MétaCan
Menu
Back to cohort
Record W1883550131

Adaptive Thread Management For Power, Temperature, And Reliability In Future Microprocessors

2010· article· en· W1883550131 on OpenAlex
Jonathan A. Winter

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.

fundA Canadian funder is recorded on the work.
no affNo Canadian affiliation: this work is invisible to an affiliation-only frame.
No Canadian affiliation. An affiliation-only frame, the usual design, would never have seen this work. It is one of the works that make the case for inverting the frame.

Bibliographic record

VenueeCommons (Cornell University) · 2010
Typearticle
Languageen
FieldComputer Science
TopicParallel Computing and Optimization Techniques
Canadian institutionsnot available
FundersNatural Sciences and Engineering Research Council of CanadaUniversity of Toronto
KeywordsThread (computing)Reliability engineeringComputer scienceReliability (semiconductor)Embedded systemEngineeringPower (physics)Operating system
DOInot available

Abstract

fetched live from OpenAlex

With continued scaling of CMOS technology, power, thermal, and reliability issues threaten to significantly limit future performance improvements. The advent of microprocessors with multiple processing units creates a new opportunity to address these concerns through low-cost adaptive thread management techniques. In this dissertation we devise two types of dynamic management schemes, thread migration and power management, which leverage the inherent architectural characteristics of future microprocessors to dramatically mitigate thermal hotspots, variations, and hard faults. These techniques are applied both within the core, in clustered simultaneous multithreaded (SMT) architectures, and among the cores of unpredictably heterogeneous chip multiprocessors (CMPs). First, we investigate dynamic thermal management (DTM) in clustered SMT architectures. We propose novel thread migration algorithms that leverage the steering mechanism inherent in clustered architectures to cool hotspots more effectively than dynamic voltage and frequency scaling (DVFS) when executing thermally nonuniform workloads. In addition, we create a DTM mechanism that combines intelligent steering with DVFS power management to achieve efficient thermal control across all workloads. In future large-scale multi-core microprocessors, hard faults and process variations will create dynamic heterogeneity, causing performance and power characteristics to differ among the cores in an unanticipated manner. Contemporary CMP thread managers are oblivious to this heterogeneity, resulting in significant performance losses and excess power dissipation. We develop operation system scheduling and global power management policies, which significantly reduce the loss in power-performance efficiency. We further explore the scalability of these algorithms to many-core architectures with four to two-hundred fifty-six cores and devise novel, scalable runtime management techniques which achieve high performance with low overhead.

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)0.000
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Theoretical or conceptual · Consensus signal: none
GenreCandidate signal: Empirical · Consensus signal: none
Teacher disagreement score0.698
Threshold uncertainty score0.544

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 science0.0010.000
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 head0.201
Teacher spread0.189 · 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