SlideShare a Scribd company logo
© 2008 IBM Corporation11 July 2008
IBM T.J. Watson Research Center
Cell/B.E. Servers: A Platform for Real Time
Scalable Computing and Visualization
Bruce D’Amora
Emerging Systems Software
Watson Research Center
Yorktown Heights, NY
2
IBM T.J. Watson Research Center
CBE for High Performance Computing and Scalable Visualization | 11 July 2008
© 2008 IBM Corporation
SW
SW
SW
SW
Workstation deployments are
complex, costly and limit
application scalability
Net-
works
Data SW
Data Center Users
Net-
work
Data
Net-
work
Data
Net-
work
Data
Net-
work
Data
Est. TCO: $2000/month per workstation
3
IBM T.J. Watson Research Center
CBE for High Performance Computing and Scalable Visualization | 11 July 2008
© 2008 IBM Corporation
Interest is converging toward remote, virtualized desktops
Compute intensive solutions
– Interactive Graphics and Dynamic Simulation require performance
scalability not achievable on a workstation
Large secure data models
Cost of ownership driving toward centrally
managed application solutions
Collaborative, interactive design review solutions
– between partners; between teams
4
IBM T.J. Watson Research Center
CBE for High Performance Computing and Scalable Visualization | 11 July 2008
© 2008 IBM Corporation
Why now?
Better network connectivity
Pressure to lower costs
Highly replicated worker stations
Globalized teaming
IP & data confidentiality protection afforded by centralization
Changing cost structure
– SW vs. HW, People vs. equipment
Microprocessor design trends: vectorization and parallelization
Applications are being re-written to take advantage of
multi-threaded and ultimately distributed platforms
5
IBM T.J. Watson Research Center
CBE for High Performance Computing and Scalable Visualization | 11 July 2008
© 2008 IBM Corporation
Scalable, high performance graphics and simulation
Interactive or real-time response time which
generally requires
Low latency and high bandwidth communications
I/O (network, storage system, etc.)
memory
The biggest inhibitors to remote scalable
application solutions are the need for:
6
IBM T.J. Watson Research Center
CBE for High Performance Computing and Scalable Visualization | 11 July 2008
© 2008 IBM Corporation
Cell/B.E. is a building block for homogeneous and
heterogeneous HPC platforms
Homogeneous systems may be appropriate
– HPC applications like seismic processing where large batches of data must be
processed by proprietary image processing pipelines
Heterogeneous application design points
– Client/server systems splitting application so compute intensive tasks on
server; GUI on PC/workstation/handheld
– Application components executing on different server nodes
Accelerator strategy becoming more prevalent
– Cell/B.E., Power, Intel mixed server nodes
– PCI-express accelerators such as GPUs
7
IBM T.J. Watson Research Center
CBE for High Performance Computing and Scalable Visualization | 11 July 2008
© 2008 IBM Corporation
IBM® PowerXCell™ 8i Processor Accelerator Board
IBM
PowerXCell
8i Processor
IBM
South
Bridge
Soldered
DDR
Soldered
DDR
PCI E Connector
Ethernet
IBM
South-
Bridge
IBM
PowerXCell
8i Processor
(2.8 GHz)
DDR2
Memory
Bcom
5704
PCI-x
PCI-e
x16
1 Gb E
1 Gb E
PCI-e Form Factor Board
IBM PowerXCell 8i
processor
2.8 GHz
2 or 4 GB DDR2
I/O
– One PCI-e x16 Bus
– Two 1 Gb Ethernet
– 8 General Purpose I/O’s
Software
– IBM SDK V 3.0
– Fedora 7
8
IBM T.J. Watson Research Center
CBE for High Performance Computing and Scalable Visualization | 11 July 2008
© 2008 IBM Corporation
System Z and Cell/B.E. Vision is a ‘Marriage’ of Two
Technologies that Perfectly Complement Each Other
Integrated and/
or networked
Cell/B.E.
Solution Characteristics
– Mission critical functionality
– Significant amounts of data storage
– Significant transactional content
– Computational intense operations
– Critical need for governance, i.e. security,
auditing, compliance, availability
– High potential for integration with other
applications
– Reduced power and space
Examples
– Financial analytics
– Medical Imaging
– Computer Vision
– Bioinformatics
– Real Time Ray Tracing
– Virtual Worlds
Z today Z tomorrow
QS20, QS21,
QS2x
Preserves the same
programming model between
Network & Integrated
Cell Blade
Digital
Media
Financial
Services
Sector
Digital Video
Surveillance
Aerospace
and Defense
Electronic
Design
Automation
Chemicals &
Petroleum
Information Based
Medicine
Courtesy of Hoplon
9
IBM T.J. Watson Research Center
CBE for High Performance Computing and Scalable Visualization | 11 July 2008
© 2008 IBM Corporation
Cell/B.E. plus AMD equals Roadrunner
Cluster of 18 Connected Units (CU)
• 6,912 AMD dual-core Opterons
• 12,960 IBM Cell/B.E. eDP accelerators
• 9.8 teraflops peak (Opteron)
• 1.33 petaflops peak (Cell/B.E. eDP)
• 1PF sustained Linpack
InfiniBand 4x DDR fabric
• all-optical cables
• 384 GB/s (bi-directional)
• 3.45 TB/s (bidirectional)
80 TB aggregate memory
• 28 TB Opteron
• 52 TB Cell/B.E.
• 216 GB/s sustained File System I/O:
• 216x2 10G Ethernets to Panasas
Software
RHEL and Fedora Linux
IBM SDK for Multicore Acceleration
xCAT Cluster Management
• System-wide GbE network
3.9 MW Power:
• 0.35 GF/Watt
Area:
• 296 racks
• 5500 ft2
10
IBM T.J. Watson Research Center
CBE for High Performance Computing and Scalable Visualization | 11 July 2008
© 2008 IBM Corporation
Cell/B.E. Real-time Application Solutions
PC/Workstations,
consoles,
handhelds
Scalable Network BackboneGbE/IB
IB/GbE
BC-H
IBM TotalStorage
IBMQS2x
IBMQS2x
IBMQS2x
IBMHS2x
IBMQS2x
IBMQS2x
IBMQS2x
IBMQS2x
IBMQS2x
IBMHS2x
IBMQS2x
IBMQS2x
IBMQS2x
IBMQS2x
IBMQS2x
IBMHS2x
IBMQS2x
IBMQS2x
Online Gaming and Rendering
BladeCenter H chassis rack mounted, 4 per rack
Heterogeneous solutions with Cell/B.E. and Intel blade servers
GbE or Infiniband attached storage
PC, Workstation, Game Consoles, Settop box, Handheld clients
Wired or wirelessly connected
11
IBM T.J. Watson Research Center
CBE for High Performance Computing and Scalable Visualization | 11 July 2008
© 2008 IBM Corporation
Game Engine based on Cell/B.E. servers and Intel workstations
Enables the next generation virtual environments
– Current online video games engines perform limited amount of physical simulation
– Not enough client CPU resources
– No distributed physics
– Bandwidth & Latency between processing nodes prohibitive to achieving real time performance
Enable complex visual F/X movies on large servers - Machinima
12
IBM T.J. Watson Research Center
CBE for High Performance Computing and Scalable Visualization | 11 July 2008
© 2008 IBM Corporation
Intel/AMD PCs
Game Consoles
Network
Scene
Management
Simulation
User
Input
Client
Server
Server Model
of World
State
Users View
Client Model
of World
State
Entity State Input
from Server
World State Input
from Client
Entity State
Verification
Input from
Client
Simulation
Online Game Client/Server Execution Flow
Rendering
Rendering
Composite
Render Input
from Server
IBM QS2x Blades
13
IBM T.J. Watson Research Center
CBE for High Performance Computing and Scalable Visualization | 11 July 2008
© 2008 IBM Corporation
Scaling Virtual Worlds
Game Framework
Network Manager
Player Login
Event Handlers Client
Intel/AMD PCs
Consoles
Handhelds
Client Renderer
Interpolator
Local Database
Event Handler
IBM Cell/B.E. Servers
Higher compute power required
.
.
Rigid Body
Zone0Zone1Zonen
Collisions
Rigid Body Collisions
Network Compression
Arithmetic Coding
Network Compression
Arithmetic Coding
Divide Virtual
world into
geographic
zones
Intel/AMD/Power Servers
Lower compute power required
Virtual World
State Manager
Rigid Body Collisions
Network Compression
Arithmetic Coding
Virtual World
State Manager
Virtual World
State Manager
Server-side State: Distributed Model
14
IBM T.J. Watson Research Center
CBE for High Performance Computing and Scalable Visualization | 11 July 2008
© 2008 IBM Corporation
Scaling Virtual Worlds
IBM Cell/B.E. Servers
Higher compute power required
.
.
Virtual World
State Manager
Game Framework
Network Manager
Blade 0
Rigid Body
Blade 1
Blade n
Node0Node1Noden
Client Renderer
Interpolator
Local Database
Collisions
Rigid Body Collisions
Rigid Body Collisions
Player Login
Event Handlers
Network Compression
Arithmetic Coding
Network Compression
Arithmetic Coding
Network Compression
Arithmetic Coding
Dispatch work (elements
of Virtual World) to
Compute Nodes
Event Handler
Intel/AMD/Power/Z Servers
Lower compute power required
More transaction processing
Client
Intel/AMD PCs
Consoles
Handhelds
Server-side State: Centralized Model
15
IBM T.J. Watson Research Center
CBE for High Performance Computing and Scalable Visualization | 11 July 2008
© 2008 IBM Corporation
Server Side Physics Loop on Cell/B.E. Servers
Numerical Integration to obtain
positions of bodies for next step
Sparse Matrix solvers
Perform very well on Cell/B.E.
Collision detection
Bounding box culling
Sorting
Point/triangle
intersection
Even sorting has proven to be very fast on Cell SPUs !!
16
IBM T.J. Watson Research Center
CBE for High Performance Computing and Scalable Visualization | 11 July 2008
© 2008 IBM Corporation
UCSD Scalable City Project (Sheldon Brown, UCSD)
17
IBM T.J. Watson Research Center
CBE for High Performance Computing and Scalable Visualization | 11 July 2008
© 2008 IBM Corporation
Scalable City Pipeline: Landscape creation
18
IBM T.J. Watson Research Center
CBE for High Performance Computing and Scalable Visualization | 11 July 2008
© 2008 IBM Corporation
Scalable City Pipeline: Border Detection
19
IBM T.J. Watson Research Center
CBE for High Performance Computing and Scalable Visualization | 11 July 2008
© 2008 IBM Corporation
Scalable City Pipeline: Space Filling Curve is GeneratedScalable City Pipeline: Space Filling Curve is Generated
20
IBM T.J. Watson Research Center
CBE for High Performance Computing and Scalable Visualization | 11 July 2008
© 2008 IBM Corporation
Scalable City Pipeline: Simple Curves Converted into Road Geometry
21
IBM T.J. Watson Research Center
CBE for High Performance Computing and Scalable Visualization | 11 July 2008
© 2008 IBM Corporation
Before use of cell, this was pre-
computed (before game play
session)
Roads Integrated into Landscape
22
IBM T.J. Watson Research Center
CBE for High Performance Computing and Scalable Visualization | 11 July 2008
© 2008 IBM Corporation
Cell based computation ofCell based computation of
similar pipelinesimilar pipeline –– takes place intakes place in
about 1 second using one bladeabout 1 second using one blade
per landscape (per landscape (demodemo))
23
IBM T.J. Watson Research Center
CBE for High Performance Computing and Scalable Visualization | 11 July 2008
© 2008 IBM Corporation
IRT Rendering Engine: IBM demo technology
24
IBM T.J. Watson Research Center
CBE for High Performance Computing and Scalable Visualization | 11 July 2008
© 2008 IBM Corporation
SPE Partitions
Blade 0
Blade 1
Blade 2
Blade 3
Blade 4
Blade 5
Blade 6
GDC 07 Configuration, Seven QS20 Blades, 3.2 TFLOPS
iRT Blade Performance Scaling
0
2
4
6
8
10
12
14
16
18
20
1 2 3 4 5 6 7
Blades
Frames/Sec
1080p 1
Realtime ray tracing including
• Reflection/transparency
• detailed shadows
• ambient occlusion
• 4x4 jittered multi-sampling
Totaling up to 288 rays/pixel
Lamborghini car model
• 1.6 million polygons
• 1080p hi-def.
• demo
Also Shown at GDC
March 2007
• PS3 node w/ seven
QS20 blades
• 3 Million Triangle
Urban Landscape
• >100 Textures
• 1080p HDTV
GigabitNetwork
Scalability at processor and system level
25
IBM T.J. Watson Research Center
CBE for High Performance Computing and Scalable Visualization | 11 July 2008
© 2008 IBM Corporation
Cell/B.E. vs. G80 –
Ray tracing
Performance
IBM QS20 Blade (Two 3.2 GHz Cell Processors) - IBM Demo ray tracer
Despite comparable theoretical floating point performance:
– One Cell/BE is 4-5x faster than G80.
– One QS20 blade (two Cell/BE processors) is 8-11x faster than G80
AMD Opteron, the defacto industry workhorse for rendering is easily overwhelmed by
Cell/B.E. or G80.
Secondary Ray processing – Incoherent nature makes them difficult to process on G80, but
they are easily handled by Cell/B.E.
Cell/B.E. is half the size of the G80 and produces five times the ray-tracing performance.
Left to Right:
•2.6 GHz AMD Opteron - Saarland Ray-
tracer
•nVIDIA GeForce 8800 GTX - Saarland
Ray-tracer
•Sony Playstation3 (partial 3.2 GHz
Cell/B.E. processor running Linux) - IBM
Demo
•3.2 GHz Cell/B.E. Processor - IBM Demo
26
IBM T.J. Watson Research Center
CBE for High Performance Computing and Scalable Visualization | 11 July 2008
© 2008 IBM Corporation
Cell/B.E. enables scalable, shared architecture
with full consumer to professional potential
Sony
PLAYSTATION©3
(Cell/B.E. + GPU)
IBM Cell/B.E.
Blade
(2 Cell/B.E.)
IBM Roadrunner
(16,000 Cell/B.E.
+ AMD)
Mercury 1u Dual Cell
Sony Cell/B.E.
Computing Unit
(Cell/B.E. + GPU +
AV I/O)
Consumer Professional
High Perf
ComputingBusiness
PowerXcell 8i PCI card
(Cell/B.E. + Host)
Common OS’s, Infrastructure, Tools, Libraries, Code…
Toshiba Spurs
Engine (SPU’s.
+ Host)
27
IBM T.J. Watson Research Center
CBE for High Performance Computing and Scalable Visualization | 11 July 2008
© 2008 IBM Corporation
© Copyright International Business Machines Corporation 2007
All Rights Reserved
This document was developed for IBM offerings in the United States as of the date of publication. IBM may not make these offerings available in
other countries, and the information is subject to change without notice. Consult your local IBM business contact for information on the IBM offerings
available in your area. In no event will IBM be liable for damages arising directly or indirectly from any use of the information contained in this
document.
Information in this document concerning non-IBM products was obtained from the suppliers of these products or other public sources. Questions on
the capabilities of non-IBM products should be addressed to the suppliers of those products.
IBM may have patents or pending patent applications covering subject matter in this document. The furnishing of this document does not give you
any license to these patents. Send license inquires, in writing, to IBM Director of Licensing, IBM Corporation, New Castle Drive, Armonk, NY 10504-
1785 USA.
All statements regarding IBM future direction and intent are subject to change or withdrawal without notice, and represent goals and objectives only.
The information contained in this document has not been submitted to any formal IBM test and is provided "AS IS" with no warranties or guarantees
either expressed or implied.
All examples cited or described in this document are presented as illustrations of the manner in which some IBM products can be used and the
results that may be achieved. Actual environmental costs and performance characteristics will vary depending on individual client configurations and
conditions.
IBM Global Financing offerings are provided through IBM Credit Corporation in the United States and other IBM subsidiaries and divisions worldwide
to qualified commercial and government clients. Rates are based on a client's credit rating, financing terms, offering type, equipment type and
options, and may vary by country. Other restrictions may apply. Rates and offerings are subject to change, extension or withdrawal without notice.
IBM is not responsible for printing errors in this document that result in pricing or information inaccuracies.
All prices shown are IBM's United States suggested list prices and are subject to change without notice; reseller prices may vary.
IBM hardware products are manufactured from new parts, or new and serviceable used parts. Regardless, our warranty terms apply.
Many of the features described in this document are operating system dependent and may not be available on Linux. For more information, please
check: http://www.ibm.com/systems/p/software/whitepapers/linux_overview.html
Any performance data contained in this document was determined in a controlled environment. Actual results may vary significantly and are
dependent on many factors including system hardware configuration and software design and configuration. Some measurements quoted in this
document may have been made on development-level systems. There is no guarantee these measurements will be the same on generally-available
systems. Some measurements quoted in this document may have been estimated through extrapolation. Users of this document should verify the
applicable data for their specific environment.
Special Notices

More Related Content

What's hot

Cloud Slam Co D Presentation
Cloud Slam Co D PresentationCloud Slam Co D Presentation
Cloud Slam Co D Presentation
Srini Chari, PhD., MBA.
 
Met Office.PDF
Met Office.PDFMet Office.PDF
Met Office.PDF
Manojkumar Baranidharan
 
G111614 top-trends-sydney2019-v1910a
G111614 top-trends-sydney2019-v1910aG111614 top-trends-sydney2019-v1910a
G111614 top-trends-sydney2019-v1910a
Tony Pearson
 
Idc Reducing It Costs With Blades
Idc Reducing It Costs With BladesIdc Reducing It Costs With Blades
Idc Reducing It Costs With Blades
pankaj009
 
Cases Riverbed 2009
Cases Riverbed 2009Cases Riverbed 2009
Cases Riverbed 2009
INSPIRIT BRASIL
 
A Time Traveller's Guide to DB2: Technology Themes for 2014 and Beyond
A Time Traveller's Guide to DB2: Technology Themes for 2014 and BeyondA Time Traveller's Guide to DB2: Technology Themes for 2014 and Beyond
A Time Traveller's Guide to DB2: Technology Themes for 2014 and Beyond
Laura Hood
 
How to combine Db2 on Z, IBM Db2 Analytics Accelerator and IBM Machine Learni...
How to combine Db2 on Z, IBM Db2 Analytics Accelerator and IBM Machine Learni...How to combine Db2 on Z, IBM Db2 Analytics Accelerator and IBM Machine Learni...
How to combine Db2 on Z, IBM Db2 Analytics Accelerator and IBM Machine Learni...
Gustav Lundström
 
Key Note Session IDUG DB2 Seminar, 16th April London - Julian Stuhler .Trito...
Key Note Session  IDUG DB2 Seminar, 16th April London - Julian Stuhler .Trito...Key Note Session  IDUG DB2 Seminar, 16th April London - Julian Stuhler .Trito...
Key Note Session IDUG DB2 Seminar, 16th April London - Julian Stuhler .Trito...
Surekha Parekh
 
Solutions_using_Blades_ITO0108
Solutions_using_Blades_ITO0108Solutions_using_Blades_ITO0108
Solutions_using_Blades_ITO0108
H Nelson Stewart
 
S200516 copy-data-management-ist2020-v2001c
S200516 copy-data-management-ist2020-v2001cS200516 copy-data-management-ist2020-v2001c
S200516 copy-data-management-ist2020-v2001c
Tony Pearson
 
IBM mainframe access
IBM mainframe accessIBM mainframe access
IBM mainframe access
Maintec Technologies Inc
 
S200515 storage-insights-ist2020-v2001d
S200515 storage-insights-ist2020-v2001dS200515 storage-insights-ist2020-v2001d
S200515 storage-insights-ist2020-v2001d
Tony Pearson
 
Z109889 z4 r-storage-dfsms-vegas-v1910b
Z109889 z4 r-storage-dfsms-vegas-v1910bZ109889 z4 r-storage-dfsms-vegas-v1910b
Z109889 z4 r-storage-dfsms-vegas-v1910b
Tony Pearson
 
Blade Server Technology Daniel Nilles Herzing
Blade Server Technology  Daniel Nilles  HerzingBlade Server Technology  Daniel Nilles  Herzing
Blade Server Technology Daniel Nilles Herzing
Daniel Nilles
 
Koomeyondatacenterelectricityuse v9
Koomeyondatacenterelectricityuse v9Koomeyondatacenterelectricityuse v9
Koomeyondatacenterelectricityuse v9
Jonathan Koomey
 
OpenPOWER/POWER9 AI webinar
OpenPOWER/POWER9 AI webinar OpenPOWER/POWER9 AI webinar
OpenPOWER/POWER9 AI webinar
Ganesan Narayanasamy
 
Ibm db2update2019 machine learning and db2 ai
Ibm db2update2019 machine learning and db2 aiIbm db2update2019 machine learning and db2 ai
Ibm db2update2019 machine learning and db2 ai
Gustav Lundström
 
Collaborate07kmohiuddin
Collaborate07kmohiuddinCollaborate07kmohiuddin
Collaborate07kmohiuddin
Sal Marcus
 

What's hot (18)

Cloud Slam Co D Presentation
Cloud Slam Co D PresentationCloud Slam Co D Presentation
Cloud Slam Co D Presentation
 
Met Office.PDF
Met Office.PDFMet Office.PDF
Met Office.PDF
 
G111614 top-trends-sydney2019-v1910a
G111614 top-trends-sydney2019-v1910aG111614 top-trends-sydney2019-v1910a
G111614 top-trends-sydney2019-v1910a
 
Idc Reducing It Costs With Blades
Idc Reducing It Costs With BladesIdc Reducing It Costs With Blades
Idc Reducing It Costs With Blades
 
Cases Riverbed 2009
Cases Riverbed 2009Cases Riverbed 2009
Cases Riverbed 2009
 
A Time Traveller's Guide to DB2: Technology Themes for 2014 and Beyond
A Time Traveller's Guide to DB2: Technology Themes for 2014 and BeyondA Time Traveller's Guide to DB2: Technology Themes for 2014 and Beyond
A Time Traveller's Guide to DB2: Technology Themes for 2014 and Beyond
 
How to combine Db2 on Z, IBM Db2 Analytics Accelerator and IBM Machine Learni...
How to combine Db2 on Z, IBM Db2 Analytics Accelerator and IBM Machine Learni...How to combine Db2 on Z, IBM Db2 Analytics Accelerator and IBM Machine Learni...
How to combine Db2 on Z, IBM Db2 Analytics Accelerator and IBM Machine Learni...
 
Key Note Session IDUG DB2 Seminar, 16th April London - Julian Stuhler .Trito...
Key Note Session  IDUG DB2 Seminar, 16th April London - Julian Stuhler .Trito...Key Note Session  IDUG DB2 Seminar, 16th April London - Julian Stuhler .Trito...
Key Note Session IDUG DB2 Seminar, 16th April London - Julian Stuhler .Trito...
 
Solutions_using_Blades_ITO0108
Solutions_using_Blades_ITO0108Solutions_using_Blades_ITO0108
Solutions_using_Blades_ITO0108
 
S200516 copy-data-management-ist2020-v2001c
S200516 copy-data-management-ist2020-v2001cS200516 copy-data-management-ist2020-v2001c
S200516 copy-data-management-ist2020-v2001c
 
IBM mainframe access
IBM mainframe accessIBM mainframe access
IBM mainframe access
 
S200515 storage-insights-ist2020-v2001d
S200515 storage-insights-ist2020-v2001dS200515 storage-insights-ist2020-v2001d
S200515 storage-insights-ist2020-v2001d
 
Z109889 z4 r-storage-dfsms-vegas-v1910b
Z109889 z4 r-storage-dfsms-vegas-v1910bZ109889 z4 r-storage-dfsms-vegas-v1910b
Z109889 z4 r-storage-dfsms-vegas-v1910b
 
Blade Server Technology Daniel Nilles Herzing
Blade Server Technology  Daniel Nilles  HerzingBlade Server Technology  Daniel Nilles  Herzing
Blade Server Technology Daniel Nilles Herzing
 
Koomeyondatacenterelectricityuse v9
Koomeyondatacenterelectricityuse v9Koomeyondatacenterelectricityuse v9
Koomeyondatacenterelectricityuse v9
 
OpenPOWER/POWER9 AI webinar
OpenPOWER/POWER9 AI webinar OpenPOWER/POWER9 AI webinar
OpenPOWER/POWER9 AI webinar
 
Ibm db2update2019 machine learning and db2 ai
Ibm db2update2019 machine learning and db2 aiIbm db2update2019 machine learning and db2 ai
Ibm db2update2019 machine learning and db2 ai
 
Collaborate07kmohiuddin
Collaborate07kmohiuddinCollaborate07kmohiuddin
Collaborate07kmohiuddin
 

Similar to Cell/B.E. Servers: A Platform for Real Time Scalable Computing and Visualization

Cell Today and Tomorrow - IBM Systems and Technology Group
Cell Today and Tomorrow - IBM Systems and Technology GroupCell Today and Tomorrow - IBM Systems and Technology Group
Cell Today and Tomorrow - IBM Systems and Technology Group
Slide_N
 
Cell Broadband EngineTM: and Cell/B.E. based blade technology
Cell Broadband EngineTM: and Cell/B.E. based blade technologyCell Broadband EngineTM: and Cell/B.E. based blade technology
Cell Broadband EngineTM: and Cell/B.E. based blade technology
Slide_N
 
Accelerating Cloud Services - Intel
Accelerating Cloud Services - IntelAccelerating Cloud Services - Intel
Accelerating Cloud Services - Intel
Amazon Web Services
 
Micro Server Design - Open Compute Project
Micro Server Design - Open Compute ProjectMicro Server Design - Open Compute Project
Micro Server Design - Open Compute Project
Hitesh Jani
 
Cell Technology for Graphics and Visualization
Cell Technology for Graphics and VisualizationCell Technology for Graphics and Visualization
Cell Technology for Graphics and Visualization
Slide_N
 
10 zig combined presentation deck v3
10 zig combined presentation deck v310 zig combined presentation deck v3
10 zig combined presentation deck v3
Jennifer Phillips
 
Cisco at v mworld 2015 vmworld sf 2015 brannon theater 20150829
Cisco at v mworld 2015 vmworld sf 2015 brannon theater 20150829Cisco at v mworld 2015 vmworld sf 2015 brannon theater 20150829
Cisco at v mworld 2015 vmworld sf 2015 brannon theater 20150829
ldangelo0772
 
IBM HPC Transformation with AI
IBM HPC Transformation with AI IBM HPC Transformation with AI
IBM HPC Transformation with AI
Ganesan Narayanasamy
 
Database as a Service - Tutorial @ICDE 2010
Database as a Service - Tutorial @ICDE 2010Database as a Service - Tutorial @ICDE 2010
Database as a Service - Tutorial @ICDE 2010
DBIS @ Ilmenau University of Technology
 
An energy, memory, and performance analysis
An energy, memory, and performance analysisAn energy, memory, and performance analysis
An energy, memory, and performance analysis
Elisabeth Stahl
 
Ch1 1
Ch1 1Ch1 1
IBM Special Announcement session Intel #IDF2013 September 10, 2013
IBM Special Announcement session Intel #IDF2013 September 10, 2013IBM Special Announcement session Intel #IDF2013 September 10, 2013
IBM Special Announcement session Intel #IDF2013 September 10, 2013
Cliff Kinard
 
Jan 2011 Presentation
Jan 2011 PresentationJan 2011 Presentation
Jan 2011 Presentation
RamanDua
 
Re-Imagining the Data Center with Intel
Re-Imagining the Data Center with IntelRe-Imagining the Data Center with Intel
Re-Imagining the Data Center with Intel
Intel IT Center
 
Windows Server 2008 R2 Hyper V
Windows Server 2008 R2 Hyper VWindows Server 2008 R2 Hyper V
Windows Server 2008 R2 Hyper V
Amit Gatenyo
 
“The Future of AI is Here Today: Deep Dive into Qualcomm’s On-Device AI Offer...
“The Future of AI is Here Today: Deep Dive into Qualcomm’s On-Device AI Offer...“The Future of AI is Here Today: Deep Dive into Qualcomm’s On-Device AI Offer...
“The Future of AI is Here Today: Deep Dive into Qualcomm’s On-Device AI Offer...
Edge AI and Vision Alliance
 
Michael Gschwind, Cell Broadband Engine: Exploiting multiple levels of parall...
Michael Gschwind, Cell Broadband Engine: Exploiting multiple levels of parall...Michael Gschwind, Cell Broadband Engine: Exploiting multiple levels of parall...
Michael Gschwind, Cell Broadband Engine: Exploiting multiple levels of parall...
Michael Gschwind
 
Deeplearningusingcloudpakfordata
DeeplearningusingcloudpakfordataDeeplearningusingcloudpakfordata
Deeplearningusingcloudpakfordata
Ganesan Narayanasamy
 
Ibm symp14 referentin_barbara koch_power_8 launch bk
Ibm symp14 referentin_barbara koch_power_8 launch bkIbm symp14 referentin_barbara koch_power_8 launch bk
Ibm symp14 referentin_barbara koch_power_8 launch bk
IBM Switzerland
 
From Grid to Cloud
From Grid to CloudFrom Grid to Cloud
From Grid to Cloud
gojkoadzic
 

Similar to Cell/B.E. Servers: A Platform for Real Time Scalable Computing and Visualization (20)

Cell Today and Tomorrow - IBM Systems and Technology Group
Cell Today and Tomorrow - IBM Systems and Technology GroupCell Today and Tomorrow - IBM Systems and Technology Group
Cell Today and Tomorrow - IBM Systems and Technology Group
 
Cell Broadband EngineTM: and Cell/B.E. based blade technology
Cell Broadband EngineTM: and Cell/B.E. based blade technologyCell Broadband EngineTM: and Cell/B.E. based blade technology
Cell Broadband EngineTM: and Cell/B.E. based blade technology
 
Accelerating Cloud Services - Intel
Accelerating Cloud Services - IntelAccelerating Cloud Services - Intel
Accelerating Cloud Services - Intel
 
Micro Server Design - Open Compute Project
Micro Server Design - Open Compute ProjectMicro Server Design - Open Compute Project
Micro Server Design - Open Compute Project
 
Cell Technology for Graphics and Visualization
Cell Technology for Graphics and VisualizationCell Technology for Graphics and Visualization
Cell Technology for Graphics and Visualization
 
10 zig combined presentation deck v3
10 zig combined presentation deck v310 zig combined presentation deck v3
10 zig combined presentation deck v3
 
Cisco at v mworld 2015 vmworld sf 2015 brannon theater 20150829
Cisco at v mworld 2015 vmworld sf 2015 brannon theater 20150829Cisco at v mworld 2015 vmworld sf 2015 brannon theater 20150829
Cisco at v mworld 2015 vmworld sf 2015 brannon theater 20150829
 
IBM HPC Transformation with AI
IBM HPC Transformation with AI IBM HPC Transformation with AI
IBM HPC Transformation with AI
 
Database as a Service - Tutorial @ICDE 2010
Database as a Service - Tutorial @ICDE 2010Database as a Service - Tutorial @ICDE 2010
Database as a Service - Tutorial @ICDE 2010
 
An energy, memory, and performance analysis
An energy, memory, and performance analysisAn energy, memory, and performance analysis
An energy, memory, and performance analysis
 
Ch1 1
Ch1 1Ch1 1
Ch1 1
 
IBM Special Announcement session Intel #IDF2013 September 10, 2013
IBM Special Announcement session Intel #IDF2013 September 10, 2013IBM Special Announcement session Intel #IDF2013 September 10, 2013
IBM Special Announcement session Intel #IDF2013 September 10, 2013
 
Jan 2011 Presentation
Jan 2011 PresentationJan 2011 Presentation
Jan 2011 Presentation
 
Re-Imagining the Data Center with Intel
Re-Imagining the Data Center with IntelRe-Imagining the Data Center with Intel
Re-Imagining the Data Center with Intel
 
Windows Server 2008 R2 Hyper V
Windows Server 2008 R2 Hyper VWindows Server 2008 R2 Hyper V
Windows Server 2008 R2 Hyper V
 
“The Future of AI is Here Today: Deep Dive into Qualcomm’s On-Device AI Offer...
“The Future of AI is Here Today: Deep Dive into Qualcomm’s On-Device AI Offer...“The Future of AI is Here Today: Deep Dive into Qualcomm’s On-Device AI Offer...
“The Future of AI is Here Today: Deep Dive into Qualcomm’s On-Device AI Offer...
 
Michael Gschwind, Cell Broadband Engine: Exploiting multiple levels of parall...
Michael Gschwind, Cell Broadband Engine: Exploiting multiple levels of parall...Michael Gschwind, Cell Broadband Engine: Exploiting multiple levels of parall...
Michael Gschwind, Cell Broadband Engine: Exploiting multiple levels of parall...
 
Deeplearningusingcloudpakfordata
DeeplearningusingcloudpakfordataDeeplearningusingcloudpakfordata
Deeplearningusingcloudpakfordata
 
Ibm symp14 referentin_barbara koch_power_8 launch bk
Ibm symp14 referentin_barbara koch_power_8 launch bkIbm symp14 referentin_barbara koch_power_8 launch bk
Ibm symp14 referentin_barbara koch_power_8 launch bk
 
From Grid to Cloud
From Grid to CloudFrom Grid to Cloud
From Grid to Cloud
 

More from Slide_N

Efficient Usage of Compute Shaders on Xbox One and PS4
Efficient Usage of Compute Shaders on Xbox One and PS4Efficient Usage of Compute Shaders on Xbox One and PS4
Efficient Usage of Compute Shaders on Xbox One and PS4
Slide_N
 
Future Commodity Chip Called CELL for HPC
Future Commodity Chip Called CELL for HPCFuture Commodity Chip Called CELL for HPC
Future Commodity Chip Called CELL for HPC
Slide_N
 
Common Software Models and Platform for Cell and SpursEngine
Common Software Models and Platform for Cell and SpursEngineCommon Software Models and Platform for Cell and SpursEngine
Common Software Models and Platform for Cell and SpursEngine
Slide_N
 
Toshiba's Approach to Consumer Product Applications by Cell and Desire/Challe...
Toshiba's Approach to Consumer Product Applications by Cell and Desire/Challe...Toshiba's Approach to Consumer Product Applications by Cell and Desire/Challe...
Toshiba's Approach to Consumer Product Applications by Cell and Desire/Challe...
Slide_N
 
Towards Cell Broadband Engine - Together with Playstation
Towards Cell Broadband Engine  - Together with PlaystationTowards Cell Broadband Engine  - Together with Playstation
Towards Cell Broadband Engine - Together with Playstation
Slide_N
 
SpursEngine A High-performance Stream Processor Derived from Cell/B.E. for Me...
SpursEngine A High-performance Stream Processor Derived from Cell/B.E. for Me...SpursEngine A High-performance Stream Processor Derived from Cell/B.E. for Me...
SpursEngine A High-performance Stream Processor Derived from Cell/B.E. for Me...
Slide_N
 
Parallel Vector Tile-Optimized Library (PVTOL) Architecture-v3.pdf
Parallel Vector Tile-Optimized Library (PVTOL) Architecture-v3.pdfParallel Vector Tile-Optimized Library (PVTOL) Architecture-v3.pdf
Parallel Vector Tile-Optimized Library (PVTOL) Architecture-v3.pdf
Slide_N
 
Experiences with PlayStation VR - Sony Interactive Entertainment
Experiences with PlayStation VR  - Sony Interactive EntertainmentExperiences with PlayStation VR  - Sony Interactive Entertainment
Experiences with PlayStation VR - Sony Interactive Entertainment
Slide_N
 
SPU-based Deferred Shading for Battlefield 3 on Playstation 3
SPU-based Deferred Shading for Battlefield 3 on Playstation 3SPU-based Deferred Shading for Battlefield 3 on Playstation 3
SPU-based Deferred Shading for Battlefield 3 on Playstation 3
Slide_N
 
Filtering Approaches for Real-Time Anti-Aliasing
Filtering Approaches for Real-Time Anti-AliasingFiltering Approaches for Real-Time Anti-Aliasing
Filtering Approaches for Real-Time Anti-Aliasing
Slide_N
 
Chip Multiprocessing and the Cell Broadband Engine.pdf
Chip Multiprocessing and the Cell Broadband Engine.pdfChip Multiprocessing and the Cell Broadband Engine.pdf
Chip Multiprocessing and the Cell Broadband Engine.pdf
Slide_N
 
New Millennium for Computer Entertainment - Kutaragi
New Millennium for Computer Entertainment - KutaragiNew Millennium for Computer Entertainment - Kutaragi
New Millennium for Computer Entertainment - Kutaragi
Slide_N
 
Sony Transformation 60 - Kutaragi
Sony Transformation 60 - KutaragiSony Transformation 60 - Kutaragi
Sony Transformation 60 - Kutaragi
Slide_N
 
Sony Transformation 60
Sony Transformation 60 Sony Transformation 60
Sony Transformation 60
Slide_N
 
Moving Innovative Game Technology from the Lab to the Living Room
Moving Innovative Game Technology from the Lab to the Living RoomMoving Innovative Game Technology from the Lab to the Living Room
Moving Innovative Game Technology from the Lab to the Living Room
Slide_N
 
The Technology behind PlayStation 2
The Technology behind PlayStation 2The Technology behind PlayStation 2
The Technology behind PlayStation 2
Slide_N
 
Industry Trends in Microprocessor Design
Industry Trends in Microprocessor DesignIndustry Trends in Microprocessor Design
Industry Trends in Microprocessor Design
Slide_N
 
Translating GPU Binaries to Tiered SIMD Architectures with Ocelot
Translating GPU Binaries to Tiered SIMD Architectures with OcelotTranslating GPU Binaries to Tiered SIMD Architectures with Ocelot
Translating GPU Binaries to Tiered SIMD Architectures with Ocelot
Slide_N
 
Cellular Neural Networks: Theory
Cellular Neural Networks: TheoryCellular Neural Networks: Theory
Cellular Neural Networks: Theory
Slide_N
 
Network Processing on an SPE Core in Cell Broadband EngineTM
Network Processing on an SPE Core in Cell Broadband EngineTMNetwork Processing on an SPE Core in Cell Broadband EngineTM
Network Processing on an SPE Core in Cell Broadband EngineTM
Slide_N
 

More from Slide_N (20)

Efficient Usage of Compute Shaders on Xbox One and PS4
Efficient Usage of Compute Shaders on Xbox One and PS4Efficient Usage of Compute Shaders on Xbox One and PS4
Efficient Usage of Compute Shaders on Xbox One and PS4
 
Future Commodity Chip Called CELL for HPC
Future Commodity Chip Called CELL for HPCFuture Commodity Chip Called CELL for HPC
Future Commodity Chip Called CELL for HPC
 
Common Software Models and Platform for Cell and SpursEngine
Common Software Models and Platform for Cell and SpursEngineCommon Software Models and Platform for Cell and SpursEngine
Common Software Models and Platform for Cell and SpursEngine
 
Toshiba's Approach to Consumer Product Applications by Cell and Desire/Challe...
Toshiba's Approach to Consumer Product Applications by Cell and Desire/Challe...Toshiba's Approach to Consumer Product Applications by Cell and Desire/Challe...
Toshiba's Approach to Consumer Product Applications by Cell and Desire/Challe...
 
Towards Cell Broadband Engine - Together with Playstation
Towards Cell Broadband Engine  - Together with PlaystationTowards Cell Broadband Engine  - Together with Playstation
Towards Cell Broadband Engine - Together with Playstation
 
SpursEngine A High-performance Stream Processor Derived from Cell/B.E. for Me...
SpursEngine A High-performance Stream Processor Derived from Cell/B.E. for Me...SpursEngine A High-performance Stream Processor Derived from Cell/B.E. for Me...
SpursEngine A High-performance Stream Processor Derived from Cell/B.E. for Me...
 
Parallel Vector Tile-Optimized Library (PVTOL) Architecture-v3.pdf
Parallel Vector Tile-Optimized Library (PVTOL) Architecture-v3.pdfParallel Vector Tile-Optimized Library (PVTOL) Architecture-v3.pdf
Parallel Vector Tile-Optimized Library (PVTOL) Architecture-v3.pdf
 
Experiences with PlayStation VR - Sony Interactive Entertainment
Experiences with PlayStation VR  - Sony Interactive EntertainmentExperiences with PlayStation VR  - Sony Interactive Entertainment
Experiences with PlayStation VR - Sony Interactive Entertainment
 
SPU-based Deferred Shading for Battlefield 3 on Playstation 3
SPU-based Deferred Shading for Battlefield 3 on Playstation 3SPU-based Deferred Shading for Battlefield 3 on Playstation 3
SPU-based Deferred Shading for Battlefield 3 on Playstation 3
 
Filtering Approaches for Real-Time Anti-Aliasing
Filtering Approaches for Real-Time Anti-AliasingFiltering Approaches for Real-Time Anti-Aliasing
Filtering Approaches for Real-Time Anti-Aliasing
 
Chip Multiprocessing and the Cell Broadband Engine.pdf
Chip Multiprocessing and the Cell Broadband Engine.pdfChip Multiprocessing and the Cell Broadband Engine.pdf
Chip Multiprocessing and the Cell Broadband Engine.pdf
 
New Millennium for Computer Entertainment - Kutaragi
New Millennium for Computer Entertainment - KutaragiNew Millennium for Computer Entertainment - Kutaragi
New Millennium for Computer Entertainment - Kutaragi
 
Sony Transformation 60 - Kutaragi
Sony Transformation 60 - KutaragiSony Transformation 60 - Kutaragi
Sony Transformation 60 - Kutaragi
 
Sony Transformation 60
Sony Transformation 60 Sony Transformation 60
Sony Transformation 60
 
Moving Innovative Game Technology from the Lab to the Living Room
Moving Innovative Game Technology from the Lab to the Living RoomMoving Innovative Game Technology from the Lab to the Living Room
Moving Innovative Game Technology from the Lab to the Living Room
 
The Technology behind PlayStation 2
The Technology behind PlayStation 2The Technology behind PlayStation 2
The Technology behind PlayStation 2
 
Industry Trends in Microprocessor Design
Industry Trends in Microprocessor DesignIndustry Trends in Microprocessor Design
Industry Trends in Microprocessor Design
 
Translating GPU Binaries to Tiered SIMD Architectures with Ocelot
Translating GPU Binaries to Tiered SIMD Architectures with OcelotTranslating GPU Binaries to Tiered SIMD Architectures with Ocelot
Translating GPU Binaries to Tiered SIMD Architectures with Ocelot
 
Cellular Neural Networks: Theory
Cellular Neural Networks: TheoryCellular Neural Networks: Theory
Cellular Neural Networks: Theory
 
Network Processing on an SPE Core in Cell Broadband EngineTM
Network Processing on an SPE Core in Cell Broadband EngineTMNetwork Processing on an SPE Core in Cell Broadband EngineTM
Network Processing on an SPE Core in Cell Broadband EngineTM
 

Recently uploaded

Building Production Ready Search Pipelines with Spark and Milvus
Building Production Ready Search Pipelines with Spark and MilvusBuilding Production Ready Search Pipelines with Spark and Milvus
Building Production Ready Search Pipelines with Spark and Milvus
Zilliz
 
Salesforce Integration for Bonterra Impact Management (fka Social Solutions A...
Salesforce Integration for Bonterra Impact Management (fka Social Solutions A...Salesforce Integration for Bonterra Impact Management (fka Social Solutions A...
Salesforce Integration for Bonterra Impact Management (fka Social Solutions A...
Jeffrey Haguewood
 
How to Interpret Trends in the Kalyan Rajdhani Mix Chart.pdf
How to Interpret Trends in the Kalyan Rajdhani Mix Chart.pdfHow to Interpret Trends in the Kalyan Rajdhani Mix Chart.pdf
How to Interpret Trends in the Kalyan Rajdhani Mix Chart.pdf
Chart Kalyan
 
Serial Arm Control in Real Time Presentation
Serial Arm Control in Real Time PresentationSerial Arm Control in Real Time Presentation
Serial Arm Control in Real Time Presentation
tolgahangng
 
GenAI Pilot Implementation in the organizations
GenAI Pilot Implementation in the organizationsGenAI Pilot Implementation in the organizations
GenAI Pilot Implementation in the organizations
kumardaparthi1024
 
WeTestAthens: Postman's AI & Automation Techniques
WeTestAthens: Postman's AI & Automation TechniquesWeTestAthens: Postman's AI & Automation Techniques
WeTestAthens: Postman's AI & Automation Techniques
Postman
 
Monitoring and Managing Anomaly Detection on OpenShift.pdf
Monitoring and Managing Anomaly Detection on OpenShift.pdfMonitoring and Managing Anomaly Detection on OpenShift.pdf
Monitoring and Managing Anomaly Detection on OpenShift.pdf
Tosin Akinosho
 
HCL Notes und Domino Lizenzkostenreduzierung in der Welt von DLAU
HCL Notes und Domino Lizenzkostenreduzierung in der Welt von DLAUHCL Notes und Domino Lizenzkostenreduzierung in der Welt von DLAU
HCL Notes und Domino Lizenzkostenreduzierung in der Welt von DLAU
panagenda
 
Skybuffer SAM4U tool for SAP license adoption
Skybuffer SAM4U tool for SAP license adoptionSkybuffer SAM4U tool for SAP license adoption
Skybuffer SAM4U tool for SAP license adoption
Tatiana Kojar
 
Main news related to the CCS TSI 2023 (2023/1695)
Main news related to the CCS TSI 2023 (2023/1695)Main news related to the CCS TSI 2023 (2023/1695)
Main news related to the CCS TSI 2023 (2023/1695)
Jakub Marek
 
AI 101: An Introduction to the Basics and Impact of Artificial Intelligence
AI 101: An Introduction to the Basics and Impact of Artificial IntelligenceAI 101: An Introduction to the Basics and Impact of Artificial Intelligence
AI 101: An Introduction to the Basics and Impact of Artificial Intelligence
IndexBug
 
5th LF Energy Power Grid Model Meet-up Slides
5th LF Energy Power Grid Model Meet-up Slides5th LF Energy Power Grid Model Meet-up Slides
5th LF Energy Power Grid Model Meet-up Slides
DanBrown980551
 
Project Management Semester Long Project - Acuity
Project Management Semester Long Project - AcuityProject Management Semester Long Project - Acuity
Project Management Semester Long Project - Acuity
jpupo2018
 
Ocean lotus Threat actors project by John Sitima 2024 (1).pptx
Ocean lotus Threat actors project by John Sitima 2024 (1).pptxOcean lotus Threat actors project by John Sitima 2024 (1).pptx
Ocean lotus Threat actors project by John Sitima 2024 (1).pptx
SitimaJohn
 
Introduction of Cybersecurity with OSS at Code Europe 2024
Introduction of Cybersecurity with OSS  at Code Europe 2024Introduction of Cybersecurity with OSS  at Code Europe 2024
Introduction of Cybersecurity with OSS at Code Europe 2024
Hiroshi SHIBATA
 
Fueling AI with Great Data with Airbyte Webinar
Fueling AI with Great Data with Airbyte WebinarFueling AI with Great Data with Airbyte Webinar
Fueling AI with Great Data with Airbyte Webinar
Zilliz
 
Choosing The Best AWS Service For Your Website + API.pptx
Choosing The Best AWS Service For Your Website + API.pptxChoosing The Best AWS Service For Your Website + API.pptx
Choosing The Best AWS Service For Your Website + API.pptx
Brandon Minnick, MBA
 
Mariano G Tinti - Decoding SpaceX
Mariano G Tinti - Decoding SpaceXMariano G Tinti - Decoding SpaceX
Mariano G Tinti - Decoding SpaceX
Mariano Tinti
 
Driving Business Innovation: Latest Generative AI Advancements & Success Story
Driving Business Innovation: Latest Generative AI Advancements & Success StoryDriving Business Innovation: Latest Generative AI Advancements & Success Story
Driving Business Innovation: Latest Generative AI Advancements & Success Story
Safe Software
 
OpenID AuthZEN Interop Read Out - Authorization
OpenID AuthZEN Interop Read Out - AuthorizationOpenID AuthZEN Interop Read Out - Authorization
OpenID AuthZEN Interop Read Out - Authorization
David Brossard
 

Recently uploaded (20)

Building Production Ready Search Pipelines with Spark and Milvus
Building Production Ready Search Pipelines with Spark and MilvusBuilding Production Ready Search Pipelines with Spark and Milvus
Building Production Ready Search Pipelines with Spark and Milvus
 
Salesforce Integration for Bonterra Impact Management (fka Social Solutions A...
Salesforce Integration for Bonterra Impact Management (fka Social Solutions A...Salesforce Integration for Bonterra Impact Management (fka Social Solutions A...
Salesforce Integration for Bonterra Impact Management (fka Social Solutions A...
 
How to Interpret Trends in the Kalyan Rajdhani Mix Chart.pdf
How to Interpret Trends in the Kalyan Rajdhani Mix Chart.pdfHow to Interpret Trends in the Kalyan Rajdhani Mix Chart.pdf
How to Interpret Trends in the Kalyan Rajdhani Mix Chart.pdf
 
Serial Arm Control in Real Time Presentation
Serial Arm Control in Real Time PresentationSerial Arm Control in Real Time Presentation
Serial Arm Control in Real Time Presentation
 
GenAI Pilot Implementation in the organizations
GenAI Pilot Implementation in the organizationsGenAI Pilot Implementation in the organizations
GenAI Pilot Implementation in the organizations
 
WeTestAthens: Postman's AI & Automation Techniques
WeTestAthens: Postman's AI & Automation TechniquesWeTestAthens: Postman's AI & Automation Techniques
WeTestAthens: Postman's AI & Automation Techniques
 
Monitoring and Managing Anomaly Detection on OpenShift.pdf
Monitoring and Managing Anomaly Detection on OpenShift.pdfMonitoring and Managing Anomaly Detection on OpenShift.pdf
Monitoring and Managing Anomaly Detection on OpenShift.pdf
 
HCL Notes und Domino Lizenzkostenreduzierung in der Welt von DLAU
HCL Notes und Domino Lizenzkostenreduzierung in der Welt von DLAUHCL Notes und Domino Lizenzkostenreduzierung in der Welt von DLAU
HCL Notes und Domino Lizenzkostenreduzierung in der Welt von DLAU
 
Skybuffer SAM4U tool for SAP license adoption
Skybuffer SAM4U tool for SAP license adoptionSkybuffer SAM4U tool for SAP license adoption
Skybuffer SAM4U tool for SAP license adoption
 
Main news related to the CCS TSI 2023 (2023/1695)
Main news related to the CCS TSI 2023 (2023/1695)Main news related to the CCS TSI 2023 (2023/1695)
Main news related to the CCS TSI 2023 (2023/1695)
 
AI 101: An Introduction to the Basics and Impact of Artificial Intelligence
AI 101: An Introduction to the Basics and Impact of Artificial IntelligenceAI 101: An Introduction to the Basics and Impact of Artificial Intelligence
AI 101: An Introduction to the Basics and Impact of Artificial Intelligence
 
5th LF Energy Power Grid Model Meet-up Slides
5th LF Energy Power Grid Model Meet-up Slides5th LF Energy Power Grid Model Meet-up Slides
5th LF Energy Power Grid Model Meet-up Slides
 
Project Management Semester Long Project - Acuity
Project Management Semester Long Project - AcuityProject Management Semester Long Project - Acuity
Project Management Semester Long Project - Acuity
 
Ocean lotus Threat actors project by John Sitima 2024 (1).pptx
Ocean lotus Threat actors project by John Sitima 2024 (1).pptxOcean lotus Threat actors project by John Sitima 2024 (1).pptx
Ocean lotus Threat actors project by John Sitima 2024 (1).pptx
 
Introduction of Cybersecurity with OSS at Code Europe 2024
Introduction of Cybersecurity with OSS  at Code Europe 2024Introduction of Cybersecurity with OSS  at Code Europe 2024
Introduction of Cybersecurity with OSS at Code Europe 2024
 
Fueling AI with Great Data with Airbyte Webinar
Fueling AI with Great Data with Airbyte WebinarFueling AI with Great Data with Airbyte Webinar
Fueling AI with Great Data with Airbyte Webinar
 
Choosing The Best AWS Service For Your Website + API.pptx
Choosing The Best AWS Service For Your Website + API.pptxChoosing The Best AWS Service For Your Website + API.pptx
Choosing The Best AWS Service For Your Website + API.pptx
 
Mariano G Tinti - Decoding SpaceX
Mariano G Tinti - Decoding SpaceXMariano G Tinti - Decoding SpaceX
Mariano G Tinti - Decoding SpaceX
 
Driving Business Innovation: Latest Generative AI Advancements & Success Story
Driving Business Innovation: Latest Generative AI Advancements & Success StoryDriving Business Innovation: Latest Generative AI Advancements & Success Story
Driving Business Innovation: Latest Generative AI Advancements & Success Story
 
OpenID AuthZEN Interop Read Out - Authorization
OpenID AuthZEN Interop Read Out - AuthorizationOpenID AuthZEN Interop Read Out - Authorization
OpenID AuthZEN Interop Read Out - Authorization
 

Cell/B.E. Servers: A Platform for Real Time Scalable Computing and Visualization

  • 1. © 2008 IBM Corporation11 July 2008 IBM T.J. Watson Research Center Cell/B.E. Servers: A Platform for Real Time Scalable Computing and Visualization Bruce D’Amora Emerging Systems Software Watson Research Center Yorktown Heights, NY
  • 2. 2 IBM T.J. Watson Research Center CBE for High Performance Computing and Scalable Visualization | 11 July 2008 © 2008 IBM Corporation SW SW SW SW Workstation deployments are complex, costly and limit application scalability Net- works Data SW Data Center Users Net- work Data Net- work Data Net- work Data Net- work Data Est. TCO: $2000/month per workstation
  • 3. 3 IBM T.J. Watson Research Center CBE for High Performance Computing and Scalable Visualization | 11 July 2008 © 2008 IBM Corporation Interest is converging toward remote, virtualized desktops Compute intensive solutions – Interactive Graphics and Dynamic Simulation require performance scalability not achievable on a workstation Large secure data models Cost of ownership driving toward centrally managed application solutions Collaborative, interactive design review solutions – between partners; between teams
  • 4. 4 IBM T.J. Watson Research Center CBE for High Performance Computing and Scalable Visualization | 11 July 2008 © 2008 IBM Corporation Why now? Better network connectivity Pressure to lower costs Highly replicated worker stations Globalized teaming IP & data confidentiality protection afforded by centralization Changing cost structure – SW vs. HW, People vs. equipment Microprocessor design trends: vectorization and parallelization Applications are being re-written to take advantage of multi-threaded and ultimately distributed platforms
  • 5. 5 IBM T.J. Watson Research Center CBE for High Performance Computing and Scalable Visualization | 11 July 2008 © 2008 IBM Corporation Scalable, high performance graphics and simulation Interactive or real-time response time which generally requires Low latency and high bandwidth communications I/O (network, storage system, etc.) memory The biggest inhibitors to remote scalable application solutions are the need for:
  • 6. 6 IBM T.J. Watson Research Center CBE for High Performance Computing and Scalable Visualization | 11 July 2008 © 2008 IBM Corporation Cell/B.E. is a building block for homogeneous and heterogeneous HPC platforms Homogeneous systems may be appropriate – HPC applications like seismic processing where large batches of data must be processed by proprietary image processing pipelines Heterogeneous application design points – Client/server systems splitting application so compute intensive tasks on server; GUI on PC/workstation/handheld – Application components executing on different server nodes Accelerator strategy becoming more prevalent – Cell/B.E., Power, Intel mixed server nodes – PCI-express accelerators such as GPUs
  • 7. 7 IBM T.J. Watson Research Center CBE for High Performance Computing and Scalable Visualization | 11 July 2008 © 2008 IBM Corporation IBM® PowerXCell™ 8i Processor Accelerator Board IBM PowerXCell 8i Processor IBM South Bridge Soldered DDR Soldered DDR PCI E Connector Ethernet IBM South- Bridge IBM PowerXCell 8i Processor (2.8 GHz) DDR2 Memory Bcom 5704 PCI-x PCI-e x16 1 Gb E 1 Gb E PCI-e Form Factor Board IBM PowerXCell 8i processor 2.8 GHz 2 or 4 GB DDR2 I/O – One PCI-e x16 Bus – Two 1 Gb Ethernet – 8 General Purpose I/O’s Software – IBM SDK V 3.0 – Fedora 7
  • 8. 8 IBM T.J. Watson Research Center CBE for High Performance Computing and Scalable Visualization | 11 July 2008 © 2008 IBM Corporation System Z and Cell/B.E. Vision is a ‘Marriage’ of Two Technologies that Perfectly Complement Each Other Integrated and/ or networked Cell/B.E. Solution Characteristics – Mission critical functionality – Significant amounts of data storage – Significant transactional content – Computational intense operations – Critical need for governance, i.e. security, auditing, compliance, availability – High potential for integration with other applications – Reduced power and space Examples – Financial analytics – Medical Imaging – Computer Vision – Bioinformatics – Real Time Ray Tracing – Virtual Worlds Z today Z tomorrow QS20, QS21, QS2x Preserves the same programming model between Network & Integrated Cell Blade Digital Media Financial Services Sector Digital Video Surveillance Aerospace and Defense Electronic Design Automation Chemicals & Petroleum Information Based Medicine Courtesy of Hoplon
  • 9. 9 IBM T.J. Watson Research Center CBE for High Performance Computing and Scalable Visualization | 11 July 2008 © 2008 IBM Corporation Cell/B.E. plus AMD equals Roadrunner Cluster of 18 Connected Units (CU) • 6,912 AMD dual-core Opterons • 12,960 IBM Cell/B.E. eDP accelerators • 9.8 teraflops peak (Opteron) • 1.33 petaflops peak (Cell/B.E. eDP) • 1PF sustained Linpack InfiniBand 4x DDR fabric • all-optical cables • 384 GB/s (bi-directional) • 3.45 TB/s (bidirectional) 80 TB aggregate memory • 28 TB Opteron • 52 TB Cell/B.E. • 216 GB/s sustained File System I/O: • 216x2 10G Ethernets to Panasas Software RHEL and Fedora Linux IBM SDK for Multicore Acceleration xCAT Cluster Management • System-wide GbE network 3.9 MW Power: • 0.35 GF/Watt Area: • 296 racks • 5500 ft2
  • 10. 10 IBM T.J. Watson Research Center CBE for High Performance Computing and Scalable Visualization | 11 July 2008 © 2008 IBM Corporation Cell/B.E. Real-time Application Solutions PC/Workstations, consoles, handhelds Scalable Network BackboneGbE/IB IB/GbE BC-H IBM TotalStorage IBMQS2x IBMQS2x IBMQS2x IBMHS2x IBMQS2x IBMQS2x IBMQS2x IBMQS2x IBMQS2x IBMHS2x IBMQS2x IBMQS2x IBMQS2x IBMQS2x IBMQS2x IBMHS2x IBMQS2x IBMQS2x Online Gaming and Rendering BladeCenter H chassis rack mounted, 4 per rack Heterogeneous solutions with Cell/B.E. and Intel blade servers GbE or Infiniband attached storage PC, Workstation, Game Consoles, Settop box, Handheld clients Wired or wirelessly connected
  • 11. 11 IBM T.J. Watson Research Center CBE for High Performance Computing and Scalable Visualization | 11 July 2008 © 2008 IBM Corporation Game Engine based on Cell/B.E. servers and Intel workstations Enables the next generation virtual environments – Current online video games engines perform limited amount of physical simulation – Not enough client CPU resources – No distributed physics – Bandwidth & Latency between processing nodes prohibitive to achieving real time performance Enable complex visual F/X movies on large servers - Machinima
  • 12. 12 IBM T.J. Watson Research Center CBE for High Performance Computing and Scalable Visualization | 11 July 2008 © 2008 IBM Corporation Intel/AMD PCs Game Consoles Network Scene Management Simulation User Input Client Server Server Model of World State Users View Client Model of World State Entity State Input from Server World State Input from Client Entity State Verification Input from Client Simulation Online Game Client/Server Execution Flow Rendering Rendering Composite Render Input from Server IBM QS2x Blades
  • 13. 13 IBM T.J. Watson Research Center CBE for High Performance Computing and Scalable Visualization | 11 July 2008 © 2008 IBM Corporation Scaling Virtual Worlds Game Framework Network Manager Player Login Event Handlers Client Intel/AMD PCs Consoles Handhelds Client Renderer Interpolator Local Database Event Handler IBM Cell/B.E. Servers Higher compute power required . . Rigid Body Zone0Zone1Zonen Collisions Rigid Body Collisions Network Compression Arithmetic Coding Network Compression Arithmetic Coding Divide Virtual world into geographic zones Intel/AMD/Power Servers Lower compute power required Virtual World State Manager Rigid Body Collisions Network Compression Arithmetic Coding Virtual World State Manager Virtual World State Manager Server-side State: Distributed Model
  • 14. 14 IBM T.J. Watson Research Center CBE for High Performance Computing and Scalable Visualization | 11 July 2008 © 2008 IBM Corporation Scaling Virtual Worlds IBM Cell/B.E. Servers Higher compute power required . . Virtual World State Manager Game Framework Network Manager Blade 0 Rigid Body Blade 1 Blade n Node0Node1Noden Client Renderer Interpolator Local Database Collisions Rigid Body Collisions Rigid Body Collisions Player Login Event Handlers Network Compression Arithmetic Coding Network Compression Arithmetic Coding Network Compression Arithmetic Coding Dispatch work (elements of Virtual World) to Compute Nodes Event Handler Intel/AMD/Power/Z Servers Lower compute power required More transaction processing Client Intel/AMD PCs Consoles Handhelds Server-side State: Centralized Model
  • 15. 15 IBM T.J. Watson Research Center CBE for High Performance Computing and Scalable Visualization | 11 July 2008 © 2008 IBM Corporation Server Side Physics Loop on Cell/B.E. Servers Numerical Integration to obtain positions of bodies for next step Sparse Matrix solvers Perform very well on Cell/B.E. Collision detection Bounding box culling Sorting Point/triangle intersection Even sorting has proven to be very fast on Cell SPUs !!
  • 16. 16 IBM T.J. Watson Research Center CBE for High Performance Computing and Scalable Visualization | 11 July 2008 © 2008 IBM Corporation UCSD Scalable City Project (Sheldon Brown, UCSD)
  • 17. 17 IBM T.J. Watson Research Center CBE for High Performance Computing and Scalable Visualization | 11 July 2008 © 2008 IBM Corporation Scalable City Pipeline: Landscape creation
  • 18. 18 IBM T.J. Watson Research Center CBE for High Performance Computing and Scalable Visualization | 11 July 2008 © 2008 IBM Corporation Scalable City Pipeline: Border Detection
  • 19. 19 IBM T.J. Watson Research Center CBE for High Performance Computing and Scalable Visualization | 11 July 2008 © 2008 IBM Corporation Scalable City Pipeline: Space Filling Curve is GeneratedScalable City Pipeline: Space Filling Curve is Generated
  • 20. 20 IBM T.J. Watson Research Center CBE for High Performance Computing and Scalable Visualization | 11 July 2008 © 2008 IBM Corporation Scalable City Pipeline: Simple Curves Converted into Road Geometry
  • 21. 21 IBM T.J. Watson Research Center CBE for High Performance Computing and Scalable Visualization | 11 July 2008 © 2008 IBM Corporation Before use of cell, this was pre- computed (before game play session) Roads Integrated into Landscape
  • 22. 22 IBM T.J. Watson Research Center CBE for High Performance Computing and Scalable Visualization | 11 July 2008 © 2008 IBM Corporation Cell based computation ofCell based computation of similar pipelinesimilar pipeline –– takes place intakes place in about 1 second using one bladeabout 1 second using one blade per landscape (per landscape (demodemo))
  • 23. 23 IBM T.J. Watson Research Center CBE for High Performance Computing and Scalable Visualization | 11 July 2008 © 2008 IBM Corporation IRT Rendering Engine: IBM demo technology
  • 24. 24 IBM T.J. Watson Research Center CBE for High Performance Computing and Scalable Visualization | 11 July 2008 © 2008 IBM Corporation SPE Partitions Blade 0 Blade 1 Blade 2 Blade 3 Blade 4 Blade 5 Blade 6 GDC 07 Configuration, Seven QS20 Blades, 3.2 TFLOPS iRT Blade Performance Scaling 0 2 4 6 8 10 12 14 16 18 20 1 2 3 4 5 6 7 Blades Frames/Sec 1080p 1 Realtime ray tracing including • Reflection/transparency • detailed shadows • ambient occlusion • 4x4 jittered multi-sampling Totaling up to 288 rays/pixel Lamborghini car model • 1.6 million polygons • 1080p hi-def. • demo Also Shown at GDC March 2007 • PS3 node w/ seven QS20 blades • 3 Million Triangle Urban Landscape • >100 Textures • 1080p HDTV GigabitNetwork Scalability at processor and system level
  • 25. 25 IBM T.J. Watson Research Center CBE for High Performance Computing and Scalable Visualization | 11 July 2008 © 2008 IBM Corporation Cell/B.E. vs. G80 – Ray tracing Performance IBM QS20 Blade (Two 3.2 GHz Cell Processors) - IBM Demo ray tracer Despite comparable theoretical floating point performance: – One Cell/BE is 4-5x faster than G80. – One QS20 blade (two Cell/BE processors) is 8-11x faster than G80 AMD Opteron, the defacto industry workhorse for rendering is easily overwhelmed by Cell/B.E. or G80. Secondary Ray processing – Incoherent nature makes them difficult to process on G80, but they are easily handled by Cell/B.E. Cell/B.E. is half the size of the G80 and produces five times the ray-tracing performance. Left to Right: •2.6 GHz AMD Opteron - Saarland Ray- tracer •nVIDIA GeForce 8800 GTX - Saarland Ray-tracer •Sony Playstation3 (partial 3.2 GHz Cell/B.E. processor running Linux) - IBM Demo •3.2 GHz Cell/B.E. Processor - IBM Demo
  • 26. 26 IBM T.J. Watson Research Center CBE for High Performance Computing and Scalable Visualization | 11 July 2008 © 2008 IBM Corporation Cell/B.E. enables scalable, shared architecture with full consumer to professional potential Sony PLAYSTATION©3 (Cell/B.E. + GPU) IBM Cell/B.E. Blade (2 Cell/B.E.) IBM Roadrunner (16,000 Cell/B.E. + AMD) Mercury 1u Dual Cell Sony Cell/B.E. Computing Unit (Cell/B.E. + GPU + AV I/O) Consumer Professional High Perf ComputingBusiness PowerXcell 8i PCI card (Cell/B.E. + Host) Common OS’s, Infrastructure, Tools, Libraries, Code… Toshiba Spurs Engine (SPU’s. + Host)
  • 27. 27 IBM T.J. Watson Research Center CBE for High Performance Computing and Scalable Visualization | 11 July 2008 © 2008 IBM Corporation © Copyright International Business Machines Corporation 2007 All Rights Reserved This document was developed for IBM offerings in the United States as of the date of publication. IBM may not make these offerings available in other countries, and the information is subject to change without notice. Consult your local IBM business contact for information on the IBM offerings available in your area. In no event will IBM be liable for damages arising directly or indirectly from any use of the information contained in this document. Information in this document concerning non-IBM products was obtained from the suppliers of these products or other public sources. Questions on the capabilities of non-IBM products should be addressed to the suppliers of those products. IBM may have patents or pending patent applications covering subject matter in this document. The furnishing of this document does not give you any license to these patents. Send license inquires, in writing, to IBM Director of Licensing, IBM Corporation, New Castle Drive, Armonk, NY 10504- 1785 USA. All statements regarding IBM future direction and intent are subject to change or withdrawal without notice, and represent goals and objectives only. The information contained in this document has not been submitted to any formal IBM test and is provided "AS IS" with no warranties or guarantees either expressed or implied. All examples cited or described in this document are presented as illustrations of the manner in which some IBM products can be used and the results that may be achieved. Actual environmental costs and performance characteristics will vary depending on individual client configurations and conditions. IBM Global Financing offerings are provided through IBM Credit Corporation in the United States and other IBM subsidiaries and divisions worldwide to qualified commercial and government clients. Rates are based on a client's credit rating, financing terms, offering type, equipment type and options, and may vary by country. Other restrictions may apply. Rates and offerings are subject to change, extension or withdrawal without notice. IBM is not responsible for printing errors in this document that result in pricing or information inaccuracies. All prices shown are IBM's United States suggested list prices and are subject to change without notice; reseller prices may vary. IBM hardware products are manufactured from new parts, or new and serviceable used parts. Regardless, our warranty terms apply. Many of the features described in this document are operating system dependent and may not be available on Linux. For more information, please check: http://www.ibm.com/systems/p/software/whitepapers/linux_overview.html Any performance data contained in this document was determined in a controlled environment. Actual results may vary significantly and are dependent on many factors including system hardware configuration and software design and configuration. Some measurements quoted in this document may have been made on development-level systems. There is no guarantee these measurements will be the same on generally-available systems. Some measurements quoted in this document may have been estimated through extrapolation. Users of this document should verify the applicable data for their specific environment. Special Notices