Eric Sloof - NTPRO.NL

https://www.youtube.com/c/cloudnativefdn/videos

VMware Container for Folding@Home is a docker container for running folding at home client. This container is supported on both Docker standalone clients and on a Kubernetes Cluster. Optional command line toggle GPU support on or off as well as all other common FAH client command line in puts.

The Folding@Home container is configured to automatically join Team VMware ID 52737. Everyone is welcome to join! Check out http://vmwa.re/fah for team and individual statistics.

Intel Optane persistent memory (PMem) is a memory technology that can deliver larger capacities compared to traditional DDR4 memory.

Increasing the memory capacity in a VMware vSphere virtualized datacenter by operating in Memory Mode allows for an increase in VM density and a subsequent increase in application capacity and performance in the same datacenter footprint.

VMware vSphere 6.7 EP10 and later versions support Intel Optane persistent memory 100 series (Intel Optane PMem) with 2nd Generation Intel Xeon Scalable processors in App Direct and Memory Mode.

Past experiments have found some performance challenges with Memory Mode in certain circumstances as seen in a previous whitepaper.

A new persistent memory BIOS profile for Memory Mode has been developed that improves performance in some of these cases.

This paper explores the performance seen with this new BIOS profile using an Enterprise Java workload and a modified version of the VMware VMmark 3.1 benchmark.

VMware vSphere Bitfusion leverages its client-server architecture to share GPUs among multiple clients. Bitfusion dynamically allocates a GPU or multiple GPUs to a remote client and releases the GPU resources when they are no longer being used. Bitfusion works across AI frameworks, clouds, networks, and in environments such as virtual machines, containers, and notebooks

This paper includes recommendations for getting the best performance from VMware vSphere Bitfusion. Hardware and software tunings are included.