Name
Abstract | ||
|---|---|---|
Emerging new memory technologies are driving the development of computer storage architectures in recent years. Non-volatile main memories (NVMMs) offer higher density than SRAM, access performance and persistency like DRAM and disk respectively. It can also persist data at memory level. However, traditional file systems are block-based and designed for the disk access features, which fail to leverage the byte-addressable characteristic of NVMMs efficiently. Consequently, there have been some NVMM file systems focusing on the study of data consistency in memory level. For instance, compared with the block-based file systems, the design of NOVA optimizes the system architecture by avoiding the page cache in DRAM and the block layer, while providing strong data consistency guarantees at the same time. In this paper, we compare state-of-the-art NVMM file systems such as PMFS and NOVA with several mainstream traditional file systems. Then, we evaluate these file systems with specific workloads and analyze the experimental results to draw some insightful conclusions. As an example, NOVA can provide write performance close to PMFS, and strong data consistency as NILFS2. When I/O size comes to 4MB, its random write performance achieves 2.2 times of Ext4, 16.5 times of NILFS2 and 4.4 times of F2FS respectively. |
Year | Venue | Field |
|---|---|---|
2017 | DASC/PiCom/DataCom/CyberSciTech | Dram,File system,Computer data storage,Computer science,ext4,Static random-access memory,Page cache,Big data,Operating system,Data consistency |
DocType | Citations | PageRank |
Conference | 0 | 0.34 |
References | Authors | |
0 | 6 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Yang Li | 1 | 659 | 125.00 |
| Fang Liu | 2 | 1188 | 125.46 |
| Nong Xiao | 3 | 649 | 116.15 |
| Songping Yu | 4 | 8 | 4.89 |
| Shuo Li | 5 | 5 | 2.48 |
| Yuxuan Xing | 6 | 3 | 5.10 |