Many data-intensive applications need large memory to boost system performance. The expansion of DRAM is restricted by its high power consumption and price per bit. Flash as an existing technology of Non-Volatile Memory (NVM) can make up for the drawbacks of DRAM. In this paper, we propose a hybrid main memory architecture named SSDRAM that expands RAM with flash-based SSD. SSDRAM implements a runtime library to provide several transparent interfaces for applications. Unlike using SSD as system swap device which manages data at a page level, SSDRAM works at an application object granularity to boost the efficiency of accessing data on SSD. It provides a flexible memory partition and multi-mapping strategy to manage the physical memory by micro-pages. Experimental results with a number of data-intensive workloads show that SSDRAM can provide up to 3.3 times performance improvement over SSD-swap.
Liyu WANG
Institute of Microelectronics of Chinese Academy of Sciences
Qiang WANG
Institute of Microelectronics of Chinese Academy of Sciences
Lan CHEN
Institute of Microelectronics of Chinese Academy of Sciences
Xiaoran HAO
Institute of Microelectronics of Chinese Academy of Sciences
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Liyu WANG, Qiang WANG, Lan CHEN, Xiaoran HAO, "Fine-Grained Data Management for DRAM/SSD Hybrid Main Memory Architecture" in IEICE TRANSACTIONS on Information,
vol. E99-D, no. 12, pp. 3172-3176, December 2016, doi: 10.1587/transinf.2016EDL8105.
Abstract: Many data-intensive applications need large memory to boost system performance. The expansion of DRAM is restricted by its high power consumption and price per bit. Flash as an existing technology of Non-Volatile Memory (NVM) can make up for the drawbacks of DRAM. In this paper, we propose a hybrid main memory architecture named SSDRAM that expands RAM with flash-based SSD. SSDRAM implements a runtime library to provide several transparent interfaces for applications. Unlike using SSD as system swap device which manages data at a page level, SSDRAM works at an application object granularity to boost the efficiency of accessing data on SSD. It provides a flexible memory partition and multi-mapping strategy to manage the physical memory by micro-pages. Experimental results with a number of data-intensive workloads show that SSDRAM can provide up to 3.3 times performance improvement over SSD-swap.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2016EDL8105/_p
Copy
@ARTICLE{e99-d_12_3172,
author={Liyu WANG, Qiang WANG, Lan CHEN, Xiaoran HAO, },
journal={IEICE TRANSACTIONS on Information},
title={Fine-Grained Data Management for DRAM/SSD Hybrid Main Memory Architecture},
year={2016},
volume={E99-D},
number={12},
pages={3172-3176},
abstract={Many data-intensive applications need large memory to boost system performance. The expansion of DRAM is restricted by its high power consumption and price per bit. Flash as an existing technology of Non-Volatile Memory (NVM) can make up for the drawbacks of DRAM. In this paper, we propose a hybrid main memory architecture named SSDRAM that expands RAM with flash-based SSD. SSDRAM implements a runtime library to provide several transparent interfaces for applications. Unlike using SSD as system swap device which manages data at a page level, SSDRAM works at an application object granularity to boost the efficiency of accessing data on SSD. It provides a flexible memory partition and multi-mapping strategy to manage the physical memory by micro-pages. Experimental results with a number of data-intensive workloads show that SSDRAM can provide up to 3.3 times performance improvement over SSD-swap.},
keywords={},
doi={10.1587/transinf.2016EDL8105},
ISSN={1745-1361},
month={December},}
Copy
TY - JOUR
TI - Fine-Grained Data Management for DRAM/SSD Hybrid Main Memory Architecture
T2 - IEICE TRANSACTIONS on Information
SP - 3172
EP - 3176
AU - Liyu WANG
AU - Qiang WANG
AU - Lan CHEN
AU - Xiaoran HAO
PY - 2016
DO - 10.1587/transinf.2016EDL8105
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E99-D
IS - 12
JA - IEICE TRANSACTIONS on Information
Y1 - December 2016
AB - Many data-intensive applications need large memory to boost system performance. The expansion of DRAM is restricted by its high power consumption and price per bit. Flash as an existing technology of Non-Volatile Memory (NVM) can make up for the drawbacks of DRAM. In this paper, we propose a hybrid main memory architecture named SSDRAM that expands RAM with flash-based SSD. SSDRAM implements a runtime library to provide several transparent interfaces for applications. Unlike using SSD as system swap device which manages data at a page level, SSDRAM works at an application object granularity to boost the efficiency of accessing data on SSD. It provides a flexible memory partition and multi-mapping strategy to manage the physical memory by micro-pages. Experimental results with a number of data-intensive workloads show that SSDRAM can provide up to 3.3 times performance improvement over SSD-swap.
ER -