HPC system uses AI classifier to find file system slowdowns
IOSage: Detecting HPC I/O Bottleneck Patterns to Guide LLM Recommendations
Distributed, Parallel, and Cluster ComputingMachine LearningPerformance
Summary
High-performance computing (HPC) applications often don't use all the available file system speed because it's hard to spot and fix the problems. The authors created IOSage, a tool that first uses a trained classifier to detect specific input/output bottleneck patterns from logs. Then, an AI model gets these patterns and extra info to suggest improvements. This approach works better than having the AI analyze logs alone, speeding up runs by up to over four times in their tests.
What this means in practice
- •For hpc system administrators: Identify and fix file system bottlenecks more accurately using IOSage's AI-assisted pattern detection and recommendations.
- •For cloud infrastructure teams: Optimize storage resource use in cloud HPC platforms by applying IOSage's method to analyze I/O logs and suggest configuration changes.
Authors
Mahdi Banisharif, Tabassum Mahmud, Ahmad Tarraf, Suren Byna, Ali Jannesari
Abstract
HPC applications often leave much of their parallel file system's bandwidth unused, and turning their Darshan logs into concrete I/O fixes still requires a scarce I/O expert. Rule-based tools flag counters against thresholds set by experts, and in tools built on large language models (LLMs), such as ION and IOAgent, the LLM decides which bottlenecks a log contains. We present IOSage, in which a classifier trained on many labeled logs, not the LLM, makes that decision, since a single log gives an LLM no baseline for what is abnormal. The LLM receives the I/O bottleneck patterns the classifier detects, the job's configuration files, per-file Darshan evidence, and for each pattern a card built from benchmark training runs, and writes recommended changes. Applied by hand and measured in repeated paired runs, the changes IOSage names speed up median total wall time 4.38x for an openPMD kernel (4.02x on a second storage tier) and 1.42x for hourly WRF restarts. For 5 of the 6 application scenarios with a helpful change, IOSage with Claude Sonnet 4 names it, including the change for 10-minute WRF restarts (up to 1.93x faster); with GPT-4o and Llama 3.1 as well, the classifier and cards name more helpful changes than the LLM alone. On 78 benchmark test runs without a pattern, the LLM deciding alone proposes changes for 64; with the classifier, for 9. The classifier reaches 0.929 Micro-F1 on 488 labeled benchmark test runs (0.916 as the gate; 0.832 on the 158 from configurations and jobs outside training), above Drishti's rules, WisIO, and IOAgent's pipeline without its literature retrieval. On TraceBench's nine expert-labeled traces from production systems, it has the highest precision of the tools compared (0.82), with 2 false positives against 12 for the keyword-mapped LLM diagnoses shipped with TraceBench and 5 for Drishti, on the five issue labels that match registered patterns.