ACAM: Application Aware Adaptive Cache Management for Shared LLC

Mahto, Sujit Kr; Newton,

doi:10.1007/978-981-10-7470-7_32

Cited by 1 publication

(1 citation statement)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Frequency based LRU Eviction based on frequency of cache access Natarajan et al (2013) Sharing-aware policy Reuse of shared LLC blocks by crossthreads Warrier et al (2013) Application-aware replacement policy Depriving high access rate application to override low access rate application Li et al (2014) Reference affinity scheme Allocating frequently used data closely Lathigara et al (2015) Application behaviour-aware replacement policy (ABRip) Predictions of re-reference interval of the block based on application behaviour and block locality Mahto et al (2017) AB-aware adaptive cache management policy (ACAM) Eviction based on application behaviour towards cache access Thread level parallelism aware partitioning (TAP) Eviction based on UCP and RRIP Mekkat et al (2013) LLC management policy GPUs tolerance towards the latency in accessing memory The research community has proposed several techniques to address main memory contention which we discuss next.…”

Section: Replacement Policy Referencementioning

confidence: 99%

A review on shared resource contention in multicores and its mitigating techniques

Jain¹,

Surve²

2020

IJHPSA

View full text Add to dashboard Cite

Chip multiprocessor (CMP) systems have become inevitable to meet high computing demands. In such systems sharing of resources is imperative for better resource utilisation. The challenge arises when various application programs running on neighbouring cores compete for these resources concurrently and introduce contention. We aim to present in a simple, lucid and captivating manner a review of previous work on contention in multicores due to various shared resources like shared caches, main memory, memory bus bandwidth, prefetchers etc. The work investigates key ideas proposed by the research community to alleviate resource contention due to these various resources, under a single umbrella. The prime objective of the study is to throw light upon the fact that, alone a single shared component is not a dominant reason for performance degradation in CMPs, rather all elements in the memory hierarchy introduce resource contention thereby affecting performance cumulatively. The work presented would assist novice readers, researchers and academicians to further serve to propose optimal policies to address contention in designing multicore applications, considering the overall impact of these resources on the performance of multicore systems.

show abstract