Significant advances have recently been made in various forms of solid-state memory such as NAND flash, magnetic RAM (MRAM), phase-change memory (PRAM), and FRAM. Solid state memory offers several advantages over hard disks: lower access latencies for random requests, smaller form factors, lower power consumption, lack of noise, and higher robustness to vibrations and temperature. In particular, recent improvements in the design and performance of NAND flash memory has resulted in it being employed in many embedded and consumer devices. Small form-factor hard disks have already been replaced by flash memory in some consumer devices like music players. More recently, flash drives with capacities in the 128-256GB range have become available and have been used in certain laptops as the secondary storage media. The cost-per-byte for a hard disk remains an order of magnitude lower than for flash memory and disks are likely to maintain this advantage over at least the next decade. Therefore, completely replacing hard disks with flash devices in enterprise-scale storage systems is likely to remain infeasible for the foreseeable future. At the same time, however, flash devices are significantly cheaper than main memory technologies that play a crucial role in improving the performance of disk-based systems via caching and buffering. Despite the higher cost compared to hard disks, the above mentioned advantages of flash memory naturally lead to following question: can flash memory, within the constraints imposed by cost issues, be used with hard disks to design improved enterprise storage?
In this project, we propose to study the utility and feasibility of hybrid enterprise storage systems that employ flash memory in conjunction with hard disk drives. Our study will go beyond exploring the use of flash as a mere cache for the disk as has been done in most existing research. We will investigate whether flash memory should really be treated as a storage medium that is complementary to the hard disk as opposed to merely being subordinate to it.
Aayush Gupta (PhD student)
Youngjae Kim (Ph.D. 2009, first employer: Oak Ridge National Lab)
Di Wang (PhD student)
Copyright notice: This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author’s copyright. These works may not be reposted without the explicit permission of the copyright holder.
A Case for Heterogeneous Flash
Di Wang, Anand Sivasubramaniam, Bhuvan Urgaonkar
Technical Report, Dept. of CSE, The Pennsylvania State University, 2011.
HybridStore: A Cost-Efficient, High-Performance Storage System Combining SSDs and HDDs
Youngjae Kim, Aayush Gupta, Bhuvan Urgaonkar, Piotr Berman and Anand Sivasubramaniam
In Proceedings of the IEEE International Symposium on Modeling, Analysis and Simulation of Computer and Telecommunication Systems (MASCOTS), Singapore, July 2011.
Technical Report (CSE-08-017), Dept. of Computer Science and Engineering, The Pennsylvania State University 2008.
Abstract Paper Tech. Report
A Comprehensive Study on Energy Efficiency of Flash Memory Storages
Seonyeong Park, Youngjae Kim, Bhuvan Urgaonkar, Joonwon Lee, and Euiseong Seo
Elsevier Journal of System Architecture (JSA), Volume 57, Issue 4, Pages 354-365, April 2011. Abstract Paper
Leveraging Value Locality in Optimizing NAND Flash-based SSDs
Aayush Gupta, Raghav Pisolkar, Bhuvan Urgaonkar, and Anand Sivasubramaniam
In Proceedings of the USENIX Conference on File and Storage Technologies (FAST), San Jose, CA, February 2011. (20/74 = 27% accepted). Abstract Paper
FlashSim: A Simulator for NAND Flash-based Solid-State Drives
Youngjae Kim, Brendan Taurus, Aayush Gupta, and Bhuvan Urgaonkar
Proceedings of the First International Conference on Advances in System Simulation (SIMUL), Porto, Portugal. September 2009. Also, Technical Report CSE 09-008, Department of Computer Science and Engineering, The Pennsylvania State University, May 2009.
Abstract Tech. Report
DFTL: A Flash Translation Layer Employing Demand-based Selec tive Caching of Page-level Address Mappings
Aayush Gupta, Youngjae Kim, and Bhuvan Urgaonkar
Proceedings of the Fourteenth International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS), Washington DC, March 2009. (29/113 = 25% accepted). Also, Technical Report CSE 08-012, Department of Computer Science and Engineering, The Pennsylvania State University, September 2008.
Abstract Paper Tech. Report
Empirical Analysis of the Energy Efficiency of Flash-based SSDs
Euiseong Seo, Seonyeong Park, Bhuvan Urgaonkar
Proceedings of the First Workshop on Power Aware Computing and Systems (HotPower), held in conjunction with USENIX OSDI 2008, San Diego, CA, December 2008.
Flash based SSD Simulator
Objected Oriented Flash Simulator – an objected oriented SSD simulator
Flash Translation Layer – a simulator for the study of various FTL schemes
National Science Foundation: CCF-0811670