A resource utilization based instruction fetch policy for SMT processors

Weng, Lichen; Liu, Chen

doi:10.1016/j.micpro.2014.10.001

Cited by 2 publications

(1 citation statement)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In another work from the literature, Eyerman and Eeckhout proposed a method that estimates the execution time of threads in SMT architecture if threads are run alone [12]. Weng and Liu provided higher fetching priority to threads with fewer utilized resources by examining early stages of the pipeline as well as lowlevel data cache misses [13]. Zhang and Lin limited the number of entries each thread can have in the issue queue according to the previous allocation's impact on performance [14].…”

Section: Heuristics For Smt Resource Managementmentioning

confidence: 99%

Hyperheuristics for explicit resource partitioning in simultaneous multithreadedprocessors

Güney¹,

Poyraz²,

Küçük³

et al. 2020

Turk J Elec Eng & Comp Sci

View full text Add to dashboard Cite

In simultaneous multithreaded (SMT) processors, various data path resources are concurrently shared by many threads. A few heuristic approaches that explicitly distribute those resources among threads with the goal of improved overall performance have already been proposed. A selection hyperheuristic is a high-level search methodology that mixes a predetermined set of heuristics in an iterative framework to utilize their strengths for solving a given problem instance. In this study, we propose a set of selection hyperheuristics for selecting and executing the heuristic with the best performance at a given stage. To the best of our knowledge, this is one of the first studies implementing a hyperheuristic algorithm on hardware. The results of our experimental study show that hyperheuristics are indeed capable of improving the performance of the studied workloads. Our best performing hyperheuristic achieves better throughput than both baseline heuristics in 5 out of 12 workloads and gives about 15% peak performance gain. The average performance gains over the well-known hill-climbing and adaptive resource partitioning heuristics are about 5% and 2%, respectively.

show abstract

Section: Heuristics For Smt Resource Managementmentioning

confidence: 99%