Deployment Guide

Overview
7 VMware vSphere Bitfusion on Dell EMC PowerEdge servers | Deployment guide
2 Overview
With the new VMware vSphere Bitfusion software, graphics processing units (GPU) are no longer isolated
from other resources. GPUs are now shared in a virtualized pool of resources and you can access them
through any virtual machine in the infrastructure as shown in Figure 1. Similar to processors and storage
resources, GPU deployments can now benefit from optimized utilization, reduced Capex and Opex, and
accelerated development and deployment of R&D resources. Data scientists and AI developers can benefit
from how Bitfusion supports monitoring higher workloads.
BitFusion offers the following key features:
• Dynamic GPU attach anywhere
Bitfusion disaggregates your GPU compute and dynamically attaches GPUs anywhere in the
datacenter, just like attaching storage.
• Fractional GPUs for efficiency
Bitfusion enables use of any arbitrary fractions of GPUs. Support more users in the test and
development phase.
• Standards based accelerator access
Leverage GPUs across an infrastructure plus integrate evolving technologies as standards emerge.
• Application run time virtualization
Bitfusion attaches GPUs based on CUDA calls at run-time, maximizing utilization of GPU servers
anywhere in the network.
• Any application
Bitfusion is a transparent layer and runs with any workload in a Tensorflow or Pytorch ecosystem.