MétaCan
Menu
Back to cohort
Record W3189640247 · doi:10.1109/tcc.2021.3102593

Accelerating Distributed Learning in Non-Dedicated Environments

2021· article· en· W3189640247 on OpenAlex
Chen Chen, Qizhen Weng, Wei Wang, Baochun Li, Bo Li

Why this work is in the frame

A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.

affAt least one author lists a Canadian institution in the pinned OpenAlex snapshot.

Bibliographic record

VenueIEEE Transactions on Cloud Computing · 2021
Typearticle
Languageen
FieldComputer Science
TopicPrivacy-Preserving Technologies in Data
Canadian institutionsUniversity of Toronto
Fundersnot available
KeywordsComputer scienceLoad balancing (electrical power)Cloud computingExecution timeArtificial intelligenceSoftware deploymentPython (programming language)Key (lock)Distributed computingMachine learningOperating system

Abstract

fetched live from OpenAlex

Machine learning (ML) models are increasingly trained with distributed workers possessing heterogeneous resources. In such scenarios, model training efficiency may be negatively affected by <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">stragglers</i> —workers that run much slower than others. Efficient model training requires eliminating such stragglers, yet for modern ML workloads, existing load balancing strategies are inefficient and even infeasible. In this article, we propose a novel strategy, called <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">semi-dynamic load balancing</i> , to eliminate stragglers of distributed ML workloads. The key insight is that ML workers shall be load-balanced at <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">iteration boundaries</i> , being non-intrusive to intra-iteration execution. Based on it we further develop LB-BSP, an integrated worker coordination mechanism that adapts workers’ load to their instantaneous processing capabilities—by right-sizing the sample batches at the synchronization barriers. We have designed distinct load tuning algorithms for ML in CPU clusters, in GPU clusters as well as in federated learning setups, based on their respective characteristics. LB-BSP has been implemented as a Python module for ML frameworks like TensorFlow and PyTorch. Our EC2 deployment confirms that LB-BSP is practical, effective and light-weight, and is able to accelerating distributed training by up to 54 percent.

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.

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)0.001
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesMeta-epidemiology (narrow), Open science
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: none
GenreCandidate signal: Empirical · Consensus signal: none
Teacher disagreement score0.753
Threshold uncertainty score1.000

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0000.001
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0000.001
Science and technology studies0.0000.000
Scholarly communication0.0000.000
Open science0.0060.001
Research integrity0.0000.001
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 head0.028
GPT teacher head0.264
Teacher spread0.236 · 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