PAM: an efficient power-aware multilevel cache policy to reduce energy consumption of storage systems

Abstract

Energy consumption is one of the most significant aspects of large-scale storage systems where multilevel caches are widely used. In a typical hierarchical storage structure, upper-level storage serves as a cache for the lower level, forming a distributed multilevel cache system. In the past two decades, several classic LRU-based multilevel cache policies have been proposed to improve the overall I/O performance of storage systems. However, few power-aware multilevel cache policies focus on the storage devices in the bottom level, which consume more than 27% of the energy of the whole system [1].

To address this problem, we propose a novel power-aware multilevel cache (PAM) policy that can reduce the energy consumption of high-performance and I/O bandwidth storage devices. In our PAM policy, an appropriate number of cold dirty blocks in the upper level cache are identified and selected to flush directly to the storage devices, providing high probability extension of the lifetime of disks in standby mode. To demonstrate the effectiveness of our proposed policy, we conduct several simulations with real-world traces. Compared to existing popular cache schemes such as PALRU, PB-LRU, and Demote, PAM reduces power consumption by up to 15% under different I/O workloads, and improves energy efficiency by up to 50.5%.

Keywords

storage system / multilevel cache / energy consumption / I/O performance / hint

References

[1] Parolini L, Sinopoli B, Krogh B H. Reducing data center energy consumption via coordinated cooling and load management. In: Proceedings of the 2008 Conference on Cluster Computing. 2008

[2] Appuswamy R, Moolenbroek D C, Tanenbaum A S. Cache, cache everywhere, flushing all hits down the sink: on exclusivity in multilevel, hybrid caches. In: Proceedings of the 29th IEEE Symposium on Mass Storage Systems and Technologies. 2013, 1–14

[3] Lampson B W. Hints for computer system design. IEEE Software, 1984, 1(1): 11–28

[4] Sarkar P, Hartman J H. Efficient cooperative caching using hints. In: Proceedings of Symposium on Operating Systems Design and Implementation, 1996, 35–46

[5] Patterson R H, Gibson G A, Ginting E. Informed prefetching and caching. Symposium on Operating Systems Principles, 1995, 29(5): 224–244

[6] Sarkar P, Hartman J. Hint-based cooperative caching. ACM Transactions on Computer Systems, 2000, 18(4): 387–419

[7] Wong T M,Wilkes J. My cache or yours?: making storage more exclusive. In: Proceedings of USENIX Annual Technical Conference. 2002, 161–175

[8] Bairavasundaram L, Sivathanu M, Arpaci-Dusseau A, Arpaci-Dusseau R. X-ray: a non-invasive exclusive caching mechanism for raids. In: Proceedings of Annual International Symposium on Computer Architecture. 2004, 176–187

[9] Chen Z, Zhou Y, Li K. Eviction-based cache placement for storage caches. In: Proceedings of USENIX Annual Technical Conference. 2003

[10] Zhou Y, Chen Z, Li K. Second-level buffer cache management. IEEE Transactions on Parallel and Distributed Systems, 2004, 15(6): 505–519

[11] He X, Kosa M, Scott S, Engelmann C. A unified multiple-level cache for high performance storage systems. International Journal of High Performance Computing and Networking, 2007, 5(1): 97–109

[12] Wu C T, He X, Cao Q, Xie C. Hint-k: an efficient multi-level cache using k-step hints. In: Proceedings of the 39th IEEE International Conference on Parallel Processing. 2010, 624–633

[13] Wu C T, He X, Cao Q, Xie C. Hint-k: an efficient multi-level cache using k-step hints. IEEE Transactions on Parallel and Distributed Systems, 2014, 25(3): 653–662

[14] Gill B S. On multi-level exclusive caching: offline optimality and why promotions are better than demotions. In: Proceedings of the 6th USENIX Conference on File and Storage Technologies. 2008

[15] Gill B S. Systems and methods for multi-level exclusive caching using hints. US Patent, 2010

[16] Yadgar G, Factor M, Schuster A. Karma: know-it-all replacement fora multilevel cache. In: Proceedings of USENIX Conference on File and Storage Technologies. 2007

[17] Yadgar G, Factor M, Li K, Schuster A. MC2: multiple clients on a multilevel cache. In: Proceedings of the IEEE International Conference on Distributed Computing Systems. 2008, 722–730

[18] Yadgar G, Factor M, Li K, Schuster A. Management of multilevel, multiclient cache hierarchies with application hints. ACM Transactions on Computer Systems, 2011, 29(2): 5

[19] Jiang S, Zhang X. ULC: a file block placement and replacement protocol to effectively exploit hierarchical locality in multi-level buffer caches. In: Proceedings of the 24th International Conferences on Distributed Computing Systems. 2004, 168–177

[20] Liu X, Aboulnaga A, Salem K, Li X. CLIC: client-informed caching for storage servers. In: Proceedings of the USENIX FAST. 2009, 297–310

[21] Zhou Y, Philbin J, Li K. The multi-queue replacement algorithm for second level buffer caches. In: Proceedings of USENIX Annual Technical Conference. 2001, 91–104

[22] Huang Q, Birman K P, Van Renesse R, Lloyd W, Kumar S, Li H C. An analysis of Facebook photocaching. In: Proceedings of the 24th ACM Symposium on Operating Systems Principles. 2013, 167–181

[23] Zhu Q, Shankar A, Zhou Y. PB-LRU: a self-tuning power aware storage cache replacement algorithm for conserving disk energy. In: Proceedings of the 18th Annual International Conference on Supercomputing. 2004, 79–88

[24] Song M. Saving disk energy in video servers by combining caching and prefetching. ACM Transactions on Multimedia Computing, Communications,and Applications, 2014, 10(1s): 15

[25] Yue J, Zhu Y, Cai Z. An energy-oriented evaluation of buffer cache algorithms using parallel I/O workloads. IEEE Transactions on Parallel and Distributed Systems, 2008, 19(11): 1565–1578

[26] Chai Y, Fan W. Three-state disk model for high quality and energy efficient streaming media servers. In: Proceedings of the 11th IEEE International Symposium on Autonomous Decentralized Systems. 2013, 1–8

[27] Zhu Q, David F M, Devaraj C F, Li Z, Zhou Y, Cao P. Reducing energy consumption of disk storage using power-aware cache management. In: Proceedings of the 10th International Symposium on High Performance Computer Architecture. 2004, 118

[28] Malina J, Boyle W. Hybrid drive changing power mode of disk channel when frequency of write data exceeds a threshold. US Patent 8, 670, 205, 2014

[29] Coker K, Boyle W. Disk drive executing log structured writes to physical zones based on power mode. US Patent 8, 576, 511, 2013

[30] Gargash N, Frantz A, Salsbery B, Barrett C. Dynamic low power mode implementation for computing devices. US Patent 9, 235, 251, 2016

[31] Lu Y, Micheli G. Comparing system-level power management policies. IEEE Design & Test, 2001, 18(2): 10–19

[32] Douglis F, Krishnan P, Marsh B. Thwarting the power-hungry disk. In: Proceedings of the Winter USENIX. 1994

[33] Irani S, Shukla S, Gupta R. Competitive analysis of dynamic power management strategies for systems with multiple power saving state. In: Proceedings of the Conference on Design, Automation, and Test in Europe. 2002, 117–123

[34] Gurumurthi S, Sivasubramaniam A. DRPM: dynamic speed control for power management in server class disks. International Symposium on Computer Architecture, 2003, 31(2): 169–179

[35] Denning P. The working set model for program behavior. Communications of the ACM, 1968, 11(5): 323–333

[36] Shasha D, Johnson T. 2Q: a low overhead high performance buffer management replacement algorithm. In: Proceedings of the 20th International Conference on Very Large Databases. 1994, 439–450

[37] O’neil E, O’neil P, Weikum G. The LRU-K page replacement algorithm for database disk buffering. ACM SIGMOD Record, 1993, 22(2): 297–306

[38] Megiddo N, Modha D S. ARC: a self-tuning, low overhead replacement cache. In: Proceedings of USENIX Conference on File and Storage Technologies. 2003, 115–130

[39] Robinson J, Devarakonda M. Data cache management using frequency-based replacement. Measurement and Modeling of Computer Systems, 1990, 18(1): 134–142

[40] Kim J, Choi J, Kim J, Noh S, Min S, Cho Y, Kim C. A low-overhead high-performance unified buffer management scheme that exploits sequential and looping references. In: Proceedings of the 4th Conference on Symposium on Operating System Design & Implementation. 2000

[41] Zhou F, von Behren J, Brewer E. AMP: program context specific buffer caching. In: Proceedings of USENIX Annual Technical Conference. 2005, 371–374

[42] Gill B, Modha D. WOW: wise ordering for writes- combining spatial and temporal locality in non-volatile caches. In: Proceedings of the 4th USENIX Conference on File and Storage Technologies. 2005

[43] Gill B, Ko M, Debnath B. STOW: a spatially and temporally optimized write caching algorithm. In: Proceedings of the 2009 Conference on USENIX Annual Technical Conference. 2009

[44] Zhu Y, Jiang H. RACE: a robust adaptive caching strategy for buffer cache. IEEE Transactions on Computers, 2008, 57(1): 25–40

[45] Gniady C, Butt A, Hu Y. Program-counter-based pattern classification in buffer caching. Operating Systems Design and Implementation, 2004, 395–408

[46] Bansal S, Modha D. CAR: clock with adaptive replacement. In: Proceedings of the 3rd USENIX Conference on File and Storage Technologies. 2004, 187–200

[47] Li X, Aboulnaga A, Salem K, Gao S. Second-tier cache management using write hints. In: Proceedings of the 4th USENIX Conference on File and Storage Technologies. 2005, 115–128

[48] Gudu D, Hardt M. Arm cluster for performant and energy efficient storage. Computational Sustainability, 2016, 265–276

[49] Rush E, Altiparmak N. Exploiting replication for energy efficiency of heterogeneous storage systems. IEEE Transactions on Parallel and Distributed Systems, 2015, 26(10): 2734–2749

[50] Zhan L, Men L, Xu P, Jian K. Design and implementation of SSD aware heterogeneous cache algorithm: a two-level caching algorithm for raid storage systems. In: Proceedings of IEEE International Conference on Cloud Computing and Big Data Analysis. 2016, 66–71

[51] Chang H, Chang Y, Kuan Y, Huang X, Kuo T, Li H. Pattern-aware write-back strategy to minimize energy consumption of PCM-based storage systems. In: Proceedings of the 5th Non-Volatile Memory Systems and Applications Symposium. 2016

[52] Meng X D, Zheng L, Li L. PAM: an efficient power-aware multi-level cache policy to reduce energy consumption of coftware defined network. In: Proceedings of the 1st International Conference on Industrial Networks and Intelligent Systems. 2015, 18–23

[53] Valentini G, Lassonde W. An overview of energy efficiency techniques in cluster computing systems. Cluster Computing, 2013, 16(1): 3–15

[54] Nimgaonkar S, Gomathisankaran M, Mohanty S. TSV: a novel energy efficient memory integrity verification scheme for embedded systems. Journal of Systems Architecture, 2013, 59(7): 400–411

[55] Deng Q, Meisner D. Multiscale: memory system DVFS with multiple memory controllers. In: Proceedings of International Symposium on Low Power Electronics and Design. 2012, 297–302

[56] Maniraj B, Jaghannath K, Swamy D. Control of reduced rating dynamic voltage restorer with a battery energy storage system. Imperial Journal of Interdisciplinary Research, 2016, 2(9): 618–621

[57] Deng Q, Meisner D. Memscale: active low-power modes for main memory. Architectural Support for Programming Languages and Operating Systems, 2011, 46(3): 225–238

[58] David H, Fallin C, Gorbatov E, Hanebutte U R, Mutlu O. Memory power management via dynamic voltage voltage/frequency scaling. In: Proceedings of the 8th ACM International Conference on Autonomic Computing. 2011, 31–40

[59] Bucy J, Schindler J, Schlosser S W, Ganger G R. The disksim simulation environment version 4.0 reference manual (cmupdl-08-101). Parallel Data Laboratory, 2008, 26

[60] Goyal P, Modha D, Tewari R. CacheCOW: providing QoS for storage system caches. Measurement and Modeling of Computer Systems, 2003, 31(1): 306–307

RIGHTS & PERMISSIONS

2018 Higher Education Press and Springer-Verlag GmbH Germany, part of Springer Nature