SlideShare a Scribd company logo
1 of 9
SR-IOV: The Key Enabling Technology for
Fully Virtualized HPC Clusters!
Glenn K. Lockwood!
Christopher Irving!
Philip M. Papadopoulos!
Mahidhar Tatineni!
Rick Wagner!
SAN DIEGO SUPERCOMPUTER CENTER
Single Root I/O Virtualization in HPC!
•  Problem: complex workflows demand increasing
flexibility from HPC platforms"
•  Virtualization = flexibility"
•  Virtualization = IO performance loss (e.g.,
excessive DMA interrupts)"
•  Solution: SR-IOV and Mellanox ConnectX-3
InfiniBand HCAs "
•  One physical function (PF) à multiple virtual
functions (VF), each with own DMA streams,
memory space, interrupts"
•  Allows DMA to bypass hypervisor to VMs!
SAN DIEGO SUPERCOMPUTER CENTER
High-Performance Virtualization
on Comet !
•  Mellanox FDR InfiniBand HCAs with SR-IOV"
•  Rocks and OpenStack Nova to manage VMs"
•  Flexibility to support complex science
gateways and web-based workflow engines"
•  custom compute appliances and virtual clusters
developed with FutureGrid and their existing
expertise"
•  backed by virtualized Lustre running over virtualized
InfiniBand"
SAN DIEGO SUPERCOMPUTER CENTER
Hardware/Software Configuration of Test Cluster !
Native, SR-IOV!

Amazon EC2!

Platform"

•  Rocks 6.1 (EL6)"
•  Virtualization via kvm	
  

•  Amazon Linux 2013.03 (EL6)"
•  cc2.8xlarge Instances"

CPUs"

•  2x Xeon E5-2660 (2.2GHz)"
•  16 cores per node"

•  2x Xeon E5-2670 (2.6GHz)"
•  16 cores per node"

RAM"

•  64 GB DDR3 DRAM"

•  60.5 DDR3 DRAM"

Interconnect" •  QDR4X InfiniBand"
•  10 GbE"
•  Mellanox ConnectX-3 (MT27500)" •  common placement group"
•  Intel VT-d, SR-IOV enabled in
firmware, kernel, drivers"
•  mlx4_core	
  1.1"
•  Mellanox OFED 2.0"
•  HCA firmware 2.11.1192"
SAN DIEGO SUPERCOMPUTER CENTER
50x less latency than Amazon EC2!
•  SR-IOV!
•  < 30% overhead for M <
128 bytes"
•  < 10% overhead for
eager send/recv"
•  Overhead à 0% for
bandwidth-limited regime"
•  Amazon EC2!
•  > 5000% worse latency"
•  Time dependent (noisy)"
5
	

SAN DIEGO SUPERCOMPUTER CENTER

OSU Microbenchmarks (3.9, osu_latency)"
10x more bandwidth than Amazon EC2!
•  SR-IOV!
•  < 2% bandwidth loss
over entire range"
•  > 95% peak bandwidth"
•  Amazon EC2!
•  < 35% peak bandwidth"
•  900% to 2500% worse
bandwidth than
virtualized InfiniBand"
6
	

SAN DIEGO SUPERCOMPUTER CENTER

OSU Microbenchmarks (3.9, osu_bw)"
Weather Modeling – 15% Overhead!
•  96-core (6-node)
calculation"
•  Nearest-neighbor
communication"
•  Scalable algorithms"
•  SR-IOV incurs modest
(15%) performance hit"
•  ...but still still 20%
faster*** than Amazon"
SAN DIEGO SUPERCOMPUTER CENTER

WRF 3.4.1 – 3hr forecast"
*** 20% faster despite SR-IOV cluster having 20% slower CPUs"
Quantum ESPRESSO: 5x Faster than EC2!
•  48-core, 3 node calc"
•  CG matrix inversion
(irregular comm.)"
•  3D FFT matrix
transposes (All-to-all
communication)"
•  28% slower w/ SR-IOV"
•  SR-IOV still > 500%
faster*** than EC2"
SAN DIEGO SUPERCOMPUTER CENTER

Quantum Espresso 5.0.2 – DEISA AUSURF112 benchmark"
*** 20% faster despite SR-IOV cluster having 20% slower CPUs"
Conclusions!
•  SR-IOV: huge step forward in high-performance virtualization"
•  Shows substantial improvement in latency over Amazon EC2, and
it provides nearly 0 bandwidth overhead!
•  Benchmark application performance confirms this: significant
improvement over EC2"

•  SR-IOV: lowers performance barrier to virtualizing the
interconnect and makes fully virtualized HPC clusters
viable!
•  Comet will deliver virtualized HPC to new/non-traditional
communities that need flexibility without major loss of performance!
SAN DIEGO SUPERCOMPUTER CENTER

More Related Content

What's hot

05.2 virtio introduction
05.2 virtio introduction05.2 virtio introduction
05.2 virtio introductionzenixls2
 
Continguous Memory Allocator in the Linux Kernel
Continguous Memory Allocator in the Linux KernelContinguous Memory Allocator in the Linux Kernel
Continguous Memory Allocator in the Linux KernelKernel TLV
 
Linux MMAP & Ioremap introduction
Linux MMAP & Ioremap introductionLinux MMAP & Ioremap introduction
Linux MMAP & Ioremap introductionGene Chang
 
The Linux Kernel Scheduler (For Beginners) - SFO17-421
The Linux Kernel Scheduler (For Beginners) - SFO17-421The Linux Kernel Scheduler (For Beginners) - SFO17-421
The Linux Kernel Scheduler (For Beginners) - SFO17-421Linaro
 
Moving to PCI Express based SSD with NVM Express
Moving to PCI Express based SSD with NVM ExpressMoving to PCI Express based SSD with NVM Express
Moving to PCI Express based SSD with NVM ExpressOdinot Stanislas
 
Message Signaled Interrupts
Message Signaled InterruptsMessage Signaled Interrupts
Message Signaled InterruptsAnshuman Biswal
 
Jagan Teki - U-boot from scratch
Jagan Teki - U-boot from scratchJagan Teki - U-boot from scratch
Jagan Teki - U-boot from scratchlinuxlab_conf
 
DMA Survival Guide
DMA Survival GuideDMA Survival Guide
DMA Survival GuideKernel TLV
 
Troubleshooting Linux Kernel Modules And Device Drivers
Troubleshooting Linux Kernel Modules And Device DriversTroubleshooting Linux Kernel Modules And Device Drivers
Troubleshooting Linux Kernel Modules And Device DriversSatpal Parmar
 
AMD Chiplet Architecture for High-Performance Server and Desktop Products
AMD Chiplet Architecture for High-Performance Server and Desktop ProductsAMD Chiplet Architecture for High-Performance Server and Desktop Products
AMD Chiplet Architecture for High-Performance Server and Desktop ProductsAMD
 
The Theory and Implementation of DVFS on Linux
The Theory and Implementation of DVFS on LinuxThe Theory and Implementation of DVFS on Linux
The Theory and Implementation of DVFS on LinuxPicker Weng
 

What's hot (20)

I2c drivers
I2c driversI2c drivers
I2c drivers
 
Linux PV on HVM
Linux PV on HVMLinux PV on HVM
Linux PV on HVM
 
USB Drivers
USB DriversUSB Drivers
USB Drivers
 
05.2 virtio introduction
05.2 virtio introduction05.2 virtio introduction
05.2 virtio introduction
 
Continguous Memory Allocator in the Linux Kernel
Continguous Memory Allocator in the Linux KernelContinguous Memory Allocator in the Linux Kernel
Continguous Memory Allocator in the Linux Kernel
 
Qemu Introduction
Qemu IntroductionQemu Introduction
Qemu Introduction
 
Linux Porting
Linux PortingLinux Porting
Linux Porting
 
Platform Drivers
Platform DriversPlatform Drivers
Platform Drivers
 
BusyBox for Embedded Linux
BusyBox for Embedded LinuxBusyBox for Embedded Linux
BusyBox for Embedded Linux
 
Linux MMAP & Ioremap introduction
Linux MMAP & Ioremap introductionLinux MMAP & Ioremap introduction
Linux MMAP & Ioremap introduction
 
The Linux Kernel Scheduler (For Beginners) - SFO17-421
The Linux Kernel Scheduler (For Beginners) - SFO17-421The Linux Kernel Scheduler (For Beginners) - SFO17-421
The Linux Kernel Scheduler (For Beginners) - SFO17-421
 
BeagleBone Black Bootloaders
BeagleBone Black BootloadersBeagleBone Black Bootloaders
BeagleBone Black Bootloaders
 
Moving to PCI Express based SSD with NVM Express
Moving to PCI Express based SSD with NVM ExpressMoving to PCI Express based SSD with NVM Express
Moving to PCI Express based SSD with NVM Express
 
Message Signaled Interrupts
Message Signaled InterruptsMessage Signaled Interrupts
Message Signaled Interrupts
 
Jagan Teki - U-boot from scratch
Jagan Teki - U-boot from scratchJagan Teki - U-boot from scratch
Jagan Teki - U-boot from scratch
 
DMA Survival Guide
DMA Survival GuideDMA Survival Guide
DMA Survival Guide
 
Troubleshooting Linux Kernel Modules And Device Drivers
Troubleshooting Linux Kernel Modules And Device DriversTroubleshooting Linux Kernel Modules And Device Drivers
Troubleshooting Linux Kernel Modules And Device Drivers
 
AMD Chiplet Architecture for High-Performance Server and Desktop Products
AMD Chiplet Architecture for High-Performance Server and Desktop ProductsAMD Chiplet Architecture for High-Performance Server and Desktop Products
AMD Chiplet Architecture for High-Performance Server and Desktop Products
 
Introduction to Linux
Introduction to LinuxIntroduction to Linux
Introduction to Linux
 
The Theory and Implementation of DVFS on Linux
The Theory and Implementation of DVFS on LinuxThe Theory and Implementation of DVFS on Linux
The Theory and Implementation of DVFS on Linux
 

Viewers also liked

Optimizing ARM cortex a and cortex-m based heterogeneous multiprocessor syste...
Optimizing ARM cortex a and cortex-m based heterogeneous multiprocessor syste...Optimizing ARM cortex a and cortex-m based heterogeneous multiprocessor syste...
Optimizing ARM cortex a and cortex-m based heterogeneous multiprocessor syste...Arm
 
A practical approach to securing embedded and io t platforms
A practical approach to securing embedded and io t platformsA practical approach to securing embedded and io t platforms
A practical approach to securing embedded and io t platformsArm
 
Developing functional safety systems with arm architecture solutions stroud
Developing functional safety systems with arm architecture solutions   stroudDeveloping functional safety systems with arm architecture solutions   stroud
Developing functional safety systems with arm architecture solutions stroudArm
 
Efficient software development with heterogeneous devices
Efficient software development with heterogeneous devicesEfficient software development with heterogeneous devices
Efficient software development with heterogeneous devicesArm
 
The Proto-Burst Buffer: Experience with the flash-based file system on SDSC's...
The Proto-Burst Buffer: Experience with the flash-based file system on SDSC's...The Proto-Burst Buffer: Experience with the flash-based file system on SDSC's...
The Proto-Burst Buffer: Experience with the flash-based file system on SDSC's...Glenn K. Lockwood
 
Software development in ar mv8 m architecture - yiu
Software development in ar mv8 m architecture - yiuSoftware development in ar mv8 m architecture - yiu
Software development in ar mv8 m architecture - yiuArm
 
So you think developing an SoC needs to be complex or expensive?
So you think developing an SoC needs to be complex or expensive?So you think developing an SoC needs to be complex or expensive?
So you think developing an SoC needs to be complex or expensive?Arm
 
Microsoft Project Olympus AI Accelerator Chassis (HGX-1)
Microsoft Project Olympus AI Accelerator Chassis (HGX-1)Microsoft Project Olympus AI Accelerator Chassis (HGX-1)
Microsoft Project Olympus AI Accelerator Chassis (HGX-1)inside-BigData.com
 
Microservices Tracing With Spring Cloud and Zipkin @Szczecin JUG
Microservices Tracing With Spring Cloud and Zipkin @Szczecin JUGMicroservices Tracing With Spring Cloud and Zipkin @Szczecin JUG
Microservices Tracing With Spring Cloud and Zipkin @Szczecin JUGMarcin Grzejszczak
 

Viewers also liked (9)

Optimizing ARM cortex a and cortex-m based heterogeneous multiprocessor syste...
Optimizing ARM cortex a and cortex-m based heterogeneous multiprocessor syste...Optimizing ARM cortex a and cortex-m based heterogeneous multiprocessor syste...
Optimizing ARM cortex a and cortex-m based heterogeneous multiprocessor syste...
 
A practical approach to securing embedded and io t platforms
A practical approach to securing embedded and io t platformsA practical approach to securing embedded and io t platforms
A practical approach to securing embedded and io t platforms
 
Developing functional safety systems with arm architecture solutions stroud
Developing functional safety systems with arm architecture solutions   stroudDeveloping functional safety systems with arm architecture solutions   stroud
Developing functional safety systems with arm architecture solutions stroud
 
Efficient software development with heterogeneous devices
Efficient software development with heterogeneous devicesEfficient software development with heterogeneous devices
Efficient software development with heterogeneous devices
 
The Proto-Burst Buffer: Experience with the flash-based file system on SDSC's...
The Proto-Burst Buffer: Experience with the flash-based file system on SDSC's...The Proto-Burst Buffer: Experience with the flash-based file system on SDSC's...
The Proto-Burst Buffer: Experience with the flash-based file system on SDSC's...
 
Software development in ar mv8 m architecture - yiu
Software development in ar mv8 m architecture - yiuSoftware development in ar mv8 m architecture - yiu
Software development in ar mv8 m architecture - yiu
 
So you think developing an SoC needs to be complex or expensive?
So you think developing an SoC needs to be complex or expensive?So you think developing an SoC needs to be complex or expensive?
So you think developing an SoC needs to be complex or expensive?
 
Microsoft Project Olympus AI Accelerator Chassis (HGX-1)
Microsoft Project Olympus AI Accelerator Chassis (HGX-1)Microsoft Project Olympus AI Accelerator Chassis (HGX-1)
Microsoft Project Olympus AI Accelerator Chassis (HGX-1)
 
Microservices Tracing With Spring Cloud and Zipkin @Szczecin JUG
Microservices Tracing With Spring Cloud and Zipkin @Szczecin JUGMicroservices Tracing With Spring Cloud and Zipkin @Szczecin JUG
Microservices Tracing With Spring Cloud and Zipkin @Szczecin JUG
 

Similar to SR-IOV: The Key Enabling Technology for Fully Virtualized HPC Clusters

AIST Super Green Cloud: lessons learned from the operation and the performanc...
AIST Super Green Cloud: lessons learned from the operation and the performanc...AIST Super Green Cloud: lessons learned from the operation and the performanc...
AIST Super Green Cloud: lessons learned from the operation and the performanc...Ryousei Takano
 
Sharing High-Performance Interconnects Across Multiple Virtual Machines
Sharing High-Performance Interconnects Across Multiple Virtual MachinesSharing High-Performance Interconnects Across Multiple Virtual Machines
Sharing High-Performance Interconnects Across Multiple Virtual Machinesinside-BigData.com
 
Building Efficient HPC Clouds with MCAPICH2 and RDMA-Hadoop over SR-IOV Infin...
Building Efficient HPC Clouds with MCAPICH2 and RDMA-Hadoop over SR-IOV Infin...Building Efficient HPC Clouds with MCAPICH2 and RDMA-Hadoop over SR-IOV Infin...
Building Efficient HPC Clouds with MCAPICH2 and RDMA-Hadoop over SR-IOV Infin...inside-BigData.com
 
Anton Moldovan "Building an efficient replication system for thousands of ter...
Anton Moldovan "Building an efficient replication system for thousands of ter...Anton Moldovan "Building an efficient replication system for thousands of ter...
Anton Moldovan "Building an efficient replication system for thousands of ter...Fwdays
 
Platforms for Accelerating the Software Defined and Virtual Infrastructure
Platforms for Accelerating the Software Defined and Virtual InfrastructurePlatforms for Accelerating the Software Defined and Virtual Infrastructure
Platforms for Accelerating the Software Defined and Virtual Infrastructure6WIND
 
Arquitetura Hibrida - Integrando seu Data Center com a Nuvem da AWS
Arquitetura Hibrida - Integrando seu Data Center com a Nuvem da AWSArquitetura Hibrida - Integrando seu Data Center com a Nuvem da AWS
Arquitetura Hibrida - Integrando seu Data Center com a Nuvem da AWSAmazon Web Services LATAM
 
The von Neumann Memory Barrier and Computer Architectures for the 21st Century
The von Neumann Memory Barrier and Computer Architectures for the 21st CenturyThe von Neumann Memory Barrier and Computer Architectures for the 21st Century
The von Neumann Memory Barrier and Computer Architectures for the 21st CenturyPerry Lea
 
End to End Processing of 3.7 Million Telemetry Events per Second using Lambda...
End to End Processing of 3.7 Million Telemetry Events per Second using Lambda...End to End Processing of 3.7 Million Telemetry Events per Second using Lambda...
End to End Processing of 3.7 Million Telemetry Events per Second using Lambda...DataWorks Summit/Hadoop Summit
 
You Call that Micro, Mr. Docker? How OSv and Unikernels Help Micro-services S...
You Call that Micro, Mr. Docker? How OSv and Unikernels Help Micro-services S...You Call that Micro, Mr. Docker? How OSv and Unikernels Help Micro-services S...
You Call that Micro, Mr. Docker? How OSv and Unikernels Help Micro-services S...rhatr
 
High Performance Cloud Computing
High Performance Cloud ComputingHigh Performance Cloud Computing
High Performance Cloud ComputingAmazon Web Services
 
High Performance Cloud Computing
High Performance Cloud ComputingHigh Performance Cloud Computing
High Performance Cloud ComputingAmazon Web Services
 
#VMUGMTL - Xsigo Breakout
#VMUGMTL - Xsigo Breakout#VMUGMTL - Xsigo Breakout
#VMUGMTL - Xsigo Breakout1CloudRoad.com
 
The Pace of Innovation - Pop-up Loft Tel Aviv
The Pace of Innovation - Pop-up Loft Tel AvivThe Pace of Innovation - Pop-up Loft Tel Aviv
The Pace of Innovation - Pop-up Loft Tel AvivAmazon Web Services
 
The Impact of Hardware and Software Version Changes on Apache Kafka Performan...
The Impact of Hardware and Software Version Changes on Apache Kafka Performan...The Impact of Hardware and Software Version Changes on Apache Kafka Performan...
The Impact of Hardware and Software Version Changes on Apache Kafka Performan...Paul Brebner
 
Real-time DeepLearning on IoT Sensor Data
Real-time DeepLearning on IoT Sensor DataReal-time DeepLearning on IoT Sensor Data
Real-time DeepLearning on IoT Sensor DataRomeo Kienzler
 
Cassandra Summit 2014: Active-Active Cassandra Behind the Scenes
Cassandra Summit 2014: Active-Active Cassandra Behind the ScenesCassandra Summit 2014: Active-Active Cassandra Behind the Scenes
Cassandra Summit 2014: Active-Active Cassandra Behind the ScenesDataStax Academy
 
ISC Cloud 2013 - Cloud Architectures for HPC – Industry Case Studies
 ISC Cloud 2013 - Cloud Architectures for HPC – Industry Case Studies ISC Cloud 2013 - Cloud Architectures for HPC – Industry Case Studies
ISC Cloud 2013 - Cloud Architectures for HPC – Industry Case StudiesOpenNebula Project
 
ISC Cloud 2013 - Cloud Architectures for HPC – Industry Case Studies
 ISC Cloud 2013 - Cloud Architectures for HPC – Industry Case Studies ISC Cloud 2013 - Cloud Architectures for HPC – Industry Case Studies
ISC Cloud 2013 - Cloud Architectures for HPC – Industry Case StudiesIgnacio M. Llorente
 
ClickOS_EE80777777777777777777777777777.pptx
ClickOS_EE80777777777777777777777777777.pptxClickOS_EE80777777777777777777777777777.pptx
ClickOS_EE80777777777777777777777777777.pptxBiHongPhc
 

Similar to SR-IOV: The Key Enabling Technology for Fully Virtualized HPC Clusters (20)

AIST Super Green Cloud: lessons learned from the operation and the performanc...
AIST Super Green Cloud: lessons learned from the operation and the performanc...AIST Super Green Cloud: lessons learned from the operation and the performanc...
AIST Super Green Cloud: lessons learned from the operation and the performanc...
 
Sharing High-Performance Interconnects Across Multiple Virtual Machines
Sharing High-Performance Interconnects Across Multiple Virtual MachinesSharing High-Performance Interconnects Across Multiple Virtual Machines
Sharing High-Performance Interconnects Across Multiple Virtual Machines
 
Building Efficient HPC Clouds with MCAPICH2 and RDMA-Hadoop over SR-IOV Infin...
Building Efficient HPC Clouds with MCAPICH2 and RDMA-Hadoop over SR-IOV Infin...Building Efficient HPC Clouds with MCAPICH2 and RDMA-Hadoop over SR-IOV Infin...
Building Efficient HPC Clouds with MCAPICH2 and RDMA-Hadoop over SR-IOV Infin...
 
Anton Moldovan "Building an efficient replication system for thousands of ter...
Anton Moldovan "Building an efficient replication system for thousands of ter...Anton Moldovan "Building an efficient replication system for thousands of ter...
Anton Moldovan "Building an efficient replication system for thousands of ter...
 
Platforms for Accelerating the Software Defined and Virtual Infrastructure
Platforms for Accelerating the Software Defined and Virtual InfrastructurePlatforms for Accelerating the Software Defined and Virtual Infrastructure
Platforms for Accelerating the Software Defined and Virtual Infrastructure
 
Arquitetura Hibrida - Integrando seu Data Center com a Nuvem da AWS
Arquitetura Hibrida - Integrando seu Data Center com a Nuvem da AWSArquitetura Hibrida - Integrando seu Data Center com a Nuvem da AWS
Arquitetura Hibrida - Integrando seu Data Center com a Nuvem da AWS
 
The von Neumann Memory Barrier and Computer Architectures for the 21st Century
The von Neumann Memory Barrier and Computer Architectures for the 21st CenturyThe von Neumann Memory Barrier and Computer Architectures for the 21st Century
The von Neumann Memory Barrier and Computer Architectures for the 21st Century
 
Big Data Management at CERN: The CMS Example
Big Data Management at CERN: The CMS ExampleBig Data Management at CERN: The CMS Example
Big Data Management at CERN: The CMS Example
 
End to End Processing of 3.7 Million Telemetry Events per Second using Lambda...
End to End Processing of 3.7 Million Telemetry Events per Second using Lambda...End to End Processing of 3.7 Million Telemetry Events per Second using Lambda...
End to End Processing of 3.7 Million Telemetry Events per Second using Lambda...
 
You Call that Micro, Mr. Docker? How OSv and Unikernels Help Micro-services S...
You Call that Micro, Mr. Docker? How OSv and Unikernels Help Micro-services S...You Call that Micro, Mr. Docker? How OSv and Unikernels Help Micro-services S...
You Call that Micro, Mr. Docker? How OSv and Unikernels Help Micro-services S...
 
High Performance Cloud Computing
High Performance Cloud ComputingHigh Performance Cloud Computing
High Performance Cloud Computing
 
High Performance Cloud Computing
High Performance Cloud ComputingHigh Performance Cloud Computing
High Performance Cloud Computing
 
#VMUGMTL - Xsigo Breakout
#VMUGMTL - Xsigo Breakout#VMUGMTL - Xsigo Breakout
#VMUGMTL - Xsigo Breakout
 
The Pace of Innovation - Pop-up Loft Tel Aviv
The Pace of Innovation - Pop-up Loft Tel AvivThe Pace of Innovation - Pop-up Loft Tel Aviv
The Pace of Innovation - Pop-up Loft Tel Aviv
 
The Impact of Hardware and Software Version Changes on Apache Kafka Performan...
The Impact of Hardware and Software Version Changes on Apache Kafka Performan...The Impact of Hardware and Software Version Changes on Apache Kafka Performan...
The Impact of Hardware and Software Version Changes on Apache Kafka Performan...
 
Real-time DeepLearning on IoT Sensor Data
Real-time DeepLearning on IoT Sensor DataReal-time DeepLearning on IoT Sensor Data
Real-time DeepLearning on IoT Sensor Data
 
Cassandra Summit 2014: Active-Active Cassandra Behind the Scenes
Cassandra Summit 2014: Active-Active Cassandra Behind the ScenesCassandra Summit 2014: Active-Active Cassandra Behind the Scenes
Cassandra Summit 2014: Active-Active Cassandra Behind the Scenes
 
ISC Cloud 2013 - Cloud Architectures for HPC – Industry Case Studies
 ISC Cloud 2013 - Cloud Architectures for HPC – Industry Case Studies ISC Cloud 2013 - Cloud Architectures for HPC – Industry Case Studies
ISC Cloud 2013 - Cloud Architectures for HPC – Industry Case Studies
 
ISC Cloud 2013 - Cloud Architectures for HPC – Industry Case Studies
 ISC Cloud 2013 - Cloud Architectures for HPC – Industry Case Studies ISC Cloud 2013 - Cloud Architectures for HPC – Industry Case Studies
ISC Cloud 2013 - Cloud Architectures for HPC – Industry Case Studies
 
ClickOS_EE80777777777777777777777777777.pptx
ClickOS_EE80777777777777777777777777777.pptxClickOS_EE80777777777777777777777777777.pptx
ClickOS_EE80777777777777777777777777777.pptx
 

More from Glenn K. Lockwood

Understanding and Measuring I/O Performance
Understanding and Measuring I/O PerformanceUnderstanding and Measuring I/O Performance
Understanding and Measuring I/O PerformanceGlenn K. Lockwood
 
ASCI Terascale Simulation Requirements and Deployments
ASCI Terascale Simulation Requirements and DeploymentsASCI Terascale Simulation Requirements and Deployments
ASCI Terascale Simulation Requirements and DeploymentsGlenn K. Lockwood
 
Large-scale Genomic Analysis Enabled by Gordon
Large-scale Genomic Analysis Enabled by GordonLarge-scale Genomic Analysis Enabled by Gordon
Large-scale Genomic Analysis Enabled by GordonGlenn K. Lockwood
 
Hadoop Streaming: Programming Hadoop without Java
Hadoop Streaming: Programming Hadoop without JavaHadoop Streaming: Programming Hadoop without Java
Hadoop Streaming: Programming Hadoop without JavaGlenn K. Lockwood
 

More from Glenn K. Lockwood (7)

Understanding and Measuring I/O Performance
Understanding and Measuring I/O PerformanceUnderstanding and Measuring I/O Performance
Understanding and Measuring I/O Performance
 
Parallel R and Hadoop
Parallel R and HadoopParallel R and Hadoop
Parallel R and Hadoop
 
ASCI Terascale Simulation Requirements and Deployments
ASCI Terascale Simulation Requirements and DeploymentsASCI Terascale Simulation Requirements and Deployments
ASCI Terascale Simulation Requirements and Deployments
 
Overview of Spark for HPC
Overview of Spark for HPCOverview of Spark for HPC
Overview of Spark for HPC
 
myHadoop 0.30
myHadoop 0.30myHadoop 0.30
myHadoop 0.30
 
Large-scale Genomic Analysis Enabled by Gordon
Large-scale Genomic Analysis Enabled by GordonLarge-scale Genomic Analysis Enabled by Gordon
Large-scale Genomic Analysis Enabled by Gordon
 
Hadoop Streaming: Programming Hadoop without Java
Hadoop Streaming: Programming Hadoop without JavaHadoop Streaming: Programming Hadoop without Java
Hadoop Streaming: Programming Hadoop without Java
 

Recently uploaded

Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...shyamraj55
 
Slack Application Development 101 Slides
Slack Application Development 101 SlidesSlack Application Development 101 Slides
Slack Application Development 101 Slidespraypatel2
 
[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdfhans926745
 
My Hashitalk Indonesia April 2024 Presentation
My Hashitalk Indonesia April 2024 PresentationMy Hashitalk Indonesia April 2024 Presentation
My Hashitalk Indonesia April 2024 PresentationRidwan Fadjar
 
Histor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slideHistor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slidevu2urc
 
08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking Men08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking MenDelhi Call girls
 
Swan(sea) Song – personal research during my six years at Swansea ... and bey...
Swan(sea) Song – personal research during my six years at Swansea ... and bey...Swan(sea) Song – personal research during my six years at Swansea ... and bey...
Swan(sea) Song – personal research during my six years at Swansea ... and bey...Alan Dix
 
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdfThe Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdfEnterprise Knowledge
 
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...Neo4j
 
A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)Gabriella Davis
 
GenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day PresentationGenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day PresentationMichael W. Hawkins
 
The Codex of Business Writing Software for Real-World Solutions 2.pptx
The Codex of Business Writing Software for Real-World Solutions 2.pptxThe Codex of Business Writing Software for Real-World Solutions 2.pptx
The Codex of Business Writing Software for Real-World Solutions 2.pptxMalak Abu Hammad
 
Maximizing Board Effectiveness 2024 Webinar.pptx
Maximizing Board Effectiveness 2024 Webinar.pptxMaximizing Board Effectiveness 2024 Webinar.pptx
Maximizing Board Effectiveness 2024 Webinar.pptxOnBoard
 
The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024Rafal Los
 
Understanding the Laravel MVC Architecture
Understanding the Laravel MVC ArchitectureUnderstanding the Laravel MVC Architecture
Understanding the Laravel MVC ArchitecturePixlogix Infotech
 
SQL Database Design For Developers at php[tek] 2024
SQL Database Design For Developers at php[tek] 2024SQL Database Design For Developers at php[tek] 2024
SQL Database Design For Developers at php[tek] 2024Scott Keck-Warren
 
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptxHampshireHUG
 
A Call to Action for Generative AI in 2024
A Call to Action for Generative AI in 2024A Call to Action for Generative AI in 2024
A Call to Action for Generative AI in 2024Results
 
The Evolution of Money: Digital Transformation and CBDCs in Central Banking
The Evolution of Money: Digital Transformation and CBDCs in Central BankingThe Evolution of Money: Digital Transformation and CBDCs in Central Banking
The Evolution of Money: Digital Transformation and CBDCs in Central BankingSelcen Ozturkcan
 
Boost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityBoost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityPrincipled Technologies
 

Recently uploaded (20)

Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
 
Slack Application Development 101 Slides
Slack Application Development 101 SlidesSlack Application Development 101 Slides
Slack Application Development 101 Slides
 
[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf
 
My Hashitalk Indonesia April 2024 Presentation
My Hashitalk Indonesia April 2024 PresentationMy Hashitalk Indonesia April 2024 Presentation
My Hashitalk Indonesia April 2024 Presentation
 
Histor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slideHistor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slide
 
08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking Men08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking Men
 
Swan(sea) Song – personal research during my six years at Swansea ... and bey...
Swan(sea) Song – personal research during my six years at Swansea ... and bey...Swan(sea) Song – personal research during my six years at Swansea ... and bey...
Swan(sea) Song – personal research during my six years at Swansea ... and bey...
 
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdfThe Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
 
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
 
A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)
 
GenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day PresentationGenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day Presentation
 
The Codex of Business Writing Software for Real-World Solutions 2.pptx
The Codex of Business Writing Software for Real-World Solutions 2.pptxThe Codex of Business Writing Software for Real-World Solutions 2.pptx
The Codex of Business Writing Software for Real-World Solutions 2.pptx
 
Maximizing Board Effectiveness 2024 Webinar.pptx
Maximizing Board Effectiveness 2024 Webinar.pptxMaximizing Board Effectiveness 2024 Webinar.pptx
Maximizing Board Effectiveness 2024 Webinar.pptx
 
The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024
 
Understanding the Laravel MVC Architecture
Understanding the Laravel MVC ArchitectureUnderstanding the Laravel MVC Architecture
Understanding the Laravel MVC Architecture
 
SQL Database Design For Developers at php[tek] 2024
SQL Database Design For Developers at php[tek] 2024SQL Database Design For Developers at php[tek] 2024
SQL Database Design For Developers at php[tek] 2024
 
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
 
A Call to Action for Generative AI in 2024
A Call to Action for Generative AI in 2024A Call to Action for Generative AI in 2024
A Call to Action for Generative AI in 2024
 
The Evolution of Money: Digital Transformation and CBDCs in Central Banking
The Evolution of Money: Digital Transformation and CBDCs in Central BankingThe Evolution of Money: Digital Transformation and CBDCs in Central Banking
The Evolution of Money: Digital Transformation and CBDCs in Central Banking
 
Boost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityBoost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivity
 

SR-IOV: The Key Enabling Technology for Fully Virtualized HPC Clusters

  • 1. SR-IOV: The Key Enabling Technology for Fully Virtualized HPC Clusters! Glenn K. Lockwood! Christopher Irving! Philip M. Papadopoulos! Mahidhar Tatineni! Rick Wagner! SAN DIEGO SUPERCOMPUTER CENTER
  • 2. Single Root I/O Virtualization in HPC! •  Problem: complex workflows demand increasing flexibility from HPC platforms" •  Virtualization = flexibility" •  Virtualization = IO performance loss (e.g., excessive DMA interrupts)" •  Solution: SR-IOV and Mellanox ConnectX-3 InfiniBand HCAs " •  One physical function (PF) à multiple virtual functions (VF), each with own DMA streams, memory space, interrupts" •  Allows DMA to bypass hypervisor to VMs! SAN DIEGO SUPERCOMPUTER CENTER
  • 3. High-Performance Virtualization on Comet ! •  Mellanox FDR InfiniBand HCAs with SR-IOV" •  Rocks and OpenStack Nova to manage VMs" •  Flexibility to support complex science gateways and web-based workflow engines" •  custom compute appliances and virtual clusters developed with FutureGrid and their existing expertise" •  backed by virtualized Lustre running over virtualized InfiniBand" SAN DIEGO SUPERCOMPUTER CENTER
  • 4. Hardware/Software Configuration of Test Cluster ! Native, SR-IOV! Amazon EC2! Platform" •  Rocks 6.1 (EL6)" •  Virtualization via kvm   •  Amazon Linux 2013.03 (EL6)" •  cc2.8xlarge Instances" CPUs" •  2x Xeon E5-2660 (2.2GHz)" •  16 cores per node" •  2x Xeon E5-2670 (2.6GHz)" •  16 cores per node" RAM" •  64 GB DDR3 DRAM" •  60.5 DDR3 DRAM" Interconnect" •  QDR4X InfiniBand" •  10 GbE" •  Mellanox ConnectX-3 (MT27500)" •  common placement group" •  Intel VT-d, SR-IOV enabled in firmware, kernel, drivers" •  mlx4_core  1.1" •  Mellanox OFED 2.0" •  HCA firmware 2.11.1192" SAN DIEGO SUPERCOMPUTER CENTER
  • 5. 50x less latency than Amazon EC2! •  SR-IOV! •  < 30% overhead for M < 128 bytes" •  < 10% overhead for eager send/recv" •  Overhead à 0% for bandwidth-limited regime" •  Amazon EC2! •  > 5000% worse latency" •  Time dependent (noisy)" 5 SAN DIEGO SUPERCOMPUTER CENTER OSU Microbenchmarks (3.9, osu_latency)"
  • 6. 10x more bandwidth than Amazon EC2! •  SR-IOV! •  < 2% bandwidth loss over entire range" •  > 95% peak bandwidth" •  Amazon EC2! •  < 35% peak bandwidth" •  900% to 2500% worse bandwidth than virtualized InfiniBand" 6 SAN DIEGO SUPERCOMPUTER CENTER OSU Microbenchmarks (3.9, osu_bw)"
  • 7. Weather Modeling – 15% Overhead! •  96-core (6-node) calculation" •  Nearest-neighbor communication" •  Scalable algorithms" •  SR-IOV incurs modest (15%) performance hit" •  ...but still still 20% faster*** than Amazon" SAN DIEGO SUPERCOMPUTER CENTER WRF 3.4.1 – 3hr forecast" *** 20% faster despite SR-IOV cluster having 20% slower CPUs"
  • 8. Quantum ESPRESSO: 5x Faster than EC2! •  48-core, 3 node calc" •  CG matrix inversion (irregular comm.)" •  3D FFT matrix transposes (All-to-all communication)" •  28% slower w/ SR-IOV" •  SR-IOV still > 500% faster*** than EC2" SAN DIEGO SUPERCOMPUTER CENTER Quantum Espresso 5.0.2 – DEISA AUSURF112 benchmark" *** 20% faster despite SR-IOV cluster having 20% slower CPUs"
  • 9. Conclusions! •  SR-IOV: huge step forward in high-performance virtualization" •  Shows substantial improvement in latency over Amazon EC2, and it provides nearly 0 bandwidth overhead! •  Benchmark application performance confirms this: significant improvement over EC2" •  SR-IOV: lowers performance barrier to virtualizing the interconnect and makes fully virtualized HPC clusters viable! •  Comet will deliver virtualized HPC to new/non-traditional communities that need flexibility without major loss of performance! SAN DIEGO SUPERCOMPUTER CENTER