Categories › Data & Databases › Vector databases & embeddings
GPU fabric tools: NCCL benchmark analysis, degradation cost, cluster compatibility, XportL facts.
Remote server: your requests and data go to xportl.io. Claude Code: claude mcp add --transport http xportl https://xportl.io/mcp · Claude app: Settings → Connectors → Add custom connector.
The tools this server listed when we connected to it, in its own words.
get_xportl_overview — Returns cite-ready facts about XportL from xportl.io: what it does, the problem it solves, the seven failure classes it detects, how it collanalyze_nccl_benchmark — Analyzes nccl-tests / rccl-tests results (paste raw all_reduce_perf output, or give size/busbw points). Detects node and GPU counts from thecalculate_fabric_degradation_cost — Estimates what silent fabric degradation (slow NCCL collectives, flapping links, congestion) costs a GPU cluster: throughput lost, GPU-hourscheck_cluster_compatibility — Checks a GPU cluster against XportL's published support matrix (Linux 5.8+ for eBPF, InfiniBand / RoCE v2 / NVLink, NVIDIA and AMD GPUs, NCCrecommend_checkpoint_interval — Recommends how often to checkpoint a training job using the Young and Daly formulas, from GPU count, blocking checkpoint write time, restartrequest_enterprise_demo — Sends the user's contact details and cluster summary to the XportL enterprise team to arrange a technical assessment or demo deployment. BefNot tested by KitScout yet. When our harness runs it, pros, cons and the tasks we tried appear here, with the date and evidence.