Will the hardware come with you?
Discontinued CPUs block the install, deprecated ones start a countdown, and the network cards fail quietly. Then the choice between import, converge and side by side, which is not the binary most people treat it as.
Discontinued CPUs block the install, deprecated ones start a countdown, and the network cards fail quietly. Then the choice between import, converge and side by side, which is not the binary most people treat it as.
Host count used to predict upgrade effort and no longer does. A VCF 9.x upgrade scales with component count, which barely changes with estate size, so the smallest estates carry the worst ratio.
VCF component patches are no longer synchronised and no single page enumerates a patch set. What the 12 October release actually contained, the order it has to be applied in, and what that does to a change record.
VCF Operations certificate management now reaches standalone vCenter 8.0 Update 3 and above, and VCF 5.2, without converting those estates to VCF. What 9.1.1 added, the CA options, and the licensing line that decides whether you can use it.
VCF 9.1.1 right-sizes VCF Management Services and adds a Small HA option, but an upgraded estate keeps its old sizing until somebody runs KB 455842. What that does to the with-or-without-the-management-plane decision.
A complaint gets you a meeting. A request gets you documents. A step-by-step guide for NSW parents, with sources.
A community Management Pack for VCF Operations that is better engineered than most vendor material, and the four things worth checking before it goes near a regulated estate.
Licensing now costs the same wherever you run it, which takes one variable out of every comparison. What separates the four hyperscaler VMware services is who operates the stack, not what is underneath it.
Data residency, operational, jurisdictional and technical sovereignty are separate questions. Most estates satisfy two and assume they have satisfied all four.
Three numbers decide every infrastructure business case, and most models get two of them badly wrong. Here is how to build one a sceptical CFO cannot dismantle.
A customer asked what database monitoring would cost. Ten minutes in it was clear they already had it. Entitled, not deployed, nobody had looked.
Fifteen questions, about four minutes. Scores your estate across five areas and tells you what I'd do next in each one. Nothing is sent anywhere.
Ai
Five stages in the path, and only the last two involve the model. Retrieval is memory bound, generation is GPU bound, and they do not belong on the same hosts.
architecture
They implement the same ideas. The difference is what each optimises for and what it costs you to run, which is a less exciting answer than people want.
architecture
The arithmetic is easy. The denominator is where models fall apart, and the biggest lever has nothing to do with hardware.
architecture
Quantisation is usually framed as a quality tradeoff. The more useful framing is bandwidth, because decode re-reads every weight for every token. Tags: architecture, operations
architecture
Passthrough, time-sliced vGPU and MIG do different things to the thing that actually limits you. And fractional slices don't divide capacity the way the spreadsheet says.
architecture
Inference behaves like two workloads with opposite requirements, sharing the same silicon. Once you see that, the strange utilisation numbers start making sense.
architecture
Of all the decisions in a VCF build, tenancy is the one made worst. Not because it's hard, but because of when it gets made.
operations
You upgraded and nothing changed. Most estates land on the platform and keep operating exactly as before. Here's the sequence I'd work through instead.
lifecycle
VMware on Azure, Google Cloud and Oracle has moved to BYOL, with deadlines. The nearest one is 31 October. And it quietly changes how you should compare clouds.
architecture
VCF on EC2 bare metal inside your own VPC. Not the managed service you may be thinking of, and the networking constraints will shape your design whether you like them or not.
architecture
Run AI, containers and virtual machines on one platform, on hardware you already own. What that adds up to, and what to plan around.
architecture
A single tenant on four GPUs used 37% of the streaming multiprocessors. Four tenants used 71%, with four times the throughput. Same box, no extra hardware.