Memory System Optimzation for Energy Efficient Big Data Processing

2017 Fiscal Year Final Research Report

• Project/Area Number: 16H06677
• Research Category: Grant-in-Aid for Research Activity Start-up
• Allocation Type: Single-year Grants
• Research Field: Computer system
• Research Institution: The University of Tokyo
• Principal Investigator: Arima Eishi 東京大学, 情報基盤センター, 特任助教 (50780699)
• Research Collaborator: Schulz Martin Lawrence Livermore National Laboratory, Computer Scientist
• Project Period (FY): 2016-08-26 – 2018-03-31
• Keywords: メモリシステム / ビッグデータ / 高電力効率化 / キャッシュ / ストレージクラスメモリ

Outline of Final Research Achievements

To improve the energy efficiency of big data processing, this work focused on optimizing memory systems, the major performance/power bottlenecks during executing big data applications, by hardware/software-side approaches. In particular, this work is based on two novel approaches: (1) address translation aware cache management and (2) storage class memory aware power management. Consequently, it is quantified that a few tens of percent of energy efficiency improvement can be achieved by applying those methods.

Free Research Field

計算機システム