A MapReduce scratchpad memory for multi-core cloud computing applications

详细信息    查看全文

• 作者：Christoforos Kachris^a ; ^{ckachris@ee.duth.gr" class="auth_mail" title="E-mail the corresponding author}Author Vitae ; Georgios Ch. Sirakoulis^aAuthor Vitae ; Dimitrios Soudris^bAuthor Vitae
• 关键词：Phoenix MapReduce ; Scratchpad memory ; Multi-core ; Execution time ; Energy consumption ; Data-centers
• 刊名：Microprocessors and Microsystems
• 出版年：2015
• 出版时间：November 2015
• 年：2015
• 卷：39
• 期：8
• 页码：599-608
• 全文大小：953 K

文摘

Phoenix MapReduce is a multi-core programming framework that is used to automatically parallelize and schedule programs. This paper presents a novel scratchpad memory architecture that is used accelerate MapReduce applications by indexing and processing the key/value pairs. The proposed scratchpad memory scheme can be mapped onto programmable logic or multi-core processors chips as a coprocessor to accelerate MapReduce applications. The proposed architecture has been implemented in a Zynq FPGA with two embedded ARM cores. The performance evaluation shows that the proposed scheme can reduce up to 2.3×$\times$ the execution time and up to 1.7×$\times$ the energy consumption.