SlideShare a Scribd company logo
1 of 26
Supercomputer Performance Characterization Presented By: IQxplorer
Here are some important computer performance questions ,[object Object],[object Object],[object Object],[object Object],[object Object]
Comparative performance results have been obtained on six computers at NCSA & SDSC, all with > 1,000 processors
These computers have shared-memory nodes of widely varying size connected by different switch types ,[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object]
Performance can be better understood with a simple model ,[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object]
Run-time components depend upon system parameters & code features Differences between point-to-point & collective communication are important too
Compute, communication, & I/O speeds have been measured for many synthetic & application benchmarks ,[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object]
Normalized memory access profiles for daxpy show better memory access, but more memory contention on Blue Gene compared DataStar
Each HPCC synthetic benchmark measures one or two system parameters in varying combinations
Relative speeds are shown for HPCC benchmarks on 6 computers at 1,024p; 4 different computers are fastest depending upon benchmark; 2 of these are also slowest, depending upon benchmark   Data available soon at CIP Web site: www.ci-partnership.org
Absolute speeds are shown for HPCC & IOR benchmarks on SDSC computers; TG processors are fastest, BG & DS interconnects are fastest, & all three computers have similar I/O rates
Relative speeds are shown for 5 applications on 6 computers at various processor counts; Cobalt & DataStar are generally fastest
Good scaling is essential to take advantage of high processors counts ,[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object]
AWM 512^3 problem shows good strong scaling to 2,048p on Blue Gene & to 512p on DataStar, but not on TeraGrid cluster Data from Yifeng Cui
MILC medium problem shows superlinear speedup on Cobalt, Mercury, & DataStar at small processor counts; strong scaling ends for DataStar & Blue Gene above 2,048p
NAMD ApoA1 problem scales best on DataStar & Blue Gene; Cobalt is fastest below 512p, but the same speed as DataStar at 512p
WRF standard problem scales best on DataStar; Cobalt is fastest below 512p, but the same speed as DataStar at 512p
Communication fraction generally grows with processor count in strong scaling scans, such as for WRF standard problem on DataStar
A more careful look at Blue Gene shows many pluses ,[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object]
But there are also some minuses ,[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object]
Major applications ported and being run on BG at SDSC span various disciplines
Speed of BG relative to DataStar varies about clock speed ratio (0.47 = 0.7/1.5) for applications on ≥ 512p; CO & VN mode perform similarly (per MPI p)
DNS scaling on BG is generally better than on DataStar, but shows unusual variation;   VN mode is somewhat slower than CO mode (per MPI p) Data from  Dmitry Pekurovsky
If number of allocated processors is considered, then VN mode is faster than CO mode, and both modes show unusual variation  Data from  Dmitry Pekurovsky
IOR weak scaling scans using GPFS-WAN show BG in VN mode achieves 3.4 GB/s for writes (~DS) & 2.7 GB/s for reads (>DS)
Blue Gene has more limited applicability than DataStar, but is a good choice if the application is right   ,[object Object],[object Object],[object Object],[object Object],[object Object],[object Object]

More Related Content

What's hot

Super computers by rachna
Super computers by  rachnaSuper computers by  rachna
Super computers by rachna
Rachna Singh
 

What's hot (20)

Supercomputer - Overview
Supercomputer - OverviewSupercomputer - Overview
Supercomputer - Overview
 
Supercomputer @ manarat university by reza
Supercomputer  @ manarat university by rezaSupercomputer  @ manarat university by reza
Supercomputer @ manarat university by reza
 
Top 10 Supercomputer 2014
Top 10 Supercomputer 2014Top 10 Supercomputer 2014
Top 10 Supercomputer 2014
 
Super computers
Super computersSuper computers
Super computers
 
Super Computers
Super ComputersSuper Computers
Super Computers
 
Param yuva ii
Param yuva iiParam yuva ii
Param yuva ii
 
Evolution of modern super computers
Evolution of modern  super computersEvolution of modern  super computers
Evolution of modern super computers
 
Super computer
Super computerSuper computer
Super computer
 
Super computers by rachna
Super computers by  rachnaSuper computers by  rachna
Super computers by rachna
 
Super computers
Super computersSuper computers
Super computers
 
Super computer 2017
Super computer 2017Super computer 2017
Super computer 2017
 
Brief Definition of Supercomputers
Brief Definition of SupercomputersBrief Definition of Supercomputers
Brief Definition of Supercomputers
 
Supercomputers
SupercomputersSupercomputers
Supercomputers
 
Supercomputer ppt
Supercomputer pptSupercomputer ppt
Supercomputer ppt
 
Supercomputer
SupercomputerSupercomputer
Supercomputer
 
Super computer
Super computerSuper computer
Super computer
 
Super computer ppt
Super computer pptSuper computer ppt
Super computer ppt
 
SUPERCOMPUTER
SUPERCOMPUTERSUPERCOMPUTER
SUPERCOMPUTER
 
Super Computer
Super ComputerSuper Computer
Super Computer
 
4'th Fastest Super Computer
4'th Fastest Super Computer4'th Fastest Super Computer
4'th Fastest Super Computer
 

Viewers also liked

Laser Communication
Laser CommunicationLaser Communication
Laser Communication
Hossam Zein
 
Executive Information System
Executive Information SystemExecutive Information System
Executive Information System
Theju Paul
 
Virtual keyboard
Virtual keyboardVirtual keyboard
Virtual keyboard
Nikhil Vyas
 
Global positioning system ppt
Global positioning system pptGlobal positioning system ppt
Global positioning system ppt
Swapnil Ramgirwar
 

Viewers also liked (19)

Supercomputers
SupercomputersSupercomputers
Supercomputers
 
Презентация ИТ Кластер Сколково
Презентация  ИТ Кластер СколковоПрезентация  ИТ Кластер Сколково
Презентация ИТ Кластер Сколково
 
Laser Communications
Laser CommunicationsLaser Communications
Laser Communications
 
Laser Communication
Laser CommunicationLaser Communication
Laser Communication
 
Executive information system
Executive information systemExecutive information system
Executive information system
 
Virtual keyboard
Virtual keyboardVirtual keyboard
Virtual keyboard
 
Global Positioning System (GPS)
Global Positioning System (GPS)Global Positioning System (GPS)
Global Positioning System (GPS)
 
LASER Communication
LASER CommunicationLASER Communication
LASER Communication
 
Laser communication
Laser communicationLaser communication
Laser communication
 
Executive Information System
Executive Information SystemExecutive Information System
Executive Information System
 
Virtual keyboard
Virtual keyboardVirtual keyboard
Virtual keyboard
 
Laser Communication
Laser CommunicationLaser Communication
Laser Communication
 
Laser Communications
Laser CommunicationsLaser Communications
Laser Communications
 
Global positioning system ppt
Global positioning system pptGlobal positioning system ppt
Global positioning system ppt
 
Global Positioning System
Global Positioning SystemGlobal Positioning System
Global Positioning System
 
Global Positioning System (GPS)
Global Positioning System (GPS) Global Positioning System (GPS)
Global Positioning System (GPS)
 
Laser
LaserLaser
Laser
 
"GPS" Global Positioning System [PDF]
"GPS" Global Positioning System  [PDF]"GPS" Global Positioning System  [PDF]
"GPS" Global Positioning System [PDF]
 
Computer hardware presentation
Computer hardware presentationComputer hardware presentation
Computer hardware presentation
 

Similar to Super Computer

Conference Paper: Universal Node: Towards a high-performance NFV environment
Conference Paper: Universal Node: Towards a high-performance NFV environmentConference Paper: Universal Node: Towards a high-performance NFV environment
Conference Paper: Universal Node: Towards a high-performance NFV environment
Ericsson
 
Analysis of Multicore Performance Degradation of Scientific Applications
Analysis of Multicore Performance Degradation of Scientific ApplicationsAnalysis of Multicore Performance Degradation of Scientific Applications
Analysis of Multicore Performance Degradation of Scientific Applications
James McGalliard
 
Linac Coherent Light Source (LCLS) Data Transfer Requirements
Linac Coherent Light Source (LCLS) Data Transfer RequirementsLinac Coherent Light Source (LCLS) Data Transfer Requirements
Linac Coherent Light Source (LCLS) Data Transfer Requirements
inside-BigData.com
 
Exploring hybrid memory for gpu energy efficiency through software hardware c...
Exploring hybrid memory for gpu energy efficiency through software hardware c...Exploring hybrid memory for gpu energy efficiency through software hardware c...
Exploring hybrid memory for gpu energy efficiency through software hardware c...
Cheng-Hsuan Li
 

Similar to Super Computer (20)

Dataplane networking acceleration with OpenDataplane / Максим Уваров (Linaro)
Dataplane networking acceleration with OpenDataplane / Максим Уваров (Linaro)Dataplane networking acceleration with OpenDataplane / Максим Уваров (Linaro)
Dataplane networking acceleration with OpenDataplane / Максим Уваров (Linaro)
 
Multiscale Dataflow Computing: Competitive Advantage at the Exascale Frontier
Multiscale Dataflow Computing: Competitive Advantage at the Exascale FrontierMultiscale Dataflow Computing: Competitive Advantage at the Exascale Frontier
Multiscale Dataflow Computing: Competitive Advantage at the Exascale Frontier
 
Conference Paper: Universal Node: Towards a high-performance NFV environment
Conference Paper: Universal Node: Towards a high-performance NFV environmentConference Paper: Universal Node: Towards a high-performance NFV environment
Conference Paper: Universal Node: Towards a high-performance NFV environment
 
Configuration Optimization for Big Data Software
Configuration Optimization for Big Data SoftwareConfiguration Optimization for Big Data Software
Configuration Optimization for Big Data Software
 
Performance and Energy evaluation
Performance and Energy evaluationPerformance and Energy evaluation
Performance and Energy evaluation
 
Scalable analytics for iaas cloud availability
Scalable analytics for iaas cloud availabilityScalable analytics for iaas cloud availability
Scalable analytics for iaas cloud availability
 
Hadoop World 2011: Hadoop Network and Compute Architecture Considerations - J...
Hadoop World 2011: Hadoop Network and Compute Architecture Considerations - J...Hadoop World 2011: Hadoop Network and Compute Architecture Considerations - J...
Hadoop World 2011: Hadoop Network and Compute Architecture Considerations - J...
 
DesignCon 2015-criticalmemoryperformancemetricsforDDR4
DesignCon 2015-criticalmemoryperformancemetricsforDDR4DesignCon 2015-criticalmemoryperformancemetricsforDDR4
DesignCon 2015-criticalmemoryperformancemetricsforDDR4
 
Analysis of Multicore Performance Degradation of Scientific Applications
Analysis of Multicore Performance Degradation of Scientific ApplicationsAnalysis of Multicore Performance Degradation of Scientific Applications
Analysis of Multicore Performance Degradation of Scientific Applications
 
Prelim Slides
Prelim SlidesPrelim Slides
Prelim Slides
 
Paper
PaperPaper
Paper
 
Linac Coherent Light Source (LCLS) Data Transfer Requirements
Linac Coherent Light Source (LCLS) Data Transfer RequirementsLinac Coherent Light Source (LCLS) Data Transfer Requirements
Linac Coherent Light Source (LCLS) Data Transfer Requirements
 
Application Report: Big Data - Big Cluster Interconnects
Application Report: Big Data - Big Cluster InterconnectsApplication Report: Big Data - Big Cluster Interconnects
Application Report: Big Data - Big Cluster Interconnects
 
Exascale Capabl
Exascale CapablExascale Capabl
Exascale Capabl
 
Exploring hybrid memory for gpu energy efficiency through software hardware c...
Exploring hybrid memory for gpu energy efficiency through software hardware c...Exploring hybrid memory for gpu energy efficiency through software hardware c...
Exploring hybrid memory for gpu energy efficiency through software hardware c...
 
676.v3
676.v3676.v3
676.v3
 
1.multicore processors
1.multicore processors1.multicore processors
1.multicore processors
 
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
 
Accelerated development in Automotive E/E Systems using VisualSim Architect
Accelerated development in Automotive E/E Systems using VisualSim ArchitectAccelerated development in Automotive E/E Systems using VisualSim Architect
Accelerated development in Automotive E/E Systems using VisualSim Architect
 
The state of SQL-on-Hadoop in the Cloud
The state of SQL-on-Hadoop in the CloudThe state of SQL-on-Hadoop in the Cloud
The state of SQL-on-Hadoop in the Cloud
 

Recently uploaded

Tales from a Passkey Provider Progress from Awareness to Implementation.pptx
Tales from a Passkey Provider  Progress from Awareness to Implementation.pptxTales from a Passkey Provider  Progress from Awareness to Implementation.pptx
Tales from a Passkey Provider Progress from Awareness to Implementation.pptx
FIDO Alliance
 
Hyatt driving innovation and exceptional customer experiences with FIDO passw...
Hyatt driving innovation and exceptional customer experiences with FIDO passw...Hyatt driving innovation and exceptional customer experiences with FIDO passw...
Hyatt driving innovation and exceptional customer experiences with FIDO passw...
FIDO Alliance
 
Structuring Teams and Portfolios for Success
Structuring Teams and Portfolios for SuccessStructuring Teams and Portfolios for Success
Structuring Teams and Portfolios for Success
UXDXConf
 
Breaking Down the Flutterwave Scandal What You Need to Know.pdf
Breaking Down the Flutterwave Scandal What You Need to Know.pdfBreaking Down the Flutterwave Scandal What You Need to Know.pdf
Breaking Down the Flutterwave Scandal What You Need to Know.pdf
UK Journal
 
Easier, Faster, and More Powerful – Alles Neu macht der Mai -Wir durchleuchte...
Easier, Faster, and More Powerful – Alles Neu macht der Mai -Wir durchleuchte...Easier, Faster, and More Powerful – Alles Neu macht der Mai -Wir durchleuchte...
Easier, Faster, and More Powerful – Alles Neu macht der Mai -Wir durchleuchte...
panagenda
 

Recently uploaded (20)

1111 ChatGPT Prompts PDF Free Download - Prompts for ChatGPT
1111 ChatGPT Prompts PDF Free Download - Prompts for ChatGPT1111 ChatGPT Prompts PDF Free Download - Prompts for ChatGPT
1111 ChatGPT Prompts PDF Free Download - Prompts for ChatGPT
 
Tales from a Passkey Provider Progress from Awareness to Implementation.pptx
Tales from a Passkey Provider  Progress from Awareness to Implementation.pptxTales from a Passkey Provider  Progress from Awareness to Implementation.pptx
Tales from a Passkey Provider Progress from Awareness to Implementation.pptx
 
Design Guidelines for Passkeys 2024.pptx
Design Guidelines for Passkeys 2024.pptxDesign Guidelines for Passkeys 2024.pptx
Design Guidelines for Passkeys 2024.pptx
 
Intro to Passkeys and the State of Passwordless.pptx
Intro to Passkeys and the State of Passwordless.pptxIntro to Passkeys and the State of Passwordless.pptx
Intro to Passkeys and the State of Passwordless.pptx
 
Simplified FDO Manufacturing Flow with TPMs _ Liam at Infineon.pdf
Simplified FDO Manufacturing Flow with TPMs _ Liam at Infineon.pdfSimplified FDO Manufacturing Flow with TPMs _ Liam at Infineon.pdf
Simplified FDO Manufacturing Flow with TPMs _ Liam at Infineon.pdf
 
Secure Zero Touch enabled Edge compute with Dell NativeEdge via FDO _ Brad at...
Secure Zero Touch enabled Edge compute with Dell NativeEdge via FDO _ Brad at...Secure Zero Touch enabled Edge compute with Dell NativeEdge via FDO _ Brad at...
Secure Zero Touch enabled Edge compute with Dell NativeEdge via FDO _ Brad at...
 
Choosing the Right FDO Deployment Model for Your Application _ Geoffrey at In...
Choosing the Right FDO Deployment Model for Your Application _ Geoffrey at In...Choosing the Right FDO Deployment Model for Your Application _ Geoffrey at In...
Choosing the Right FDO Deployment Model for Your Application _ Geoffrey at In...
 
Hyatt driving innovation and exceptional customer experiences with FIDO passw...
Hyatt driving innovation and exceptional customer experiences with FIDO passw...Hyatt driving innovation and exceptional customer experiences with FIDO passw...
Hyatt driving innovation and exceptional customer experiences with FIDO passw...
 
Introduction to FIDO Authentication and Passkeys.pptx
Introduction to FIDO Authentication and Passkeys.pptxIntroduction to FIDO Authentication and Passkeys.pptx
Introduction to FIDO Authentication and Passkeys.pptx
 
2024 May Patch Tuesday
2024 May Patch Tuesday2024 May Patch Tuesday
2024 May Patch Tuesday
 
Structuring Teams and Portfolios for Success
Structuring Teams and Portfolios for SuccessStructuring Teams and Portfolios for Success
Structuring Teams and Portfolios for Success
 
Oauth 2.0 Introduction and Flows with MuleSoft
Oauth 2.0 Introduction and Flows with MuleSoftOauth 2.0 Introduction and Flows with MuleSoft
Oauth 2.0 Introduction and Flows with MuleSoft
 
Working together SRE & Platform Engineering
Working together SRE & Platform EngineeringWorking together SRE & Platform Engineering
Working together SRE & Platform Engineering
 
Introduction to FDO and How It works Applications _ Richard at FIDO Alliance.pdf
Introduction to FDO and How It works Applications _ Richard at FIDO Alliance.pdfIntroduction to FDO and How It works Applications _ Richard at FIDO Alliance.pdf
Introduction to FDO and How It works Applications _ Richard at FIDO Alliance.pdf
 
Human Expert Website Manual WCAG 2.0 2.1 2.2 Audit - Digital Accessibility Au...
Human Expert Website Manual WCAG 2.0 2.1 2.2 Audit - Digital Accessibility Au...Human Expert Website Manual WCAG 2.0 2.1 2.2 Audit - Digital Accessibility Au...
Human Expert Website Manual WCAG 2.0 2.1 2.2 Audit - Digital Accessibility Au...
 
Event-Driven Architecture Masterclass: Engineering a Robust, High-performance...
Event-Driven Architecture Masterclass: Engineering a Robust, High-performance...Event-Driven Architecture Masterclass: Engineering a Robust, High-performance...
Event-Driven Architecture Masterclass: Engineering a Robust, High-performance...
 
How Red Hat Uses FDO in Device Lifecycle _ Costin and Vitaliy at Red Hat.pdf
How Red Hat Uses FDO in Device Lifecycle _ Costin and Vitaliy at Red Hat.pdfHow Red Hat Uses FDO in Device Lifecycle _ Costin and Vitaliy at Red Hat.pdf
How Red Hat Uses FDO in Device Lifecycle _ Costin and Vitaliy at Red Hat.pdf
 
Breaking Down the Flutterwave Scandal What You Need to Know.pdf
Breaking Down the Flutterwave Scandal What You Need to Know.pdfBreaking Down the Flutterwave Scandal What You Need to Know.pdf
Breaking Down the Flutterwave Scandal What You Need to Know.pdf
 
Easier, Faster, and More Powerful – Alles Neu macht der Mai -Wir durchleuchte...
Easier, Faster, and More Powerful – Alles Neu macht der Mai -Wir durchleuchte...Easier, Faster, and More Powerful – Alles Neu macht der Mai -Wir durchleuchte...
Easier, Faster, and More Powerful – Alles Neu macht der Mai -Wir durchleuchte...
 
ADP Passwordless Journey Case Study.pptx
ADP Passwordless Journey Case Study.pptxADP Passwordless Journey Case Study.pptx
ADP Passwordless Journey Case Study.pptx
 

Super Computer

  • 2.
  • 3. Comparative performance results have been obtained on six computers at NCSA & SDSC, all with > 1,000 processors
  • 4.
  • 5.
  • 6. Run-time components depend upon system parameters & code features Differences between point-to-point & collective communication are important too
  • 7.
  • 8. Normalized memory access profiles for daxpy show better memory access, but more memory contention on Blue Gene compared DataStar
  • 9. Each HPCC synthetic benchmark measures one or two system parameters in varying combinations
  • 10. Relative speeds are shown for HPCC benchmarks on 6 computers at 1,024p; 4 different computers are fastest depending upon benchmark; 2 of these are also slowest, depending upon benchmark Data available soon at CIP Web site: www.ci-partnership.org
  • 11. Absolute speeds are shown for HPCC & IOR benchmarks on SDSC computers; TG processors are fastest, BG & DS interconnects are fastest, & all three computers have similar I/O rates
  • 12. Relative speeds are shown for 5 applications on 6 computers at various processor counts; Cobalt & DataStar are generally fastest
  • 13.
  • 14. AWM 512^3 problem shows good strong scaling to 2,048p on Blue Gene & to 512p on DataStar, but not on TeraGrid cluster Data from Yifeng Cui
  • 15. MILC medium problem shows superlinear speedup on Cobalt, Mercury, & DataStar at small processor counts; strong scaling ends for DataStar & Blue Gene above 2,048p
  • 16. NAMD ApoA1 problem scales best on DataStar & Blue Gene; Cobalt is fastest below 512p, but the same speed as DataStar at 512p
  • 17. WRF standard problem scales best on DataStar; Cobalt is fastest below 512p, but the same speed as DataStar at 512p
  • 18. Communication fraction generally grows with processor count in strong scaling scans, such as for WRF standard problem on DataStar
  • 19.
  • 20.
  • 21. Major applications ported and being run on BG at SDSC span various disciplines
  • 22. Speed of BG relative to DataStar varies about clock speed ratio (0.47 = 0.7/1.5) for applications on ≥ 512p; CO & VN mode perform similarly (per MPI p)
  • 23. DNS scaling on BG is generally better than on DataStar, but shows unusual variation; VN mode is somewhat slower than CO mode (per MPI p) Data from Dmitry Pekurovsky
  • 24. If number of allocated processors is considered, then VN mode is faster than CO mode, and both modes show unusual variation Data from Dmitry Pekurovsky
  • 25. IOR weak scaling scans using GPFS-WAN show BG in VN mode achieves 3.4 GB/s for writes (~DS) & 2.7 GB/s for reads (>DS)
  • 26.