LSTM-CRP: Algorithm-Hardware Co-Design and Implementation of Cache Replacement Policy Using Long Short-Term Memory

隐藏物 计算机科学 期限(时间) 缓存算法 并行计算 计算机硬件 CPU缓存 算法 嵌入式系统 量子力学 物理
作者
Yizhou Wang,Yishuo Meng,Jiaxing Wang,Chen Yang
出处
期刊:Big data and cognitive computing [Multidisciplinary Digital Publishing Institute]
卷期号:8 (10): 140-140 被引量:1
标识
DOI:10.3390/bdcc8100140
摘要

As deep learning has produced dramatic breakthroughs in many areas, it has motivated emerging studies on the combination between neural networks and cache replacement algorithms. However, deep learning is a poor fit for performing cache replacement in hardware implementation because its neural network models are impractically large and slow. Many studies have tried to use the guidance of the Belady algorithm to speed up the prediction of cache replacement. But it is still impractical to accurately predict the characteristics of future access addresses, introducing inaccuracy in the discrimination of complex access patterns. Therefore, this paper presents the LSTM-CRP algorithm as well as its efficient hardware implementation, which employs the long short-term memory (LSTM) for access pattern identification at run-time to guide cache replacement algorithm. LSTM-CRP first converts the address into a novel key according to the frequency of the access address and a virtual capacity of the cache, which has the advantages of low information redundancy and high timeliness. Using the key as the inputs of four offline-trained LSTM network-based predictors, LSTM-CRP can accurately classify different access patterns and identify current cache characteristics in a timely manner via an online set dueling mechanism on sampling caches. For efficient implementation, heterogeneous lightweight LSTM networks are dedicatedly constructed in LSTM-CRP to lower hardware overhead and inference delay. The experimental results show that LSTM-CRP was able to averagely improve the cache hit rate by 20.10%, 15.35%, 12.11% and 8.49% compared with LRU, RRIP, Hawkeye and Glider, respectively. Implemented on Xilinx XCVU9P FPGA at the cost of 15,973 LUTs and 1610 FF registers, LSTM-CRP was running at a 200 MHz frequency with 2.74 W power consumption.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
刚刚
徐院长发布了新的文献求助10
3秒前
华仔的应助被霖29采纳,获得10
3秒前
洁净的天德完成签到,获得积分10
4秒前
马佳闯完成签到 ,获得积分10
7秒前
future完成签到 ,获得积分10
7秒前
7秒前
152455完成签到 ,获得积分10
8秒前
蝉鸣一夏完成签到,获得积分10
8秒前
流觞曲水完成签到 ,获得积分10
10秒前
破事不糊墙完成签到,获得积分10
10秒前
犹豫大树完成签到,获得积分10
10秒前
XL完成签到,获得积分10
12秒前
guac完成签到,获得积分10
13秒前
007完成签到 ,获得积分10
14秒前
友好的晓曼完成签到,获得积分10
16秒前
清风明月完成签到,获得积分10
16秒前
sky完成签到,获得积分10
17秒前
2580852qwe完成签到 ,获得积分10
17秒前
我看看sci怎么个事完成签到 ,获得积分10
19秒前
keyIncy完成签到,获得积分10
20秒前
徐妮完成签到,获得积分10
21秒前
22秒前
暗心不语完成签到,获得积分10
22秒前
lianxin完成签到 ,获得积分10
22秒前
23秒前
25秒前
rayqiang完成签到,获得积分0
25秒前
一笑没的应助被科研通管家采纳,获得10
26秒前
rayq完成签到,获得积分10
26秒前
一笑没的应助被科研通管家采纳,获得10
26秒前
syvshc的应助被科研通管家采纳,获得50
26秒前
26秒前
大个的应助被科研通管家采纳,获得10
26秒前
zxy的应助被科研通管家采纳,获得10
26秒前
ryz99完成签到,获得积分10
27秒前
霖29发布了新的文献求助10
27秒前
Shaelyn_Cheung完成签到 ,获得积分10
28秒前
sheep发布了新的文献求助10
29秒前
小五完成签到 ,获得积分10
30秒前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
拉伸高分子熔体中缠结行为的模拟和小角中子散射研究 1700
The World Atlas of Wine 9th edition 1000
奈特基础组织学:配套组织病理学(第4版)(Netter's Essential Histology With Correlated Histopathology) 800
八旗制度の研究 600
高分子基功能界面水蒸发材料与应用研究 500
外场增强的水汽发电研究 500
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 计算机科学 工程类 有机化学 化学工程 生物化学 物理 复合材料 内科学 催化作用 光电子学 人工智能 细胞生物学 无机化学 物理化学 数学
热门帖子
关注 科研通微信公众号,转发送积分 8185571
求助须知:如何正确求助?哪些是违规求助? 9574290
关于积分的说明 21834928
捐赠科研通 7783274
什么是DOI,文献DOI怎么找? 3434577
关于科研通互助平台的介绍 2558973
邀请新用户注册赠送积分活动 2481370