Evaluating and repairing write performance on flash devices

ABSTRACT

In the last few years NAND flash storage has become more and more popular as price per GB and capacity both improve at exponential rates. Flash memory offers significant benefits compared to magnetic hard disk drives (HDDs) and DBMSs are highly likely to use flash as a general storage backend, either alone or in heterogeneous storage solutions with HDDs. Flash devices, however, respond quite differently than HDDs for common access patterns, and recent research shows a strong asymmetry between read and write performance. Moreover, flash storage devices behave unpredictably, showing a high dependence on previous IO history and usage patterns.

In this paper we investigate how a DBMS can overcome these issues to take full advantage of flash memory as persistent storage. We propose new a flash aware data layout --- append and pack --- which stabilizes device performance by eliminating random writes. We assess the impact of append and pack on OLTP workload performance using both an analytical model and micro-benchmarks, and our results suggest that significant improvements can be achieved for real workloads.

REFERENCES

• 1: The FusionIO drive. Technical specifications available at: http://www.fusionio.com/PDFs/Fusion%20Specsheet.pdf.
• 2: The LogFS file system. Available at: http://logfs.org/logfs/.
• 3: Transaction Processing Performance Council (TPC). TPC Benchmark C: Standard Specification. Available online at: http://www.tpc.org/tpcc/spec/tpcc_current.pdf.
• 4: L. Bouganim, B. T. Jónsson, and P. Bonnet. uFLIP: Understanding Flash IO Patterns. In CIDR, 2009.
• 5: Hui Dai , Michael Neufeld , Richard Han, ELF: an efficient log-structured flash file system for micro sensor nodes, Proceedings of the 2nd international conference on Embedded networked sensor systems, November 03-05, 2004, Baltimore, MD, USA [doi>10.1145/1031495.1031516]
• 6: Atsuo Kawaguchi , Shingo Nishioka , Hiroshi Motoda, A flash-memory based file system, Proceedings of the USENIX 1995 Technical Conference Proceedings on USENIX 1995 Technical Conference Proceedings, p.13-13, January 16-20, 1995, New Orleans, Louisiana
• 7: Gye-Jeong Kim , Seung-Cheon Baek , Hyun-Sook Lee , Han-Deok Lee , Moon Jeung Joe, LGeDBMS: a small DBMS for embedded system with flash memory, Proceedings of the 32nd international conference on Very large data bases, September 12-15, 2006, Seoul, Korea
• 8: 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]
• 9: 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]
• 10: Mendel Rosenblum , John K. Ousterhout, The design and implementation of a log-structured file system, ACM Transactions on Computer Systems (TOCS), v.10 n.1, p.26-52, Feb. 1992 [doi>10.1145/146941.146943]
• 11: D. Woodhouse. JFFS: The Journalling Flash File System. Ottawa Linux Symposium, 2001, available at: http://sources.redhat.com/jffs2/jffs2.pdf.

INDEX TERMS

Primary Classification:
  D. Software
  D.4 OPERATING SYSTEMS
      D.4.2 Storage Management
          Subjects: Allocation/deallocation strategies

Additional Classification:
  B. Hardware
  B.3 MEMORY STRUCTURES
      B.3.2 Design Styles
          Subjects: Mass storage (e.g., magnetic, optical, RAID)

  D. Software
  D.4 OPERATING SYSTEMS
      D.4.8 Performance
          Subjects: Operational analysis
      D.4.m Miscellaneous

General Terms:
Algorithms, Design, Management, Performance

Keywords:
data layout, database management systems, flash memory, storage virtualization