WilliamLam.com

VCF 9.1.1 - Enabling the the new AI Assistant for VCF in VCF Operations

by // 11 Comments

One feature in VMware Cloud Foundation (VCF) 9.1.1 that I am personally excited about is the new AI Assistant for VCF (Tech Preview). It is great to see AI being integrated directly into VCF for infrastructure teams, and while this is just the first release, I think it will open up many interesting possibilities for troubleshooting, reporting, operational insights, and ultimately improving operational efficiency by giving infrastructure teams valuable time back.

More importantly, as we enter a world where AI agents can operate at a scale and speed that humans simply cannot match, these capabilities will become increasingly important. AI will not only help infrastructure teams scale their operations, but also better prepare them to defend against and respond to threats operating at machine speed.

AI Assistant for VCF currently supports both cloud-based and VCF Private AI Services (PAIS) LLM model providers. For ease of setup and to quickly get started with the new capability, I will walk through the configuration using Google Gemini, which is currently the only supported cloud-based LLM provider.

[Read more...]

VCF 9.1.1 - Reducing CPU Utilization on AMD Ryzen Zen4/Zen5 by Masking the RDSEED Instruction

by // 2 Comments

While I am a huge fan of the recent AMD Ryzen Zen4/Zen5-based consumer platforms, including the versatile Minisforum MS-A2, it is well known at this point that these processors can exhibit slower entropy generation, which can lead to significantly higher CPU utilization.

I have published a number of workarounds (HERE, HERE and HERE) to address higher-than-expected CPU utilization across different VCF components, but this has continued to feel like a game of whack-a-mole, with the same underlying entropy issue manifesting itself in different services and workloads.

Right before VMware Explore, I had also been testing VMware Cloud Foundation (VCF) 9.1.1 in my lab while finalizing my presentations and workshops for the conference. After deploying certain VCF components, I noticed that CPU utilization had increased significantly. Long story short, after an extensive debugging session with VCF Engineering, we traced the increased CPU utilization to RDSEED instruction requests originating from application space. The simplest solution turned out to be masking the RDSEED instruction from the guest operating systems!

[Read more...]

VCF 9.1.1: Adopting the Reduced VCF Management Services (VCFMS) Footprint After an Upgrade

by // 8 Comments

When VCF Management Services (VCFMS) was introduced in VCF 9.1.0, its virtual machine sizing was optimized to accommodate the largest optional Day-N service. This reduced the likelihood of requiring a rollout to a larger VM size as additional services were enabled. However, for environments that did not deploy these optional Day-N services, some VCFMS virtual machines could be larger than necessary.

With VCF 9.1.1, VCFMS sizing has been optimized for the initial Day-0 deployment. In addition, individual VCFMS components have been further right-sized to ensure that each service reserves only the resources it requires, improving overall resource utilization. For example, a Simple (non-HA) deployment, you will now have one less VCFMS Worker Node (12 vCPU / 24GB memory).


It is important to note that the right-sized VCFMS worker nodes and component sizing are automatically applied for new VCF 9.1.1 deployments, but they are not automatically applied to existing deployments that are upgraded to VCF 9.1.1.

With that said, users who upgrade to VCF 9.1.1 can still realize the benefits of the reduced VCFMS footprint by running a post-upgrade script that produces the same end state as a new VCF 9.1.1 deployment. For those interested, you can also use the VCF Inspector Fling which will include the post-upgrade script workflow, so you can simply perform that using that tool as well.
[Read more...]