ACR: APPLICATION AWARE CACHE REPLACEMENT FOR SHARED CACHES IN MULTI-CORE SYSTEMS
Journal Title: International Journal of Computer Engineering & Technology (IJCET) - Year 2019, Vol 10, Issue 2
Abstract
Modern multi-core systems allow concurrent execution of different applications on a single chip. Such multicores handle the large bandwidth requirement from the processing cores by employing multiple levels of caches with one or two levels of private caches along with a shared last-level cache (LLC). In shared LLC, when applications with varying access behavior compete with each other for space, conventional single core cache replacement techniques can significantly degrade the system performance. In such scenarios, we need an efficient replacement policy for reducing the off-chip memory traffic as well as contention for the memory bandwidth. This paper proposes a novel Application-aware Cache Replacement (ACR) policy for the shared LLC. ACR policy considers the memory access behavior of the applications during the process of victim selection to prevent victimizing a low access rate application by a high-access rate application. \textcolor{red}{ It dynamically tracks the maximum life-time of cache lines in shared LLC for each concurrent application and helps in efficient utilization of the cache space. Experimental evaluation of ACR policy for 4-core systems, with 16-way set associative 4MB LLC, using SPEC CPU 2000 and 2006 benchmark suites shows a geometric mean speed-up of 8.7% over the least recently used (LRU) replacement policy. We show that the ACR policy performs better than recently proposed thread-aware dynamic re-reference interval prediction (TA-DRRIP) and protecting distance based (PDP) techniques for various 2-core, 4-core and 8-core workloads.
Authors and Affiliations
TRIPTI S WARRIER
Keywords
Related Articles
- EP ID EP46550
- DOI 10.34218/IJCET.10.2.2019.023
- Views 211
- Downloads 0
MACHINE LEARNING ALGORITHMS FOR HETEROGENEOUS DATA: A COMPARATIVE STUDY
In the present digital era massive amount of data is being continuously generated at exceptional and increasing scales. This data has become an important and indispensable part of every economy, industry, organization,...
A 3-LEVEL MULTIFACTOR AUTHENTICATION SCHEME FOR CLOUD COMPUTING
The objective of this paper is to propose a secure, user friendly and economical multi-level authentication scheme that uses multiple factors for gaining access to resource on insecure platforms and for financial trans...
RESOLVING PRIVACY CONFLICT FOR MAINTAINING PRIVACY POLICIES IN ONLINE SOCIAL NETWORKS
In recent years, the use of online social networks (OSNs) such as Facebook, Twitter etc has tremendous increased. Users see these OSNs as a useful tool to find friends and interact with them. Moreover, OSNs allow their...
DIABETES CLASSIFICATION AND PREDICTION USING ARTIFICIAL NEURAL NETWORK
The classification of data is an important field of data mining comes under supervised learning. In this approach classifier is trained on the pre-categorized data thereafter tested on unseen part called test data to e...
AN APPROACH FOR PREDICTION OF CROP YIELD USING MACHINE LEARNING AND BIG DATA TECHNIQUES
Agriculture is the primary source of livelihood which forms the backbone of our country. Current challenges of water shortages, uncontrolled cost due to demand-supply, and weather uncertainty necessitate farmers to be...