Cache memory structural design for handling big data usages

A cache is a hardware or software component in computing that stores data in order to serve future requests for that data more quickly; the data stored in a cache could be the result of an earlier computation as well as any copy of data stored elsewhere. The efficient execution of today's big data-based applications relies heavily on cache memory. Multicore processors are being used in the high-performance computer to support parallel execution of multiple applications and threads. Chip multiprocessors combine multiple multicore processors into a single chip (CMP). Each core has its own private cache memories, as well as the last-level cache is shared by all cores (LLC). When it comes to handling big data-based applications, LLC's performance is critical. This paper conducted a survey of the novel techniques proposed for efficiently handling big data-based applications in the LLC of CMPs in this paper.