & resilient• Commoditized infrastructure• Pervasive Web services• Unified deployment model across

technology platforms

5


Traditional Processing Pipeline

6

SAN

4x Fiber ChannelFile I/O

Automated Image Processing System:Receive XML WorkOrderRead Input Images from SAN ~17 GBSharpen ImageGeocorrect ImageCompress ImageWrite results to SANGenerate XML WorkerOrderResponse

Upstream System

XML Work Order

High End Silicon Graphics Server:64 Processors256 Gigbytes RAMAltix OSApproximate Cost: $2,000,000

typical of the industry


Start With Application Fabrics

7

SANSAN

Tiling FabricTransformation Fabrics

I/O sc

ala

bility

Processing scalability


Psychological Sticking Points …

• I/O Scalability• Development Complexity• Operational Complexity• Reliability

• and don’t forget the cost reductions …

reality favors the application fabric in every dimension!

8

→application fabric→application fabric→application fabric→application fabric


Appistry EAF EnergySaver

• New add-on product for fabric power management

• Customer sets policies for minimizing energy consumption, while ensuring application performance & service levels

• No impact to application developers or SLAs!

9


Conclusion

Application fabrics, increase capability, lower cost, reduce time to market—i.e., are better in every dimension—and are very well suited to geospatial intelligence applications.

www.appistry.comThe Fabric of BusinessTHE FABRIC OF BUSINESS

“ By relying on the application fabric to provide scalability, reliability and manageability, we can leave our infrastructure concerns behind and focus on providing maximum value to our customers. ”

-Ray Helmering, VP Photogrammetric Engineering GeoEye

www.appistry.comThe Fabric of BusinessTHE FABRIC OF BUSINESS

Join in the conversation at www.appistry.com/blogs/bob

Documents

Case Study Overview