Ghost-Cache CRFP Algorithm Simulator Based on Trace and File System

Abstract

Solid State Drive (SSD) is fast media that has become popular recently. Because it has a smaller size, it acts on larger and slower media as a cache. There are researches on SSD as a cache on Hard Disk (HDD) media. This paper present GCRFP algorithm and test its performance using two simulators. The first simulator is implemented by using trace input, the second using user-space file-system. The purpose of these simulators is to study the algorithm performance behavior. The findings demonstrate that GCRFP gives LARC comparable performance in both simulators.

High Reliability Through Cache Failure Minimization Technique for Workload Execution Under Multi-Core Edge-Cloud Platforms

“…∑ 𝐷 𝑘𝑝 ′ 𝑞 𝑘𝑝 𝑎 𝑘 (𝑡) + ∑ 𝐻 𝑘𝑝 𝑞 𝑘𝑝 𝑝 𝑝 (9) where ℎ 𝑘 (𝑡) represents the approximated processing-element resource utilization and 𝑞 𝑘𝑝 represents the job-rate within the operational time period 𝑀 𝑘𝑝 for the heterogeneous computing system 𝐵 𝑘 . where ℎ 𝑘 (𝑡) represents the approximated processing-element resource utilization and 𝑞 𝑘𝑝 represents the job-rate.…”

“…Comparatively, edge servers can only store a little amount of data, but they contain files that are accessed often. Last but not least, local storage takes up the least space yet is where you'll find frequently accessed data and files [9]. As edge caching provides better results and there is a probability that there would be some failure during the retrieval of the information from the multi-core edge cloud.…”