Taeho Kgil
Trevor Mudge
ABSTRACT
We propose an architecture that uses NAND flash memory to reduce main memory power in web server platforms. Our architecture uses a two level file buffer cache composed of a relatively small DRAM, which includes a primary file buffer cache, and a flash memory secondary file buffer cache. Compared to a conventional DRAM-only architecture, our architecture consumes orders of magnitude less idle power while remaining cost effective. This is a result of using flash memory, which consumes orders of magnitude less idle power than DRAM and is twice as dense. The client request behavior in web servers, allows us to show that the primary drawbacks of flash memory?endurance and long write latencies?can easily be overcome. In fact the wear-level aware management techniques that we propose are not heavily used.
- Error Correction Code in Single Level Cell NAND Flash Memories. http://www. st. com/stonline/products/literature/an/10123. pdf.Google Scholar
- Hybrid Hard Drives with Non-Volatile Flash and Longhorn. http://www. samsung. com/Products/HardDiskDrive/news/HardDiskDrive_20050425_0000117556. htm.Google Scholar
- JFFS:The Journalling Flash File System. http://sources. redhat. com/jffs2/jffs2. pdf.Google Scholar
- Micron DDR2 DRAM. http://www. micron. com/products/dram/ddr2/.Google Scholar
- The Micron system-power calculator. http://www. micron. com/products/dram/syscalc. html.Google Scholar
- Samsung NAND Flash memory datasheet. http://www. samsung. com/products/semiconductor/NANDFlash/SLC_LargeBlock/8Gbit/K9K8G08U0A/K9K8G08U0A. htm.Google Scholar
- SPECweb99 benchmark. http://www. spec. org/osg/web99/.Google Scholar
- Sun Fire T2000 Server Power Calculator. http://www. sun. com/servers/coolthreads/t2000/calc/index. jsp.Google Scholar
- TrueFFS. http://www. m-systems. com/site/en-US/Support/DeveloperZone/Software/LifespanCalc. htm.Google Scholar
- ITRS roadmap. Technical report, 2005.Google Scholar
- P. Barford and M. Crovella. Generating representative web workloads for network and server performance evaluation. In Measurement and Modeling of Computer Systems, pages 151--160, 1998. Google ScholarDigital Library
- N. L. Binkert, R. G. Dreslinski, L. R. Hsu, K. T. Lim, A. G. Saidi, and S. K. Reinhardt. The M5 simulator: Modeling networked systems. IEEE Micro, 26(4):52--60, Jul/Aug 2006. Google ScholarDigital Library
- E. L. Congduc. Packet classi cation in the NIC for improved SMP-based internet servers. In Proc. Int 'l Conf. on Networking, Feb. 2004.Google Scholar
- M. Ekman and P. Stenstr. A cost-effective main memory organization for future servers. In Proc. of the Int'l Parallel and Distributed Processing Symp., Apr 2005. Google ScholarDigital Library
- S. Gregori, A. Cabrini, O. Khouri, and G. Torelli. On-chip error correcting techniques for new-generation flash memories. 91(4), Apr 2003.Google Scholar
- S. Gupta, M. Hilbert, S. Hong, and R. Patti. Techniques for producing 3D ICs with high-density interconnect. www. tezzaron. com/about/papers/ieee_vmic_2004_finalsecure. pdf.Google Scholar
- H. Huang, P. Pillai, and K. G. Shin. Design and Implementation of Power-Aware Virtual Memory. In USENIX Annual Technical Conference, pages 57--70, 2003. Google ScholarDigital Library
- P. Kongetira, K. Aingaran, and K. Olukotun. Niagara: A 32-way multithreaded Sparc processor. IEEE Micro, 25(2):21--29, Mar. 2005. Google ScholarDigital Library
- A. R. Lebeck, X. Fan, H. Zeng, and C. S. Ellis. Power aware page allocation. In Proc. Int 'l Conf. on Arch. Support for Programming Languages and Operating Systems, pages 105--116, 2000. Google ScholarDigital Library
- J. Lee, S.-S. Lee, O.-S. Kwon, K.-H. Lee, D.-S. Byeon, I. -Y. Kim, K.-H. Lee, Y.-H. Lim, B.-S. Choi, J.-S. Lee, W.-C. Shin, J.-H. Choi, and K.-D. Suh. A 90-nm CMOS 1. 8-V 2-Gb NAND Flash Memory for Mass Storage Applications. 38(11), Nov 2003.Google Scholar
- G. MacGillivray. Process vs. density in DRAMs. http://www. eetasia. com/ARTICLES/2005SEP/B/2005SEP01_STOR_TA. pdf.Google Scholar
- C. Park, J. Seo, S. Bae, H. Kim, S. Kim, and B. Kim. A Low-cost Memory Architecture With NAND XIP for Mobile Embedded Systems. In Proc. Int'l Conf. on HW-SW Codesign and System Synthesis (CODES+ISSS), Oct 2003. Google ScholarDigital Library
Index Terms
- FlashCache: a NAND flash memory file cache for low power web servers
Recommendations
LSTAFF*: an efficient flash translation layer for large block flash memory
SAC '11: Proceedings of the 2011 ACM Symposium on Applied ComputingRecently, flash memory is widely used as a non-volatile storage for embedded applications such as cellular phones, mp3 players, digital cameras, and so on. The software layer called FTL (flash translation layer) becomes more important since it is a key ...
Storage architecture and software support for SLC/MLC combined flash memory
SAC '09: Proceedings of the 2009 ACM symposium on Applied ComputingWe propose a novel flash memory management software for SLC/MLC combined flash memories which are recently introduced to provide flexible and cost-efficient embedded storage systems. To provide a fast and large capacity of flash memory, the proposed ...
A group-based wear-leveling algorithm for large-capacity flash memory storage systems
CASES '07: Proceedings of the 2007 international conference on Compilers, architecture, and synthesis for embedded systemsAlthough NAND flash memory has become one of the most popular storage media for portable devices, it has a serious problem with respect to lifetime. Each block of NAND flash memory has a limited number of program/erase cycles, usually 10,000-100,000, ...
Comments