SlideShare a Scribd company logo
1 of 21
Intel Omni-path Architecture
Ashay Shirwadkar
Agenda
• Introduction
• End to end solution
• Software stack
• Network stack
• Features
Introduction
Product line
Fabric integration
Software stack
Host software stack
• Leverages OpenFabrics Alliance
(OFA)
• Open source elements
• Host software stack via OFA
• Intel Omni-Path FastFabric Tools,
Fabric Manager, and GUI
• OPA support is included in
standard linux distros
• Starting with RHEL 7.3 and SLES
12sp2
Libfabric
Performance Scaled Messaging 2(PSM 2)
• User level library which provides API for Intel Omni-Path HFI
• PSM is specifically designed for MPI to provide high MPI rates
• Provides matched queue (MQ), building block for tag matching send and
receive calls
• Implementation to scale upto millions of MPI ranks (WHAAAT?)
• Provides active message (AM) API to implement PGAS programming
model
• OpenSHMEM, GASNet etc.
• Connectionless with minimal on-apdapter state
• Back compatible with PSM (Intel TrueScale)
Onload vs Offload (Infiniband HCA vs Intel HFI)
PSM (Programmable I/O)
Host Driven Send
• Optimize latency and message rate for high
priority message
• PIO is done when size is less than or equal to 8kb
Eager Receive
• Data is stored in receive buffers
• Receive buffer copies buffer to application
buffer
PSM (SDMA)
Send DMA
• Optimizes bandwidth for Large messages
• 16 SDMA engines for CPU offload
• Generally done when size is greater than
16 kb
Direct data placement
• Data directly placed into application
buffer
Network Stack:
In respect to OSI stack,
• Layer 1.5: Link Transfer Protocol
• Responsible for reliable delivery of Layer 2 packets
• Flow control and link control
• Layer 2: Data Link Layer
• Fabric addressing, switching, QoS
• Partitioning support
• Layer 4-7: Application Layer
• Interface between software libraries and Intel OPA HFI
Link Transfer Protocol: FLITz and LTP
Link Transfer Protocol: Packet Integrity
Protection (PIP)
Link Transfer Protocol: Traffic Flow
Optimization (Disabled)
Link Transfer Protocol: Traffic Flow
Optimization (Enabled)
Dynamic Lane Scaling (DLS)
Data Link Layer
• Quality of Service support
• Bandwidth allocation and
traffic separation
• Protocol deadlock avoidance
(eg. request/response)
• Congestion management
• Adaptive routing
• Dispersive routing
• Partitions
• Isolation mechanism, where every packet being associated with a single
partition
Transport layer and key software
• On packet loss or data, retransmission of packets
• Software layers are provided to provide the network API to user
• PSM
• OFED Verbs
• OFI
END

More Related Content

What's hot

Tech Talk by John Casey (CTO) CPLANE_NETWORKS : High Performance OpenStack Ne...
Tech Talk by John Casey (CTO) CPLANE_NETWORKS : High Performance OpenStack Ne...Tech Talk by John Casey (CTO) CPLANE_NETWORKS : High Performance OpenStack Ne...
Tech Talk by John Casey (CTO) CPLANE_NETWORKS : High Performance OpenStack Ne...
nvirters
 
Eduroam workshop nic mitev probes - networkshop44
Eduroam workshop  nic mitev probes - networkshop44Eduroam workshop  nic mitev probes - networkshop44
Eduroam workshop nic mitev probes - networkshop44
Jisc
 
Support formobility
Support formobilitySupport formobility
Support formobility
Rahul Hada
 
Nov 2014 webinar Making The Transition From Ftp
Nov 2014 webinar Making The Transition From FtpNov 2014 webinar Making The Transition From Ftp
Nov 2014 webinar Making The Transition From Ftp
FileCatalyst
 

What's hot (20)

NP - Unit 4 - Routing - RIP, OSPF and Internet Multicasting
NP - Unit 4 - Routing - RIP, OSPF and Internet MulticastingNP - Unit 4 - Routing - RIP, OSPF and Internet Multicasting
NP - Unit 4 - Routing - RIP, OSPF and Internet Multicasting
 
message communication protocols in IoT
message communication protocols in IoTmessage communication protocols in IoT
message communication protocols in IoT
 
OPNFV Service Function Chaining
OPNFV Service Function ChainingOPNFV Service Function Chaining
OPNFV Service Function Chaining
 
web communication protocols in IoT
web communication protocols in IoTweb communication protocols in IoT
web communication protocols in IoT
 
Chapter07
Chapter07Chapter07
Chapter07
 
SFScon14: The OSS-Mailcluster of Raiffeisen OnLine GmbH
SFScon14: The OSS-Mailcluster of Raiffeisen OnLine GmbHSFScon14: The OSS-Mailcluster of Raiffeisen OnLine GmbH
SFScon14: The OSS-Mailcluster of Raiffeisen OnLine GmbH
 
The Internet Protocol Suite (commonly known as TCP/IP)
The Internet Protocol Suite (commonly known as TCP/IP) The Internet Protocol Suite (commonly known as TCP/IP)
The Internet Protocol Suite (commonly known as TCP/IP)
 
Fog & edge computing(coap )
Fog & edge computing(coap )Fog & edge computing(coap )
Fog & edge computing(coap )
 
1005 cern-active mq-v2
1005 cern-active mq-v21005 cern-active mq-v2
1005 cern-active mq-v2
 
Tech Talk by John Casey (CTO) CPLANE_NETWORKS : High Performance OpenStack Ne...
Tech Talk by John Casey (CTO) CPLANE_NETWORKS : High Performance OpenStack Ne...Tech Talk by John Casey (CTO) CPLANE_NETWORKS : High Performance OpenStack Ne...
Tech Talk by John Casey (CTO) CPLANE_NETWORKS : High Performance OpenStack Ne...
 
Eduroam workshop nic mitev probes - networkshop44
Eduroam workshop  nic mitev probes - networkshop44Eduroam workshop  nic mitev probes - networkshop44
Eduroam workshop nic mitev probes - networkshop44
 
Chapter13
Chapter13Chapter13
Chapter13
 
Support formobility
Support formobilitySupport formobility
Support formobility
 
Beyond FTP & hard drives: Accelerating LAN file transfers
Beyond FTP & hard drives: Accelerating LAN file transfersBeyond FTP & hard drives: Accelerating LAN file transfers
Beyond FTP & hard drives: Accelerating LAN file transfers
 
Nov 2014 webinar Making The Transition From Ftp
Nov 2014 webinar Making The Transition From FtpNov 2014 webinar Making The Transition From Ftp
Nov 2014 webinar Making The Transition From Ftp
 
A New Internet? Introduction to HTTP/2, QUIC and DOH
A New Internet? Introduction to HTTP/2, QUIC and DOHA New Internet? Introduction to HTTP/2, QUIC and DOH
A New Internet? Introduction to HTTP/2, QUIC and DOH
 
Equal Cost Multipath Routing in FOKUS OpenSDNCore
Equal Cost Multipath Routing in FOKUS OpenSDNCoreEqual Cost Multipath Routing in FOKUS OpenSDNCore
Equal Cost Multipath Routing in FOKUS OpenSDNCore
 
protocol and the TCP/IP suite Chapter 02
 protocol and the TCP/IP suite Chapter 02 protocol and the TCP/IP suite Chapter 02
protocol and the TCP/IP suite Chapter 02
 
Light Reading BTE_SDNtoolbox_June_2015
Light Reading BTE_SDNtoolbox_June_2015Light Reading BTE_SDNtoolbox_June_2015
Light Reading BTE_SDNtoolbox_June_2015
 
Traffic Engineering in Software-Defined Networks
Traffic Engineering in Software-Defined NetworksTraffic Engineering in Software-Defined Networks
Traffic Engineering in Software-Defined Networks
 

Similar to Intel omni path architecture

Web technologies: recap on TCP-IP
Web technologies: recap on TCP-IPWeb technologies: recap on TCP-IP
Web technologies: recap on TCP-IP
Piero Fraternali
 

Similar to Intel omni path architecture (20)

Avani
AvaniAvani
Avani
 
Unit 2 ppt-idc
Unit 2 ppt-idcUnit 2 ppt-idc
Unit 2 ppt-idc
 
OpenPOWER Acceleration of HPCC Systems
OpenPOWER Acceleration of HPCC SystemsOpenPOWER Acceleration of HPCC Systems
OpenPOWER Acceleration of HPCC Systems
 
Hyper Transport Technology
Hyper Transport TechnologyHyper Transport Technology
Hyper Transport Technology
 
Bhargava Presentation.ppt
Bhargava Presentation.pptBhargava Presentation.ppt
Bhargava Presentation.ppt
 
Bhargava Presentation.ppt
Bhargava Presentation.pptBhargava Presentation.ppt
Bhargava Presentation.ppt
 
Osi
OsiOsi
Osi
 
OSI Model
OSI ModelOSI Model
OSI Model
 
Apache Kafka
Apache KafkaApache Kafka
Apache Kafka
 
Mobility managment and io t standards
Mobility managment and io t standardsMobility managment and io t standards
Mobility managment and io t standards
 
From Device to Data Center to Insights: Architectural Considerations for the ...
From Device to Data Center to Insights: Architectural Considerations for the ...From Device to Data Center to Insights: Architectural Considerations for the ...
From Device to Data Center to Insights: Architectural Considerations for the ...
 
Web technologies: recap on TCP-IP
Web technologies: recap on TCP-IPWeb technologies: recap on TCP-IP
Web technologies: recap on TCP-IP
 
File Transfer Protocol (FTP)
File Transfer Protocol (FTP)File Transfer Protocol (FTP)
File Transfer Protocol (FTP)
 
TCP/IP Protocols
TCP/IP ProtocolsTCP/IP Protocols
TCP/IP Protocols
 
UCX: An Open Source Framework for HPC Network APIs and Beyond
UCX: An Open Source Framework for HPC Network APIs and BeyondUCX: An Open Source Framework for HPC Network APIs and Beyond
UCX: An Open Source Framework for HPC Network APIs and Beyond
 
Music city data Hail Hydrate! from stream to lake
Music city data Hail Hydrate! from stream to lakeMusic city data Hail Hydrate! from stream to lake
Music city data Hail Hydrate! from stream to lake
 
From Device to Data Center to Insights
From Device to Data Center to InsightsFrom Device to Data Center to Insights
From Device to Data Center to Insights
 
Accelerate Big Data Processing with High-Performance Computing Technologies
Accelerate Big Data Processing with High-Performance Computing TechnologiesAccelerate Big Data Processing with High-Performance Computing Technologies
Accelerate Big Data Processing with High-Performance Computing Technologies
 
ODP Presentation LinuxCon NA 2014
ODP Presentation LinuxCon NA 2014ODP Presentation LinuxCon NA 2014
ODP Presentation LinuxCon NA 2014
 
IT-NET GROUP 3 REPORT.pptx
IT-NET GROUP 3 REPORT.pptxIT-NET GROUP 3 REPORT.pptx
IT-NET GROUP 3 REPORT.pptx
 

Recently uploaded

UNIT-V FMM.HYDRAULIC TURBINE - Construction and working
UNIT-V FMM.HYDRAULIC TURBINE - Construction and workingUNIT-V FMM.HYDRAULIC TURBINE - Construction and working
UNIT-V FMM.HYDRAULIC TURBINE - Construction and working
rknatarajan
 
Call Girls in Ramesh Nagar Delhi 💯 Call Us 🔝9953056974 🔝 Escort Service
Call Girls in Ramesh Nagar Delhi 💯 Call Us 🔝9953056974 🔝 Escort ServiceCall Girls in Ramesh Nagar Delhi 💯 Call Us 🔝9953056974 🔝 Escort Service
Call Girls in Ramesh Nagar Delhi 💯 Call Us 🔝9953056974 🔝 Escort Service
9953056974 Low Rate Call Girls In Saket, Delhi NCR
 

Recently uploaded (20)

UNIT-II FMM-Flow Through Circular Conduits
UNIT-II FMM-Flow Through Circular ConduitsUNIT-II FMM-Flow Through Circular Conduits
UNIT-II FMM-Flow Through Circular Conduits
 
Generative AI or GenAI technology based PPT
Generative AI or GenAI technology based PPTGenerative AI or GenAI technology based PPT
Generative AI or GenAI technology based PPT
 
Double rodded leveling 1 pdf activity 01
Double rodded leveling 1 pdf activity 01Double rodded leveling 1 pdf activity 01
Double rodded leveling 1 pdf activity 01
 
UNIT - IV - Air Compressors and its Performance
UNIT - IV - Air Compressors and its PerformanceUNIT - IV - Air Compressors and its Performance
UNIT - IV - Air Compressors and its Performance
 
(INDIRA) Call Girl Aurangabad Call Now 8617697112 Aurangabad Escorts 24x7
(INDIRA) Call Girl Aurangabad Call Now 8617697112 Aurangabad Escorts 24x7(INDIRA) Call Girl Aurangabad Call Now 8617697112 Aurangabad Escorts 24x7
(INDIRA) Call Girl Aurangabad Call Now 8617697112 Aurangabad Escorts 24x7
 
PVC VS. FIBERGLASS (FRP) GRAVITY SEWER - UNI BELL
PVC VS. FIBERGLASS (FRP) GRAVITY SEWER - UNI BELLPVC VS. FIBERGLASS (FRP) GRAVITY SEWER - UNI BELL
PVC VS. FIBERGLASS (FRP) GRAVITY SEWER - UNI BELL
 
KubeKraft presentation @CloudNativeHooghly
KubeKraft presentation @CloudNativeHooghlyKubeKraft presentation @CloudNativeHooghly
KubeKraft presentation @CloudNativeHooghly
 
Online banking management system project.pdf
Online banking management system project.pdfOnline banking management system project.pdf
Online banking management system project.pdf
 
Java Programming :Event Handling(Types of Events)
Java Programming :Event Handling(Types of Events)Java Programming :Event Handling(Types of Events)
Java Programming :Event Handling(Types of Events)
 
CCS335 _ Neural Networks and Deep Learning Laboratory_Lab Complete Record
CCS335 _ Neural Networks and Deep Learning Laboratory_Lab Complete RecordCCS335 _ Neural Networks and Deep Learning Laboratory_Lab Complete Record
CCS335 _ Neural Networks and Deep Learning Laboratory_Lab Complete Record
 
UNIT-V FMM.HYDRAULIC TURBINE - Construction and working
UNIT-V FMM.HYDRAULIC TURBINE - Construction and workingUNIT-V FMM.HYDRAULIC TURBINE - Construction and working
UNIT-V FMM.HYDRAULIC TURBINE - Construction and working
 
Call Girls in Ramesh Nagar Delhi 💯 Call Us 🔝9953056974 🔝 Escort Service
Call Girls in Ramesh Nagar Delhi 💯 Call Us 🔝9953056974 🔝 Escort ServiceCall Girls in Ramesh Nagar Delhi 💯 Call Us 🔝9953056974 🔝 Escort Service
Call Girls in Ramesh Nagar Delhi 💯 Call Us 🔝9953056974 🔝 Escort Service
 
The Most Attractive Pune Call Girls Budhwar Peth 8250192130 Will You Miss Thi...
The Most Attractive Pune Call Girls Budhwar Peth 8250192130 Will You Miss Thi...The Most Attractive Pune Call Girls Budhwar Peth 8250192130 Will You Miss Thi...
The Most Attractive Pune Call Girls Budhwar Peth 8250192130 Will You Miss Thi...
 
ONLINE FOOD ORDER SYSTEM PROJECT REPORT.pdf
ONLINE FOOD ORDER SYSTEM PROJECT REPORT.pdfONLINE FOOD ORDER SYSTEM PROJECT REPORT.pdf
ONLINE FOOD ORDER SYSTEM PROJECT REPORT.pdf
 
The Most Attractive Pune Call Girls Manchar 8250192130 Will You Miss This Cha...
The Most Attractive Pune Call Girls Manchar 8250192130 Will You Miss This Cha...The Most Attractive Pune Call Girls Manchar 8250192130 Will You Miss This Cha...
The Most Attractive Pune Call Girls Manchar 8250192130 Will You Miss This Cha...
 
Booking open Available Pune Call Girls Koregaon Park 6297143586 Call Hot Ind...
Booking open Available Pune Call Girls Koregaon Park  6297143586 Call Hot Ind...Booking open Available Pune Call Girls Koregaon Park  6297143586 Call Hot Ind...
Booking open Available Pune Call Girls Koregaon Park 6297143586 Call Hot Ind...
 
Booking open Available Pune Call Girls Pargaon 6297143586 Call Hot Indian Gi...
Booking open Available Pune Call Girls Pargaon  6297143586 Call Hot Indian Gi...Booking open Available Pune Call Girls Pargaon  6297143586 Call Hot Indian Gi...
Booking open Available Pune Call Girls Pargaon 6297143586 Call Hot Indian Gi...
 
Thermal Engineering -unit - III & IV.ppt
Thermal Engineering -unit - III & IV.pptThermal Engineering -unit - III & IV.ppt
Thermal Engineering -unit - III & IV.ppt
 
Vivazz, Mieres Social Housing Design Spain
Vivazz, Mieres Social Housing Design SpainVivazz, Mieres Social Housing Design Spain
Vivazz, Mieres Social Housing Design Spain
 
Intze Overhead Water Tank Design by Working Stress - IS Method.pdf
Intze Overhead Water Tank  Design by Working Stress - IS Method.pdfIntze Overhead Water Tank  Design by Working Stress - IS Method.pdf
Intze Overhead Water Tank Design by Working Stress - IS Method.pdf
 

Intel omni path architecture

  • 2. Agenda • Introduction • End to end solution • Software stack • Network stack • Features
  • 7. Host software stack • Leverages OpenFabrics Alliance (OFA) • Open source elements • Host software stack via OFA • Intel Omni-Path FastFabric Tools, Fabric Manager, and GUI • OPA support is included in standard linux distros • Starting with RHEL 7.3 and SLES 12sp2
  • 9. Performance Scaled Messaging 2(PSM 2) • User level library which provides API for Intel Omni-Path HFI • PSM is specifically designed for MPI to provide high MPI rates • Provides matched queue (MQ), building block for tag matching send and receive calls • Implementation to scale upto millions of MPI ranks (WHAAAT?) • Provides active message (AM) API to implement PGAS programming model • OpenSHMEM, GASNet etc. • Connectionless with minimal on-apdapter state • Back compatible with PSM (Intel TrueScale)
  • 10. Onload vs Offload (Infiniband HCA vs Intel HFI)
  • 11. PSM (Programmable I/O) Host Driven Send • Optimize latency and message rate for high priority message • PIO is done when size is less than or equal to 8kb Eager Receive • Data is stored in receive buffers • Receive buffer copies buffer to application buffer
  • 12. PSM (SDMA) Send DMA • Optimizes bandwidth for Large messages • 16 SDMA engines for CPU offload • Generally done when size is greater than 16 kb Direct data placement • Data directly placed into application buffer
  • 13. Network Stack: In respect to OSI stack, • Layer 1.5: Link Transfer Protocol • Responsible for reliable delivery of Layer 2 packets • Flow control and link control • Layer 2: Data Link Layer • Fabric addressing, switching, QoS • Partitioning support • Layer 4-7: Application Layer • Interface between software libraries and Intel OPA HFI
  • 14. Link Transfer Protocol: FLITz and LTP
  • 15. Link Transfer Protocol: Packet Integrity Protection (PIP)
  • 16. Link Transfer Protocol: Traffic Flow Optimization (Disabled)
  • 17. Link Transfer Protocol: Traffic Flow Optimization (Enabled)
  • 19. Data Link Layer • Quality of Service support • Bandwidth allocation and traffic separation • Protocol deadlock avoidance (eg. request/response) • Congestion management • Adaptive routing • Dispersive routing • Partitions • Isolation mechanism, where every packet being associated with a single partition
  • 20. Transport layer and key software • On packet loss or data, retransmission of packets • Software layers are provided to provide the network API to user • PSM • OFED Verbs • OFI
  • 21. END

Editor's Notes

  1. Mellanox and Qlogic were in infiniband Qlogic acquisition Cray interconnect (Intellectual property)
  2. ASIC (HFI or Switch) Edge switches have less ports and protocol supported Director switches have more ports and protocol
  3. Today: card modules In future: MCP packages Therefore we will not consume PCIe slot used for OPA
  4. Heart of compute software Boot over fabric, API for management of storage, drivers, middleware Element mgmt. stack: on switches (GUI and CLI)
  5. FF Tools: Fabric bring up and debug FM: Same as Subnet Manager in Infiniband GUI: Fabric monitoring
  6. Exporting fabric communication services to applications We had libibverbs (for RDMA) Applications and providers Providers: sockets are used for fallback RDMA basics (Queue pairs, work queue elements etc.) Draw a diagram for MQ
  7. User lib for Omnipath HFI Specifically designed for MPI
  8. Rank is a logical way of numbering processes. Connection state is stored in DRAM IB: Connection state Buffering and state machines L4 is offloaded Small portion is placed in cache
  9. Small messages let CPU do the work Explaining is HFI will be put down throughput Does not make sense to offload Same with receive, yes they do copy but again no Headers are not big in comparison with WQE
  10. Direct message placement 16 SDMA engines to offload Header generator: Give template and tell which fields to change
  11. RDMA basics (Queue pairs, work queue elements etc.)
  12. 1 type bit: Body FLITZ (1), HEAD (10), Tail (100), Others (000) control flitz (retransmission) and communication flitz 2 Types of LTP NULL ltp (not stored in reply buffer) and LTP with FLITZ
  13. Infiniband uses FEC which is checked on endpoint PIP has that on hop basis (can not be disabled)
  14. In infiniband ,whole lane is taken down OPA, does not take link down, uses PIP
  15. Adaptive: identifies congestion and adjusts itself Dispersive: Disperses the packets over multiple lanes Uses multipath between endpoints Partitions: Communication is allowed within the partition Either full or limited member Managed by fabric manager
  16. Adaptive: identifies congestion and adjusts itself Dispersive: Disperses the packets over multiple lanes Uses multipath between endpoints Partitions: Communication is allowed within the partition Either full or limited member Managed by fabric manager