Abstract
State-of-the-art high density NAND flash memory has been recommended as read intensive storage device due to their excellent read performance. However, recent studies and reports show that the read latency of high density NAND flash memory is increasing. The reason comes from at least two aspects: First, high density flash generally adopts multiple bits per cell technique, where the access latency of the most significant bits is largely increased. Second, due to the reliability variation among these bits, the access latency of the most significant bits is further increased. We introduce RLV, a read performance optimization scheme is proposed to exploit the read latency variation among the multiple bits. The basic idea is that firstly identify the hotness of read data and then move them to the places with corresponding read latency. Our evaluation shows that RLV incurs negligible overhead, while improving read performance by 14% on average compared with state-of-the-arts.
| Original language | English |
|---|---|
| Title of host publication | GLSVLSI'20 Proceedings of the 2020 Great Lakes Symposium on VLSI |
| Publisher | Association for Computing Machinery |
| Pages | 411-414 |
| ISBN (Electronic) | 9781450379441 |
| DOIs | |
| Publication status | Published - Sept 2020 |
| Event | The 30th edition of the ACM Great Lakes Symposium on VLSI (GLSVLSI 2020) : “In-Memory Processing for Future Electronics” - Virtual, Online, Beijing, China Duration: 8 Sept 2020 → 11 Sept 2020 Conference number: 30 https://www.glsvlsi.org/index.html |
Publication series
| Name | Proceedings of the ACM Great Lakes Symposium on VLSI, GLSVLSI |
|---|
Conference
| Conference | The 30th edition of the ACM Great Lakes Symposium on VLSI (GLSVLSI 2020) |
|---|---|
| Place | China |
| City | Beijing |
| Period | 8/09/20 → 11/09/20 |
| Internet address |
Research Keywords
- 3D Stacking NAND
- Read latency
- Reliability