MIG: Difference between revisions

From HPCWIKI
Jump to navigation Jump to search
No edit summary
Line 2: Line 2:
NVIDIA introduced MIG(Multi-Instance GPU) since Ampere architecture.
NVIDIA introduced MIG(Multi-Instance GPU) since Ampere architecture.


MIG feature allows GPUs to be securely partitioned into up to seven separate GPU Instances for [[CUDA]] applications, so that multiple users use the same physical GPU with separate resources for optimal GPU utilization. <ref>https://docs.nvidia.com/datacenter/tesla/mig-user-guide/</ref>
MIG feature allows a single GPU into multiple fully isolated virtual GPU devices that are efficiently sized per-user-case, specifically smaller use-cases that only require a subset of GPU resources.


<div style="position: relative; width: 200px; height: auto; overflow: hidden;">
https://www.nvidia.com/content/dam/en-zz/Solutions/gtcs22/multi-instance-gpu/multi-instance-gpu-tech-works-1cc-l@2x.png
</div>


MIG ensures to providing each instance's processors have separate and isolated paths through the entire memory system - the on-chip crossbar ports, L2 cache banks, memory controllers, and DRAM address busses are all assigned uniquely to an individual instanceenhanced isolation GPU resources.
MIG ensures to providing each instance's processors have separate and isolated paths through the entire memory system - the on-chip crossbar ports, L2 cache banks, memory controllers, and DRAM address busses are all assigned uniquely to an individual instanceenhanced isolation GPU resources.
[[File:Nvidia-mig-example.png|right|frameless|nvidia-mig-example]]




With MIG, users will be able to see and schedule jobs on their new virtual GPU Instances as if they were physical GPUs. MIG works with [[Linux]] operating systems, supports containers using [[Docker]] Engine and  hypervisors such as Red Hat Virtualization and                        VMware vSphere.
Benefits of MIG on MIG featured GPU are
*Physical allocation of resourdces used by parallel GPU workloads - Secure multi-tenant environments with isolation and predictable QoS
MIG supports following configurations   
*Versatile profiles with dynamic configuration - Maximized utilization by configuring for specfic workloads
 
*[[CUDA]] programming model unchanged
*Bare-metal, including containers
*GPU pass-through virtualization to Linux guests on top of supported hypervisors
* vGPU on top of supported hypervisors




== References==
== References==
<references />
<references />

Revision as of 09:26, 22 May 2023

NVIDIA Multi-Instance GPU

NVIDIA introduced MIG(Multi-Instance GPU) since Ampere architecture.

MIG feature allows a single GPU into multiple fully isolated virtual GPU devices that are efficiently sized per-user-case, specifically smaller use-cases that only require a subset of GPU resources.


MIG ensures to providing each instance's processors have separate and isolated paths through the entire memory system - the on-chip crossbar ports, L2 cache banks, memory controllers, and DRAM address busses are all assigned uniquely to an individual instanceenhanced isolation GPU resources.

nvidia-mig-example


Benefits of MIG on MIG featured GPU are

  • Physical allocation of resourdces used by parallel GPU workloads - Secure multi-tenant environments with isolation and predictable QoS
  • Versatile profiles with dynamic configuration - Maximized utilization by configuring for specfic workloads
  • CUDA programming model unchanged


References

Retrieved from "http://wiki.hpcmate.com/index.php?title=MIG&oldid=962"