Whole GPU, Passthrough, vGPU, MIG, or Time Slicing? The Enterprise GPU Allocation Decision Matrix

by

Introduction Enterprise GPU design becomes confused when several different decisions are compressed into one question: “How should we share the GPU?” That question mixes hardware assignment, virtualization, Kubernetes scheduling, tenant isolation, business priority, and service-level commitments. The result is often a platform that advertises many GPU “slices” but cannot explain what each slice guarantees. A … Read more

VCF Deprecations and Removals: An Upgrade Readiness Guide

by

VCF 9.1 creates a familiar operational trap: a platform team can complete a release-note review and still not be ready to upgrade. The reason is that support changes do not arrive as one clean class of event. Some capabilities remain functional but are deprecated. Some hardware operates under restricted support. Some components have already been … Read more

MCP vs A2A in 2026: Which Protocol Does Your AI Architecture Actually Need?

by

The architecture distinction is visible before the protocol details begin. A2A operates horizontally between independently operated agents. MCP operates vertically between an AI application or agent and the tools, data, prompts, and enterprise capabilities it consumes. That difference is the foundation for the entire protocol decision. Enterprise AI teams are increasingly asking whether they should … Read more

Brownfield vSphere to VMware Cloud Foundation 9.1: Import, Converge, or Rebuild?

by

Introduction The phrase “import an existing vCenter” sounds safer than it really is. It suggests that VMware Cloud Foundation reads an inventory, registers a few objects, and leaves the underlying environment largely untouched. That description is incomplete. In VMware Cloud Foundation 9.1, brownfield adoption is a change in platform ownership, lifecycle control, management topology, networking … Read more

How to Execute, Verify, and Roll Back Agent Actions: Enterprise Agent Control Plane Series, Part 4

by

TL;DR An agent action is not complete when the MCP server returns a successful response. It is complete only when the runtime proves that the intended post-condition exists in the authoritative system and that no unapproved side effect occurred. The execution runtime should: The governing principle is: The runtime does not trust that an action … Read more

How Many AI Workloads Can My GPU Platform Really Support?

by

TL;DR An expected workload count is not a GPU requirement. Thirty concurrent notebooks, RAG services, inference endpoints, fine-tuning jobs, or distributed training runs can create radically different demands for GPU memory, accelerator time, CPU, system memory, storage throughput, metadata operations, network bandwidth, scheduling policy, and failure reserve. Capacity planning should begin by defining workload service … Read more

How to Deploy the NVIDIA RAG Blueprint on Kubernetes with Helm

by

TL;DR The NVIDIA RAG Blueprint is not one application pod. It is a coordinated retrieval platform that combines an ingestion service, a RAG server, NVIDIA NIM microservices, NV-Ingest, a vector database, object storage, model caches, and supporting Kubernetes operators. For the current 2.6.0 release, Elasticsearch is the default vector database and SeaweedFS is the default … Read more

Who Owns the Failure? Building a Support RACI for a Multivendor Private AI Platform

by

TL;DR A multivendor private AI platform is not operationally complete when the hardware is installed, the GPUs are visible, and the first model endpoint responds. It is complete when the organization knows who performs the first diagnostic action when any part of the stack fails. The customer should retain one accountable service owner and one … Read more

How to Keep Secrets and Sensitive Data Out of AI Prompts, Embeddings, Traces, and Logs

by

TL;DR Sensitive data protection for AI is not a prompt-writing problem. It is a data-path control problem. The safest operating model classifies data before it enters the AI workflow, blocks credentials and restricted records by default, redacts or tokenizes approved confidential data, preserves source permissions in retrieval, disables prompt and tool-content telemetry unless explicitly approved, … Read more

Operationalizing the Agent Blast Radius Model: Policy Gates, Tool Contracts, and Rollback Controls

by

A risk model is only useful if it changes what happens before the agent acts. The Agent Blast Radius Model gives teams a way to classify AI agent actions by autonomy level, tool scope, transaction impact, and rollback feasibility. That model helps architects, security teams, platform owners, and business stakeholders agree on how much freedom … Read more

VMware Cloud Foundation 9.1 Release Notes: What Changed for Operators

by

VMware Cloud Foundation 9.1 became available on May 12, 2026, as build 25377994. The release covers infrastructure efficiency, Kubernetes operations, private AI, cyber resilience, platform scale, and lifecycle improvements. The headline features are significant, but they are not the whole story. [A] The more consequential change is architectural. VCF 9.1 consolidates more fleet lifecycle, software-depot, … Read more

How to Govern the MCP Gateway and MCP Server Layer: Enterprise Agent Control Plane Series, Part 3

by

TL;DR The MCP gateway should be the governed capability-access layer beneath the trusted agent controller. It should admit approved servers, expose only permitted tools, validate schemas, enforce identity and routing rules, isolate credentials, detect capability changes, and preserve trace evidence. The gateway should not become the only security control. Each MCP server must still enforce … Read more

NSX VPC or Another Workload Domain? Choosing the Right Isolation Boundary in VCF 9.1

by

TL;DR Do not create a VMware Cloud Foundation workload domain every time a tenant, business unit, application, or security team asks for separation. In VCF 9.1, the correct design is usually to choose the smallest boundary that satisfies the strongest verified requirement. A subnet or segment separates network attachment. An NSX VPC separates application networking, … Read more

Your AI Factory Is a Data Pipeline: Where PowerScale, PowerFlex, vSAN, Object Storage, and Local NVMe Belong

by

Introduction Enterprise AI architecture is often drawn from the compute layer outward. The GPU cluster sits in the middle, Kubernetes or virtual machines surround it, and storage appears as one cylinder at the bottom labeled data. That drawing is convenient, but it hides the design problem that causes many AI platforms to underperform or become … Read more

Green Hardware Does Not Mean a Ready AI Platform: Commissioning VCF Private AI Services

by

TL;DR A rack of healthy GPU servers is not a production-ready VCF Private AI platform. Production acceptance should prove the complete service chain: supported hardware, aligned ESXi and NVIDIA software, a healthy VCF 9.1 workload domain, stable NSX and shared infrastructure services, a ready Supervisor and VKS layer, functioning tenant controls, trusted Harbor and model … Read more

How to Deploy VMware Private AI Foundation with NVIDIA on VCF 9.1

by

TL;DR Deploying VMware Private AI Foundation with NVIDIA on VCF 9.1 is not a single-product installation. It is an integrated platform deployment spanning the VCF workload domain, GPU-enabled ESXi hosts, NVIDIA drivers and licensing, vSphere Supervisor, namespaces, Private AI Services, Harbor, identity, networking, certificates, and the AI consumption model. The most important design decision happens … Read more

How to Deploy the NVIDIA RAG Blueprint on Kubernetes with Helm

by

TL;DR The NVIDIA RAG Blueprint is not one application pod. It is a coordinated retrieval platform that combines an ingestion service, a RAG server, NVIDIA NIM microservices, NV-Ingest, a vector database, object storage, model caches, and supporting Kubernetes operators. For the current 2.6.0 release, Elasticsearch is the default vector database and SeaweedFS is the default … Read more

How to Run Coding Agents Safely Inside CI/CD Pipelines

by

TL;DR A coding agent should enter a CI/CD pipeline as an untrusted change producer, not as a privileged developer account. Give it an ephemeral sandbox, read-only repository access, a writable workspace, tightly controlled network egress, no deployment credentials, and no direct path to a protected branch. The agent should produce a patch and evidence bundle. … Read more

How to Deploy NVIDIA NIM Microservices on Kubernetes with the NIM Operator

by

TL;DR NVIDIA NIM can be deployed on Kubernetes through Helm or managed declaratively through the NVIDIA NIM Operator. The operator-based path is the better fit when you want Kubernetes-native lifecycle management for model caching, GPU scheduling, health probes, service exposure, scaling, and upgrades. The practical sequence is straightforward, but the dependencies matter. Build a supported … Read more