skills.rest
KwongFuk
KwongFukCOMMUNITY
Explore all skills by this author →
GitHub

palmetto-cpu-request-advisor

Estimate and right-size Palmetto CPU requests for job submissions.

Updated Apr 10, 2026
Search Tags:#throughput#hpc#sbatch#palmetto#cpu-sizing#cpu-requests#jobstats
One-click install
npx skills add https://github.com/KwongFuk/codex-skills --skill palmetto-cpu-request-advisor
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: palmetto-cpu-request-advisor
Source: https://github.com/KwongFuk/codex-skills/tree/main/workspace/palmetto-cpu-request-advisor
Command: npx skills add https://github.com/KwongFuk/codex-skills --skill palmetto-cpu-request-advisor

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Manual guesswork for CPU sizing wastes resources, increases queue times, and can over- or under-provision workloads. The Palmetto CPU Request Advisor helps teams estimate and right-size CPU requests before submission to improve utilization and throughput.

Core Features & Use Cases

  • Estimates starting CPU allocations for CPU-only serial workloads, threaded workloads, and GPU or dataloader-heavy jobs.
  • Interprets prior evidence from jobstats to adjust requests and guide retuning decisions.
  • Provides conservative defaults and clear guidance on when to scale up or down for efficient resource use.

Quick Start

Analyze your job details or run the advisor with your workload profile to receive a recommended CPU configuration.

Frequently Asked Questions about palmetto-cpu-request-advisor

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I right-size CPU requests for Palmetto HPC jobs?▼

Right-sizing CPU requests for Palmetto HPC jobs involves analyzing prior jobstats evidence and workload shape to determine safe, performance-aligned CPU allocations instead of relying on manual guesswork.

Why does manual CPU sizing waste Palmetto cluster resources?▼

Manual CPU sizing wastes resources because guesswork over-provisions or under-provisions workloads, which increases queue times and degrades overall Palmetto cluster throughput.

Can I use jobstats to estimate CPU allocations for GPU training runs?▼

Yes, you can use jobstats evidence to estimate starting CPU allocations for GPU training runs and dataloader-heavy tasks, applying conservative defaults to guide efficient retuning decisions.

Does the Palmetto CPU Request Advisor support threaded workloads?▼

Yes, the advisor supports CPU-only serial workloads, threaded workloads, and dataloader-heavy tasks, providing clear guidance on when to scale CPU allocations up or down.

What is the best way to determine sbatch CPU requests before submitting a job?▼

The best way to determine sbatch CPU requests is analyzing workload profiles and jobstats data beforehand to produce conservative, performance-aligned CPU configurations that improve utilization.

Related Skills

Explore More Software Engineering Skills→
ririntoririnto

java-performance-concurrency

Classify Java performance bottlenecks using profiling evidence and concurrency analysis.

Community
Advanced
KwongFukKwongFuk

palmetto-slurm-fit-check

Evaluate SBATCH resource requests against Palmetto node availability.

Community
Advanced
KwongFukKwongFuk

palmetto-slurm-workflow

Validate Palmetto Slurm job submissions with cluster guardrails and execution modes.

Community
Advanced
julianobarbosajulianobarbosa
7

researching-web

Search the web via Perplexity AI and return citation-rich results.

Community
Intermediate