Deployment Guide

Pre-deployment requirements
14 VMware vSphere Bitfusion on Dell EMC PowerEdge servers | Deployment guide
4.3 Bitfusion server and client software
Bitfusion OVA is a VMware appliance prepackaged with GPU software and services. Bitfusion client package
runs on the virtual machines where, the applications make use of the GPU resources. To download the OVA
and client package, see the Download VMware vSphere Bitfusion page after logging into My VMware
account.
4.4 vCenter
Once the Bitfusion server OVA is deployed, select Bitfusion from the vCenter menu. vCenter Server 7.0 lists
the components connected to the server once the GPU hosts and client clusters are connected. By doing this,
embedded platform services are installed on the client cluster. To download the vCenter Server Appliance,
see the Download VMware vSphere page.
Note: You can download the vCenter Server Appliance from the Download VMware vSphere page after
logging in to My VMware account.
4.5 Client virtual machine
The client cluster has a virtual machince with CentOS installed with the required NVIDIA tools and drivers.
You can use this virtual machine to accesss the GPUs remotely.
Install the following components on the the CentOS virtual machine to set up the client cluster:
• Python 3 and pip3 package manager
• Compute Unified Device Architecture 10.0 Toolkit (CUDA) for Red Hat Enterprise Linux 7
• cuDNN 7 python library
• TensorFlow v1.13.1 GPU framework
• TensorFlow benchmark toolkit compatible with TensorFlow v1.13 framework
This client virtual machine is connected to the management and RDMA network through PVRDMA.
Note: For instructions to deploy the above pre-requisites on a client virtual machine, see the Running
TensorFlow on vSphere Bitfusion vSphere Bitfusion guide.
4.6 Connectivity
The Bitfusion server appliances, client virtual machine with the remote GPUs and vCenter are connected over
a dedicated management network. In addition to this, vSAN,vMotion and Hardware Acceleration
communication are all required to be connected to the client cluster.
To monitor the GPU traffic, a dedicated RDMA (RoCE) connection is established between the GPU hosts and
the client cluster hosts.
The Dell EMC PowerSwitch ToR is configured for VLANs to accommodate vSAN, vMotion and GPU data
traffic management. Two switches are set up with Virtual Link Trucking (VLT) for redundancy.
Route the Bitfusion Appliance management network subnet to access the internet and then download the
NVIDIA driver.