A Layer-Adaptable Cache Hierarchy by a Multiple-layer Bypass Mechanism

Egawa, Ryusuke; Saito, Ryoma; Sato, Masayuki; Kobayashi, Hiroaki

doi:10.1145/3337801.3337820

Cited by 6 publications

(3 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Over the years, the industry has adopted large shared cache memories to allow several applications to share hardware resources and run concurrently in a multi-task environment. Consequently, the cache memory hierarchy consumes a large portion of the total area and energy budget [Egawa et al 2019]. However, due to the heterogeneous nature of the applications, such memories are not guaranteed to be used efficiently.…”

Section: Introductionmentioning

confidence: 99%

“…Generally, these applications use a large portion of the shared resources. Consequently, identifying and avoiding noisy neighbors can improve general system performance by reducing conflicts and decreasing cache pollution [Egawa et al 2019].…”

Section: Introductionmentioning

confidence: 99%

“…Therefore, extensive studies try to identify the appropriated cache configuration [Kim et al 2019, Egawa et al 2019, using technologies such as Gated-Vdd [Powell et al 2000] and Cache Allocation Technology [Kim et al 2019], aiming to get higher energy efficiency and performance for these applications. However, adapting the LLC to individual applications remains a challenge.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

DyCa: Dynamically Adaptable Cache Bypassing Mechanism

Carmin¹,

Santos²,

Alves³

2022

Anais Do XXIII Simpósio Em Sistemas Computacionais De Alto Desempenho (SSCAD 2022)

View full text Add to dashboard Cite

As the number of cores increases, more cores and threads share the Last-Level Cache (LLC), which consumes a large portion of the chip’s total power and area. Therefore, sophisticated solutions must guarantee the best resource usage addressing cache conflicts and cache pollution problems. This work exploits the knowledge that many applications present poor temporal and spatial locality. Thus, an adaptive cache mechanism can benefit such applications, improving general system performance and decreasing energy consumption. In this paper, we propose an online and application-aware predictor to adapt the use of LLC. As a result, DyCa shows up to 22% and 21% performance increases in single and multi-program workloads, respectively.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%