Name
Abstract | ||
|---|---|---|
For many big data applications, non-volatile memory (NVM) can be utilized to store and process the data at faster rate due to its high-performance, scalable technology, DRAM-like interface and low energy requirement. NVMs such as phase change memory and memristor allow the applications to store persistent data directly in memory, and avoid data serialization and deserialization. However, NVM, like volatile memory, is susceptible to data corruption from software bugs. In this work, we present a paradigm shift from current process-based page-protection to a thread-based solution specifically designed for NVM. We have developed SafeNVM, a reliable NVM store to support application-specific data formats. SafeNVM will enable the NVM to provide strong data protection while delivering high performance access. We propose a simple hardware change in TLB and page table entry and exploit bound checking inherent to swizzled pointers. We show that SafeNVM is reliable against a collection of stray writes and the cost to achieve such protection is small. |
Year | DOI | Venue |
|---|---|---|
2017 | 10.1109/BigDataCongress.2017.18 | 2017 IEEE International Congress on Big Data (BigData Congress) |
Keywords | Field | DocType |
Big Data,NVM,Data Reliability | Interleaved memory,Semiconductor memory,Non-volatile random-access memory,Computer science,NVM Express,Memory management,Page fault,Memory map,Volatile memory,Operating system | Conference |
ISSN | ISBN | Citations |
2379-7703 | 978-1-5386-1997-1 | 0 |
PageRank | References | Authors |
0.34 | 19 | 2 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Pradeep Kumar | 1 | 31 | 2.33 |
| H. Howie Huang | 2 | 537 | 40.29 |