Pith. sign in

REVIEW 5 cited by

MIGPerf: A Comprehensive Benchmark for Deep Learning Training and Inference Workloads on Multi-Instance GPUs

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2301.00407 v1 pith:PPCG4ROK submitted 2023-01-01 cs.LG cs.PF

Huaizheng Zhang , Yuanming Li , Wencong Xiao , Yizheng Huang , Xing Di , Jianxiong Yin , Simon See , Yong Luo
show 2 more authors
Chiew Tong Lau Yang You
This is my paper
classification cs.LGcs.PF
keywords inferencetrainingbenchmarkdeeplearningmigperfresultsworkloads
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

New architecture GPUs like A100 are now equipped with multi-instance GPU (MIG) technology, which allows the GPU to be partitioned into multiple small, isolated instances. This technology provides more flexibility for users to support both deep learning training and inference workloads, but efficiently utilizing it can still be challenging. The vision of this paper is to provide a more comprehensive and practical benchmark study for MIG in order to eliminate the need for tedious manual benchmarking and tuning efforts. To achieve this vision, the paper presents MIGPerf, an open-source tool that streamlines the benchmark study for MIG. Using MIGPerf, the authors conduct a series of experiments, including deep learning training and inference characterization on MIG, GPU sharing characterization, and framework compatibility with MIG. The results of these experiments provide new insights and guidance for users to effectively employ MIG, and lay the foundation for further research on the orchestration of hybrid training and inference workloads on MIGs. The code and results are released on https://github.com/MLSysOps/MIGProfiler. This work is still in progress and more results will be published soon.

Discussion (0). Sign in to comment.

Forward citations

Cited by 5 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Performance Isolation for Inference Processes in Edge GPU Systems

    cs.OS 2026-01 conditional novelty 5.0 of 10

    On NVIDIA GPUs, MIG partitions provide the strongest temporal isolation for concurrent inference, and Green Contexts approach that level on edge devices only when power throttling is removed.

  2. Energy-Aware Scheduling for Serverless LLM Serving on Shared GPUs

    cs.DC 2026-06 unverdicted novelty 4.0 of 10

    Festina reduces energy consumption by up to 56% for serverless LLM inference on shared GPUs while keeping TTFT/TBT SLO attainment within 2% of four state-of-the-art baselines.

  3. SMART-MIG: A Learning Framework for Scalable and Energy-Efficient GPU Scheduling

    cs.DC 2026-06 unverdicted novelty 3.0 of 10

    SMART-MIG applies MF-MARL for constant-complexity MIG repartitioning plus heuristics for scheduling, reporting 18% better energy-tardiness efficiency than static partitioning and 27% above a theoretical energy lower bound.

  4. CompPow: A Case for Component-level GPU Power Management

    cs.AR 2026-05 unverdicted novelty 3.0 of 10

    CompPow makes the case that component-aware power management inside GPUs can yield 10% higher energy efficiency and 5% better performance for ML workloads.

  5. A comprehensive evaluation of spatial co-execution on GPUs using MPS and MIG technologies

    cs.DC 2026-04 unverdicted novelty 3.0 of 10

    MPS can boost performance up to 30% and cut energy 20% with careful provisioning but degrades sharply under memory contention, whereas MIG delivers steadier gains through hardware isolation at the cost of higher overh...

Pith tools