MétaCan
Menu
Back to cohort
Record W7126394018 · doi:10.21428/594757db.552e5df1

GPU Job Scheduler with Deep Reinforcement Learning

2024· article· en· W7126394018 on OpenAlexaff
Junjie Deng, Aijun An, Hajer Ayadi, Yiming Shao, Hossein Pourmodheji, Hao Zhou, Michael Feiman

Bibliographic record

Venuenot available
Typearticle
Languageen
FieldComputer Science
TopicParallel Computing and Optimization Techniques
Canadian institutionsIBM (Canada)York University
Fundersnot available
KeywordsReinforcement learningScheduling (production processes)HeuristicArtificial neural networkDeep learningGPU clusterProcess (computing)Speedup

Abstract

fetched live from OpenAlex

We present Proximity-DRL, a novel method for GPU job assignment in a GPU cluster using Deep Reinforcement Learning (DRL). Proximity-DRL considers the spatial relationships of GPUs during GPU allocation for each job, which effectively reduces the communication cost among the GPUs assigned to the job. In addition, Proximity-DRL uses a sequential decision-making process to allocate GPUs for a job, which effectively reduces the action space of the DRL model and leads to better performance than DRL models that select multiple GPUs in a single decision-making step. Trained using the Proximal Policy Optimization (PPO) algorithm to facilitate multi-GPU selection, Proximity-DRL demonstrates superior performance in optimizing average Job Completion Time (JCT), makespan, and communication costs compared to several popular heuristic schedulers and a DRL-based scheduler, highlighting its potential to speed up distributed training of deep neural network models.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame distilled prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.

metaresearch head score (Codex)0.000
metaresearch head score (Gemma)0.000
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: Simulation or modeling
GenreCandidate signal: Methods · Consensus signal: none
Teacher disagreement score0.607
Threshold uncertainty score0.322

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0000.000
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0000.000
Science and technology studies0.0000.000
Scholarly communication0.0000.000
Open science0.0000.000
Research integrity0.0000.000
Insufficient payload (model declined to judge)0.0000.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.012
GPT teacher head0.249
Teacher spread0.237 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one teacher head, not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designSimulation or modeling
Domainnot available
GenreMethods

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations0
Published2024
Admission routes1
Has abstractyes

Explore more

Same topicParallel Computing and Optimization TechniquesFrench-language works237,207