ABSTRACT
As database application performance depends on the utilization of the memory hierarchy, smart data placement plays a central role in increasing locality and in improving memory utilization. Existing techniques, however, do not optimize accesses to all levels of the memory hierarchy and for all the different workloads, because each storage level uses different technology (cache, memory, disks) and each application accesses data using different patterns. Clotho is a new buffer pool and storage management architecture that decouples in-memory page layout from data organization on non-volatile storage devices to enable independent data layout design at each level of the storage hierarchy. Clotho can maximize cache and memory utilization by (a) transparently using appropriate data layouts in memory and non-volatile storage, and (b) dynamically synthesizing data pages to follow application access patterns at each level as needed. Clotho creates in-memory pages individually tailored for compound and dynamically changing workloads, and enables efficient use of different storage technologies (e.g., disk arrays or MEMS-based storage devices). This paper describes the Clotho design and prototype implementation and evaluates its performance under a variety of workloads using both disk arrays and simulated MEMS-based storage devices.
- {1} A. Ailamaki, D. J. DeWitt, M. D. Hill, and M. Skounakis. Weaving relations for cache performance. International Conference on Very Large Databases, pages 169-180. Morgan Kaufmann Publishing, Inc., 2001. Google ScholarDigital Library
- {2} A. Ailamaki, D. J. DeWitt, M. D. Hill, and D. A. Wood. DBMSs on a modern processor: where does time go? International Conference on Very Large Databases, pages 266-277. Morgan Kaufmann Publishing, Inc., 1999. Google ScholarDigital Library
- {3} P. A. Boncz, S. Manegold, and M. L. Kersten. Database architecture optimized for the new bottleneck: memory access. International Conference on Very Large Databases, pages 54-65. Morgan Kaufmann Publishers, Inc., 1999. Google ScholarDigital Library
- {4} M. J. Carey et al. Shoring up persistent applications. ACM SIGMOD International Conference on Management of Data. 1994. Google ScholarDigital Library
- {5} L. R. Carley et al. Single-chip computers with microelectromechanical systems-based magnetic memory. Journal of Applied Physics, 87(9):6680-6685, 1 May 2000.Google ScholarCross Ref
- {6} X. Chen and X. Zhang. Coordinated data prefetching by utilizing reference information at both proxy and web servers. Performance Evaluation Review, 29(2):32-38. ACM, September 2001. Google ScholarDigital Library
- {7} G. P. Copeland and S. Khoshafian. A decomposition storage model. ACM SIGMOD International Conference on Management of Data, pages 268-279. ACM Press, 1985. Google ScholarDigital Library
- {8} The DiskSim Simulation Environment (Version 3.0). http://-www.pdl.cmu.edu/DiskSim/index.html.Google Scholar
- {9} R. A. Hankins and J. M. Patel. Data morphing: an adaptive, cache-conscious storage technique. International Conference on Very Large Databases, pages 1-12. VLDB, 2003. Google ScholarDigital Library
- {10} S. Manegold, P. A. Boncz, and M. L. Kersten. Generic database cost models for hierarchical memory systems. International Conference on Very Large Databases, pages 191-202. Morgan Kaufmann Publishers, Inc., 2002. Google ScholarDigital Library
- {11} C. Mohan et al. ARIES: a transaction recovery method supporting fine-granularity locking and partial rollbacks using write-ahead logging. ACM Transactions on Database Systems, 17(1):94-162, March 1992. Google ScholarDigital Library
- {12} D. A. Patterson, G. Gibson, and R. H. Katz. A case for redundant arrays of inexpensive disks (RAID). ACM SIGMOD International Conference on Management of Data, pages 109-116, 1-3 June 1988. Google ScholarDigital Library
- {13} R. Ramakrishnan and J. Gehrke. Database management systems, number 3rd edition. McGraw-Hill, 2003. Google ScholarDigital Library
- {14} R. Ramamurthy, D. J. DeWitt, and Q. Su. A case for fractured mirrors. International Conference on Very Large Databases, pages 430-441. Morgan Kaufmann Publishers, Inc., 2002. Google ScholarDigital Library
- {15} J. Schindler, A. Ailamaki, and G. R. Ganger. Lachesis: robust database storage management based on device-specific performance characteristics. International Conference on Very Large Databases, pages 706-717. Morgan Kaufmann Publishing, Inc., 2003. Google ScholarDigital Library
- {16} J. Schindler et al. Atropos: a disk array volume manager for orchestrated use of disks. Conference on File and Storage Technologies, pages 159-172. USENIX Association, 2004. Google ScholarDigital Library
- {17} J. Schindler, J. L. Griffin, C. R. Lumb, and G. R. Ganger. Track-aligned extents: matching access patterns to disk drive characteristics. Conference on File and Storage Technologies, pages 259-274. USENIX Association, 2002. Google ScholarDigital Library
- {18} S. W. Schlosser, J. L. Griffin, D. F. Nagle, and G. R. Ganger. Designing computer systems with MEMS-based storage. Architectural Support for Programming Languages and Operating Systems. 2000. Google ScholarDigital Library
- {19} S. W. Schlosser, J. Schindler, A. Ailamaki, and G. R. Ganger. Exposing and exploiting internal parallelism in MEMS-based storage. Technical Report CMU-CS-03-125. Carnegie-Mellon University, Pittsburgh, PA, March 2003.Google Scholar
- {20} M. Shao and A. Ailamaki. DBMbench: Microbenchmarking database systems in a small, yet real world. Technical Report CMU-CS-03-161. Carnegie-Mellon University, Pittsburgh, PA, October 2003.Google Scholar
- {21} P. Vettiger et al. The "Millipede" - more than one thousand tips for future AFM data storage. IBM Journal of Research and Development , 44(3):323-340, 2000. Google ScholarDigital Library
- {22} H. Yu, D. Agrawal, and A. E. Abbadi. Tabular placement of relational data on MEMS-based storage devices. International Conference on Very Large Databases, pages 680-693, 2003. Google ScholarDigital Library
- {23} J. Zhou and K. A. Ross. A Multi-resolution Block Storage Model for Database Design. International Database Engineering & Applications Symposium, 2003.Google Scholar
Index Terms
- Clotho: decoupling memory page layout from storage organization
Recommendations
Clotho: an elastic MapReduce workload/runtime co-design
ARM '13: Proceedings of the 12th International Workshop on Adaptive and Reflective MiddlewareThe resource management of a multi-tenant MapReduce cluster can be hard given unpredictable user demands. Conventional resource management scheme would inevitably create a fair amount of spare resource fragments in the system. On the other hand, ...
Design and Optimization of Large Size and Low Overhead Off-Chip Caches
Large off-chip L3 caches can significantly improve the performance of memory-intensive applications. However, conventional L3 SRAM caches are facing two issues as those applications require increasingly large caches. First, an SRAM cache has a limited ...
TLB Improvements for Chip Multiprocessors: Inter-Core Cooperative Prefetchers and Shared Last-Level TLBs
Translation Lookaside Buffers (TLBs) are critical to overall system performance. Much past research has addressed uniprocessor TLBs, lowering access times and miss rates. However, as Chip MultiProcessors (CMPs) become ubiquitous, TLB design and ...
Comments