Graphcore IPU-POD64

  1. Home
  2. Solutions
  3. Graphcore IPU-POD64
 

Explore new horizons with AI at scale

IPU-POD64 delivers ultimate flexibility to maximise all available space and power in your datacentre, no matter how it is provisioned. IPU-POD64 brings together world-class IPU compute with a choice of best in class datacenter technologies and systems in flexible, pre-qualified configurations, to ensure your datacentre is operating with maximum efficiency and performance.

BOOK A TEST DRIVEDOWNLOAD DATASHEETKEY FEATURES
Graphcore IPU-POD64

 

 
 

GRAPHCORE FAMILY

 

  • BOSTON GRAPHCORE POPLAR® Server

    The Boston Graphcore Poplar Server is fully qualified by Graphcore as the default server to host for IPU-POD
    systems.

    FIND OUT MORE

  • GRAPHCORE
    IPU-M2000

    The IPU-M2000 is the fundamental compute engine for IPU-based machine intelligence, built with the powerful Colossus MK2 IPU.

    FIND OUT MORE

  • GRAPHCORE
    IPU-POD16

    For innovators making new AI breakthroughs, the Graphcore IPU-POD16 opens up a new world of machine intelligence innovation.

    FIND OUT MORE

  • GRAPHCORE
    IPU-POD64

    IPU-POD64 delivers ultimate flexibility to maximise all available space and power in your datacentre, no matter how it is provisioned.

    FIND OUT MORE

 

 

 

 

 

Graphcore's Mk2 IPU systems

IPU-POD64: The multi-IPU-Machine scaleout solution for datacentres with large AI compute needs. Connect thousands of machines for large Machine Intelligence problems, or multiple, concurrent workloads. Featuring ultra-high bandwidth, low-latency communication, enabled by Graphcore's breakthrough IPU-Fabric technology.

DOWNLOAD DATASHEETBOOK A TEST DRIVE

 

 

Graphcore IPU-POD64

Key features

  • 64 x GC200 Mk2 IPUs
  • 16 x IPU-M2000
  • 16 petaFLOPS FP16.16, 4 petaFLOPS FP32 Performance
  • 94,208 IPU Cores
  • 565,440 Threads
  • 2.8Tbps IPU-Fabric
  • Air-Cooled

Contact us about your requirements

 

KEY FEATURES


 

UNPRECEDENTED SCALABILITY

Simplify and streamline your AI datacenter scale-out with pre-configured, pre-approved IPU-POD reference designs

 


 

POWERFUL AI
COMPUTE

Challenge the status quo by choosing powerful, parallel AI compute and differentiate your business with new AI breakthroughs

 


 

SOFTWARE DESIGNED FOR AI AT SCALE

Poplar flexibly and simply compiles AI models across any number of IPUs giving you back precious development time

 


 

SIMPLE, POWERFUL BUILT-IN NETWORKING

IPU-Fabric is designed from the ground-up for AI to provide close-to-constant communication latency, ready to extend with near-limitless scale

 

 

 

Graphcore Poplar

Introducing Poplar®

The Poplar SDK is a complete software stack, which was co-designed from scratch with the IPU, to implement our graph toolchain in an easy to use and flexible software development environment. At a high level, Poplar is fully integrated with standard machine learning frameworks so developers can port existing models easily, and get up and running out-of-the-box with new applications in a familiar environment.

DOWNLOAD WHITEPAPER

 

 

Download Our Latest Datasheet

IPU-POD64 brings together powerful IPU compute with a choice of best in class data center technologies and systems from leading technology providers in flexible, pre-qualified configurations, to ensure your data center is operating with maximum efficiency and performance, while making your data center AI deployments simpler and faster.

Download datasheet

Get in touch to discuss our range of solutions

+91 22 5002 3262
[email protected]
Contact Us

Find your solution

Test out any of our solutions at Boston Labs

To help our clients make informed decisions about new technologies, we have opened up our research & development facilities and actively encourage customers to try the latest platforms using their own tools and if necessary together with their existing hardware. Remote access is also available

Contact us

ISC 2024

Latest Event

ISC 2024 | 13th - 15th May 2024, Congress Center, Hamburg

International Super Computing is a can't miss event for anyone interested in HPC, tech, and more.

more info