doca-bench

doca-bench is a Runtime diagnostics & profiling skill for AI agents, published by NVIDIA in the skills repository on GitHub. Skillspace rates doca-bench D (69/100) and it passed the safety scan; the rating is based on static analysis only.

Use this skill when the user is running `doca_bench` (DOCA ≥ 2.7.0) — the cross-library micro-benchmark harness — to measure throughput, bulk-latency, precision-latency, or max-bandwidth (the four `benchmark_mode` enum values the binary ships) of a DOCA library (RDMA, COMPRESS, AES-GCM, SHA, DMA, EC, ETH, Comch, GPUNetIO) on host or BlueField Arm, probe the granular- build query for which libraries the install exposes, capture a baseline four-tuple (command, version, device, environment), or diagnose a bench failure in the config-syntax, device-binding, library/workload-precondition, or measurement-soundness layer. Trigger even without the word "doca_bench" — typical implicit forms include "how fast is compress on my BlueField", "what RDMA throughput can this NIC do", "measure AES-GCM latency", or "baseline before a firmware update". Refuse and route elsewhere for application-level end-to-end timing, custom benchmark programs, DOCA install/upgrade, or patching the binary.

Grade

D

Quality

69 / 100

Safety

Passed

GitHub stars

2,517

Unproven: rated by static analysis only, not yet compared head-to-head.

Our two independent classifiers differ on this skill's category, so treat the placement as provisional.

This rating describes the version of the repository we crawled; it may have changed since, so go by its current files.

Skillspace indexes and rates this skill; it does not host or endorse it. Run third-party skills at your own risk.