Rdma Roce

Use when the user asks about RDMA, RoCEv2, InfiniBand, GPUDirect RDMA, ibverbs, PFC, ECN/DCQCN, lossless Ethernet or NCCL falling back to NET/Socket. Produces a fabric checklist, validation commands (perftest, nccl-tests) and NCCL env settings. Not for choosing DDP/FSDP parallelism (use distributed-training).

45 stars 6 forksTesting & QA

Install

skills CLI (Claude Code, Cursor, Codex and more)
npx skills add j4flmao/agent-skills --skill rdma-roce
Manual install
git clone --depth 1 https://github.com/j4flmao/agent-skills.git /tmp/agent-skills
mkdir -p ~/.claude/skills
cp -r /tmp/agent-skills/skills/hardware-acceleration/rdma-roce ~/.claude/skills/rdma-roce

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# RDMA and RoCEv2

## Purpose
RDMA lets a NIC read and write registered memory on a remote host without involving the remote CPU or the kernel in the data path. It is the transport under NCCL multi-node collectives. InfiniBand provides it natively with credit-based lossless links; RoCEv2 carries the InfiniBand transport over UDP/IP on Ethernet and depends on careful congestion configuration. Most "slow multi-node training" problems are fabric misconfiguration.

## Agent Protocol

### Trigger
"rdma", "roce", "rocev2", "infiniband", "gpudirect", "ibverbs", "queue pair", "pfc", "ecn", "dcqcn", "lossless ethernet", "NCCL_IB_*", "nccl net/socket".

### Input Context
- NIC model and firmware, switch model, topology (rail-optimized, fat-tree), link speeds.
- GPUs per node, NIC-GPU PCIe affinity, OS/driver stack (MLNX_OFED / DOCA, nvidia-peermem or dma-buf).
- Symptoms: NCCL logs, `nccl-tests` bandwidth, switch counters.

### Output Artifact
Fabric checklist + validation commands + NCCL environment settings. No file unless requested.

### Response Format
```
Transport seen: {IB|RoCEv2|Socket} GDR: {on|off}
Measured: {busbw GB/s} vs expected {GB/s}
Finding: {misconfig}
Fix: {setting} Verify: {command}
```

### Completion Criteria
- [ ] NCCL uses the IB/RoCE transport with GPUDirect RDMA (not Socket).
- [ ] `nccl-tests` bus bandwidth near line rate for large messages.
- [ ] Pause/ECN counters checked under load; no PFC storms.

## Verbs Data Path
Read the full skill on GitHub

Details

Skill name
rdma-roce
Path
skills/hardware-acceleration/rdma-roce/SKILL.md
License
MIT
Repo updated
17 November 2026

FAQ

What does the Rdma Roce skill do?

Use when the user asks about RDMA, RoCEv2, InfiniBand, GPUDirect RDMA, ibverbs, PFC, ECN/DCQCN, lossless Ethernet or NCCL falling back to NET/Socket. Produces a fabric checklist, validation commands (perftest, nccl-tests) and NCCL env settings. Not for choosing DDP/FSDP parallelism (use distributed-training).

How do I install the Rdma Roce skill?

skills CLI (Claude Code, Cursor, Codex and more): npx skills add j4flmao/agent-skills --skill rdma-roce. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is Rdma Roce popular?

It lives in j4flmao/agent-skills, which has 45 GitHub stars and 6 forks.

Related skills

Agent Observability

45

j4flmao/agent-skills

Use when the user asks about agent observability, tracing agent reasoning, debugging an agent loop, OpenTelemetry GenAI spans for LLM calls, LLM cost tracking per agent, agent audit logs or hallucination/eval monitoring in production. Produces instrumentation config, trace schema, dashboards, alerts and cost attribution. Not for generic service tracing without LLMs (use devops-opentelemetry).

DevOps & Cloud

AI Agent Architectures

45

j4flmao/agent-skills

Use when the user asks to build an AI agent, design an agent that calls tools and keeps memory, write an agent system prompt or pick an agent framework (Claude Agent SDK, OpenAI Agents SDK, LangGraph). Produces an architecture brief covering loop, tools, memory, guardrails and evals, routing depth to child skills. Not for workflow patterns (use ai-agentic-workflows).

AI, LLMs & Agents

AI Agent Core Loop

45

j4flmao/agent-skills

Use when the user asks about the agent loop, ReAct vs plan-and-solve, re-planning, Reflexion, an agent stuck in a loop, max iterations or termination conditions. Produces a loop spec or patch with state, success predicate, caps and no-progress detection. Not for overall agent design (use ai-agent-architectures) or tool schemas (use tool-calling-principles).

AI, LLMs & Agents

Scala Play

45

j4flmao/agent-skills

Use when the user asks about Play Framework, Scala Play controllers, Play JSON Reads/Writes, Slick, Guice DI in Play or Pekko streams in Play 3. Produces Play 3 services with async actions and tests. Not for Scala language basics (use lang-scala).

Testing & QA

Frontend Testing

45

j4flmao/agent-skills

Use when the user asks about frontend testing, component tests with Testing Library, Vitest or Jest, userEvent, mocking APIs with MSW, Playwright or Cypress for critical journeys, and what to test in the UI. Produces a test strategy and component/E2E tests. Not for Storybook stories (use frontend-storybook) or cross-team E2E suites (use quality-e2e-testing).

Testing & QA

Green

49

marcoemrich/agentic_coding_lab

TDD Green Phase - Implement minimal code to make the failing test pass

Testing & QA

Browse the most popular Claude skills, all testing & qa skills, or write your own with the skills generator.