| MacroTrend: A Write-Efficient Cache Algorithm for NVM-Based Read Cache |
| |
| 作者姓名： | 鲍宁 柴云鹏 秦啸 王传雯 |
| |
| 作者单位： | Key Laboratory of Data Engineering and Knowledge Engineering;School of Information;Samuel Ginn College of Engineering |
| |
| 基金项目： | supported by the National Key Research and Development Program of China under Grant No.2019YFE0198600;the National Natural Science Foundation of China under Grant Nos.61972402,61972275,and 61732014. |
| |
| 摘 要： | The future storage systems are expected to contain a wide variety of storage media and layers due to the rapid development of NVM(non-volatile memory)techniques.For NVM-based read caches,many kinds of NVM devices cannot stand frequent data updates due to limited write endurance or high energy consumption of writing.However,traditional cache algorithms have to update cached blocks frequently because it is difficult for them to predict long-term popularity according to such limited information about data blocks,such as only a single value or a queue that reflects frequency or recency.In this paper,we propose a new MacroTrend(macroscopic trend)prediction method to discover long-term hot blocks through blocks'macro trends illustrated by their access count histograms.And then a new cache replacement algorithm is designed based on the MacroTrend prediction to greatly reduce the write amount while improving the hit ratio.We conduct extensive experiments driven by a series of real-world traces and find that compared with LRU,MacroTrend can reduce the write amounts of NVM cache devices significantly with similar hit ratios,leading to longer NVM lifetime or less energy consumption.
|
| 关 键 词： | non-volatile memory(NVM) solid state disk(SSD) CACHE ENDURANCE |
| 收稿时间： | 2019-11-13 |
MacroTrend: A Write-Efficient Cache Algorithm for NVM-Based Read Cache |
| |
| Ning Bao,Yun-Peng Chai,Xiao Qin,Chuan-Wen Wang.MacroTrend: A Write-Efficient Cache Algorithm for NVM-Based Read Cache[J].Journal of Computer Science and Technology,2022,37(1):207-230. |
| |
| Authors: | Ning Bao Yun-Peng Chai Xiao Qin Chuan-Wen Wang |
| |
| Affiliation: | 1.Key Laboratory of Data Engineering and Knowledge Engineering, Ministry of Education, Beijing 100872, China;2.School of Information, Renmin University of China, Beijing 100872, China;3.Samuel Ginn College of Engineering, Auburn University, Alabama 36830, U.S.A. |
| |
| Abstract: | The future storage systems are expected to contain a wide variety of storage media and layers due to the rapid development of NVM (non-volatile memory) techniques. For NVM-based read caches, many kinds of NVM devices cannot stand frequent data updates due to limited write endurance or high energy consumption of writing. However, traditional cache algorithms have to update cached blocks frequently because it is difficult for them to predict long-term popularity according to such limited information about data blocks, such as only a single value or a queue that reflects frequency or recency.
In this paper, we propose a new MacroTrend (macroscopic trend) prediction method to discover long-term hot blocks through blocks' macro trends illustrated by their access count histograms. And then a new cache replacement algorithm is designed based on the MacroTrend prediction to greatly reduce the write amount while improving the hit ratio. We conduct extensive experiments driven by a series of real-world traces and find that compared with LRU, MacroTrend can reduce the write amounts of NVM cache devices significantly with similar hit ratios, leading to longer NVM lifetime or less energy consumption. |
| |
| Keywords: | non-volatile memory (NVM) solid state disk (SSD) cache enduranc |
| 本文献已被 维普 等数据库收录！ |
| 点击此处可从《计算机科学技术学报》浏览原始摘要信息 |
|
点击此处可从《计算机科学技术学报》下载全文 |
