Budget Amount *help |
¥4,680,000 (Direct Cost: ¥3,600,000、Indirect Cost: ¥1,080,000)
Fiscal Year 2016: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Fiscal Year 2015: ¥780,000 (Direct Cost: ¥600,000、Indirect Cost: ¥180,000)
Fiscal Year 2014: ¥2,730,000 (Direct Cost: ¥2,100,000、Indirect Cost: ¥630,000)
|
Outline of Final Research Achievements |
Higher and higher performance computers are still required. Unfortunately, the improvement in memory performance is delayed behind that in CPU performance. Thus, the system performance required by IoT and deep learning has not been satisfied. Considering the background, the goal of this study is to realize high-performance and low-power memory systems. A target application is a NoSQL database system that utilizes Key-Value Store. We propose a low-power IoT device that has a flash-based main memory. We evaluate it by implementing Redis database on its main memory and by using YCSB benchmark suite. From simulations, we found the power reduction of 20%.
|