Advances in flash memory SSD technology for enterprise database applications

ABSTRACT

The past few decades have witnessed a chronic and widening imbalance among processor bandwidth, disk capacity, and access speed of disk. According to Amdhal's law, the performance enhancement possible with a given improvement is limited by the amount that the improved feature is used. This implies that the performance enhancement of an OLTP system would be seriously limited without a considerable improvement in I/O throughput. Since the market debut of flash memory SSD a few years ago, we have made a continued effort to overcome its poor random write performance and to provide stable and sufficient I/O bandwidth. In this paper, we present three different flash memory SSD models prototyped recently by Samsung Electronics. We then show how the flash memory SSD technology has advanced to reverse the widening trend of performance gap between processors and storage devices. We also demonstrate that even a single flash memory drive can outperform a level-0 RAID with eight enterprise class 15k-RPM disk drives with respect to transaction throughput, cost effectiveness and energy consumption.

REFERENCES

• 1: Amir Ban and Ramat Hasharon. Flash File System Optimized for Page-Mode Flash Technologies, August 1999. United States Patent 5937425.
• 2: Transaction Processing Performance Council. TPC Benchmark C Standard Specification (Revision 5. 9). http://www.tpc.org, June 2007.
• 3: Brian Dees. Native Command Queuing-Advanced Performance in Desktop Storage. IEEE Potentials 24(4):4--7, Oct/Nov 2005.
• 4: Rules of Thumb in Data Engineering, Proceedings of the 16th International Conference on Data Engineering, p.3, February 28-March 03, 2000
• 5: John L. Hennessy , David A. Patterson, Computer Architecture, Fourth Edition: A Quantitative Approach, Morgan Kaufmann Publishers Inc., San Francisco, CA, 2006
• 6: W. Hsu , A. J. Smith, Characteristics of I/O traffic in personal computer and server workloads, IBM Systems Journal, v.42 n.2, p.347-372, April 2003
• 7: W. Hsu , A. J. Smith, The performance impact of I/O optimizations and disk improvements, IBM Journal of Research and Development, v.48 n.2, p.255-289, March 2004
• 8: IBM. IBM Power 595 Server Model 9119-FHA Using AIX 5L Version 5. 3 and DB2 Enterprise 9. 5, June 2008. TPC Benchmark C Full Disclosure Report First Edition.
• 9: Sang-Won Lee , Bongki Moon, Design of flash-based DBMS: an in-page logging approach, Proceedings of the 2007 ACM SIGMOD international conference on Management of data, June 11-14, 2007, Beijing, China [doi>10.1145/1247480.1247488]
• 10: Sang-Won Lee , Bongki Moon , Chanik Park , Jae-Myung Kim , Sang-Woo Kim, A case for flash memory ssd in enterprise database applications, Proceedings of the 2008 ACM SIGMOD international conference on Management of data, June 09-12, 2008, Vancouver, Canada [doi>10.1145/1376616.1376723]
• 11: Oracle. ORION:Oracle I/O Numbers Calibration Tool. http://www.oracle.com/technology/software/-tech/orion/.
• 12: Seon-yeong Park , Dawoon Jung , Jeong-uk Kang , Jin-soo Kim , Joonwon Lee, CFLRU: a replacement algorithm for flash memory, Proceedings of the 2006 international conference on Compilers, architecture and synthesis for embedded systems, October 22-25, 2006, Seoul, Korea [doi>10.1145/1176760.1176789]
• 13: David A. Patterson, Latency lags bandwith, Communications of the ACM, v.47 n.10, p.71-75, October 2004 [doi>10.1145/1022594.1022596]
• 14: Meikel Poess and Raghunath Othayoth Nambiar. Energy Cost, The Key Challenge of Today's Data Centers: A Power Consumption Analysis of TPC-C Results. In Proceedings of VLDB 2008.
• 15: Chris Ruemmler , John Wilkes, An introduction to disk drive modeling, Computer, v.27 n.3, p.17-28, March 1994 [doi>10.1109/2.268881]

INDEX TERMS

Primary Classification:
  H. Information Systems
  H.2 DATABASE MANAGEMENT
      H.2.2 Physical Design
          Subjects: Access methods

General Terms:
Design, Measurement, Performance

Keywords:
energy, flash-memory ssd, tpc-c benchmark