Abstract
There are trade-offs among performance, energy, and device endurance for storage systems. Designs optimized for one dimension or workload often suffer in another. Therefore, it is important to study the trade-offs to enable adaptation to workloads and dimensions. As Flash SSD has emerged, hybrid drives have been studied more closely. However, hybrids are mainly designed for high throughput, efficient energy consumption, or improving endurance—leaving quantitative study on the trade-offs unexplored. Past endurance studies also lack a concrete model to help study the trade-offs. Last, previous designs are often based on inflexible policies that cannot adapt easily to changing conditions.
We designed and developed GreenDM, a versatile hybrid drive that combines Flash-based SSDs with traditional HDDs. The SSD can be used as cache or as primary storage for hot data. We present our endurance model together with GreenDM to study these trade-offs. GreenDM presents a block interface and requires no modifications to existing software. GreenDM offers tunable parameters to enable the system to adapt to many workloads. We have designed, developed, and carefully evaluated GreenDM with a variety of workloads using commodity SSD and HDD drives. We demonstrate the importance of versatility to enable adaptation to various workloads and dimensions.
- D. G. Andersen, J. Franklin, M. Kaminsky, A. Phanishayee, L. Tan, and V. Vasudevan. 2009. FAWN: A fast array of wimpy nodes. In Proceedings of the 22nd ACM Symposium on Operating Systems Principles (SOSP’2009). ACM SIGOPS, New York, NY, 1--14. Google Scholar
Digital Library
- L. A. Barroso and U. Hölzle. 2009. The datacenter as a computer: An introduction to the design of warehouse-scale machines. Synthesis Lectures on Computer Architecture 4, 1, 1--108.Google Scholar
Cross Ref
- bcache 2015. Bcache. Retrieved July 5, 2015 from http://bcache.evilpiepirate.org/.Google Scholar
- T. Bisson, S. A. Brandt, and D. D. E. Long. 2007. A hybrid disk-aware spin-down algorithm with I/O subsystem support. In Proceedings of the 26th IEEE International Performance, Computing and Communications Conference. IEEE Computer Society, Los Alamitos, CA, 236--245.Google Scholar
Cross Ref
- A. Brown and M. Seltzer. 1997. Operating system benchmarking in the wake of Lmbench: A case study of the performance of NetBSD on the Intel x86 architecture. In Proceedings of the 1997 ACM SIGMETRICS Conference on Measurement and Modeling of Computer Systems. ACM Press, New York, NY, 214--224. Google Scholar
Digital Library
- F. Chen, D. A. Koufaty, and X. Zhang. 2011. Hystor: Making the best use of solid state drives in high performance storage systems. In Proceedings of the International Conference on Supercomputing. ACM, New York, NY, 22--32. Google Scholar
Digital Library
- D. Colarelli and D. Grunwald. 2002. Massive arrays of idle disks for storage archives. In Proceedings of the 2002 ACM/IEEE conference on Supercomputing. ACM, New York, NY, USA, 1--11. Google Scholar
Digital Library
- V. Delaluz, A. Sivasubramaniam, M. Kandemir, N. Vijaykrishnan, and M. J. Irwin. 2002. Scheduler-based DRAM energy management. In Proceedings of the 39th annual Design Automation Conference. ACM, New York, NY, USA, 697--702. Google Scholar
Digital Library
- Dell Flash Array. 2015. Compellent Flash Optimized Solutions. http://www.dell.com/us/business/p/dell-compellent-flash-optimized/pd.Google Scholar
- P. Desnoyers. 2013. What systems researchers need to know about NAND flash. In HotStorage’13: Proceedings of the 5th USENIX Workshop on Hot Topics in Storage. USENIX Association, Berkeley, CA, 1--5. Google Scholar
Digital Library
- Filebench. 2014. Filebench. http://filebench.sourceforge.net/.Google Scholar
- R. Freitas. 2009. Storage class memory: Technology, systems and applications. In Proceedings of the 2009 ACM SIGMOD International Conference on Management of Data (SIGMOD’09). ACM, New York, NY, 985--986. Google Scholar
Digital Library
- J. Gantz and D. Reinsel. 2010. The Digital Universe Decade—Are You Ready? http://www.emc.com/collateral/analyst-reports/idc-digital-universe-are-you-ready.pdf.Google Scholar
- GitHub 2015. facebook/flashcache. Retrieved July 5, 2015 from https://github.com/facebook/flashcache/.Google Scholar
- B. Guenter, N. Jain, and C. Williams. 2011. Managing cost, performance, and reliability tradeoffs for energy-aware server provisioning. In INFOCOM 2011. 30th IEEE International Conference on Computer Communications, Joint Conference of the IEEE Computer and Communications Societies. IEEE, Los Alamitos, CA, 1332--1340.Google Scholar
- J. Guerra, H. Pucha, J. Glider, W. Belluomini, and R. Rangaswami. 2011. Cost effective storage using extent based dynamic tiering. In USENIX FAST. USENIX Association, Berkeley, CA, 20--34. Google Scholar
Digital Library
- S. Gurumurthi, A. Sivasubramaniam, M. Kandemir, and H. Franke. 2003. DRPM: Dynamic speed control for power management in server class disks. In Proceedings of the 30th International Symposium on Computer Architecture. ACM Press, New York, NY, 169--179. Google Scholar
Digital Library
- W. Jiang, C. Hu, Y. Zhou, and A. Kanevsky. 2008. Are disks the dominant contributor for storage failures? A comprehensive study of storage subsystem failure characteristics. In Proceedings of the 5th UNENIX Conference on File and Storage Technologies (FAST’08). USENIX Association, San Jose, CA. 111--125. Google Scholar
Digital Library
- N. Joukov and J. Sipek. 2008. GreenFS: Making enterprise computers greener by protecting them better. In Proceedings of the 3rd ACM SIGOPS/EuroSys European Conference on Computer Systems 2008 (EuroSys’08). ACM, Glasgow, Scotland, 1--14. Google Scholar
Digital Library
- N. Joukov, A. Traeger, R. Iyer, C. P. Wright, and E. Zadok. 2006. Operating system profiling via latency analysis. In Proceedings of the 7th Symposium on Operating Systems Design and Implementation (OSDI’06). ACM SIGOPS, Seattle, WA, 89--102. Google Scholar
Digital Library
- M. Jung and M. Kandemir. 2013. Revisiting widely held SSD expectations and rethinking system-level implications. In Proceedings of the ACM SIGMETRICS/International Conference on Measurement and Modeling of Computer Systems (SIGMETRICS’13). ACM, New York, NY, USA, 203--216. Google Scholar
Digital Library
- Asim Kadav, Mahesh Balakrishnan, Vijayan Prabhakaran, and Dahlia Malkhi. 2009. Differential RAID: Rethinking RAID for SSD reliability. In HotStorage’09: Proceedings of the 1st Workshop on Hot Topics in Storage. ACM, New York, NY, 4:1--4:22.Google Scholar
- R. T. Kaushik and M. Bhandarkar. 2010. GreenHDFS: Towards an energy-conserving, storage-efficient, hybrid Hadoop compute cluster. In Proceedings of the 2010 International Conference on Power Aware Computing and Systems (HotPower’10). USENIX Association, Berkeley, CA, 1--9. Google Scholar
Digital Library
- Y. Kim, A. Gupta, B. Urgaonkar, P. Berman, and A. Sivasubramaniam. 2011. HybridStore: A cost-efficient, high-performance storage system combining SSDs and HDDs. In IEEE MASCOTS. IEEE Computer Society, Washington, DC, 227--236. Google Scholar
Digital Library
- H. Kim, S. Seshadri, C. L. Dickey, and L. Chiu. 2014. Evaluating phase change memory for enterprise storage systems: A study of caching and tiering approaches. In Proceedings of the 12th USENIX Conference on File and Storage Technologies. USENIX, Berkeley, CA, 33--45. Google Scholar
Digital Library
- J. G. Koomey. 2011. Growth in Data Center Electricity Use 2005 to 2010. Technical Report. Stanford University, Stanford, CA.Google Scholar
- R. Kothiyal, V. Tarasov, P. Sehgal, and E. Zadok. 2009. Energy and performance evaluation of lossless file data compression on server systems. In Proceedings of the Second ACM Israeli Experimental Systems Conference (SYSTOR’09). ACM, Haifa, Israel, 4:1--4:12. Google Scholar
Digital Library
- S. Lee, T. Kim, K. Kim, and J. Kim. 2012. Lifetime management of flash-based SSDs using recovery-aware dynamic throttling. In Proceedings of the 10th UNENIX Conference on File and Storage Technologies (FAST’12). USENIX Association, San Jose, CA, 26--26. Google Scholar
Digital Library
- S. Li and A. Belay. 2007. cpuidle — do nothing, efficiently.... In Proceedings of the Linux Symposium, Vol. 2. Linux Symposium, Ottawa, Ontario, Canada, 1--10.Google Scholar
- Z. Li, A. Desai, C. Bhatt, and E. Zadok. 2012a. vATM: vSphere adaptive task management. In Proceedings of the 7th International Workshop on Feedback Computing (FC’12). ACM, New York, NY, 1--7.Google Scholar
- Z. Li, K. M. Greenan, A. W. Leung, and E. Zadok. 2012b. Power consumption in enterprise-scale backup storage systems. In Proceedings of the 10th UNENIX Conference on File and Storage Technologies (FAST’12). USENIX Association, San Jose, CA, 6--13. Google Scholar
Digital Library
- Z. Li, R. Grosu, K. Muppalla, S. A. Smolka, S. D. Stoller, and E. Zadok. 2011a. Model discovery for energy-aware computing systems: An experimental evaluation. In Proceedings of the 1st Workshop on Energy Consumption and Reliability of Storage Systems (ERSS’11). IEEE Computer Society, Los Alamitos, CA, 1--6. Google Scholar
Digital Library
- Z. Li, R. Grosu, P. Sehgal, S. A. Smolka, S. D. Stoller, and E. Zadok. 2011b. On the energy consumption and performance of systems software. In Proceedings of the 4th Israeli Experimental Systems Conference (ACM SYSTOR’11). ACM, Haifa, Israel, 8:1--8:12. Google Scholar
Digital Library
- Z. Li, A. Mukker, and E. Zadok. 2014. On the importance of evaluating storage systems’ $costs. In Proceedings of the 6th USENIX Conference on Hot Topics in Storage and File Systems (HotStorage’14). USENIX, Berkeley, CA, 1--5. to appear. Google Scholar
Digital Library
- T. Luo, R. Lee, M. Mesnier, F. Chen, and X. Zhang. 2012. hStorage-DB: Heterogeneity-aware data management to exploit the full capability of hybrid storage systems. Proceedings of the VLDB Endowment 5, 10, 1076--1087. Google Scholar
Digital Library
- V. Mohan, T. Siddiqua, S. Gurumurthi, and M. R. Stan. 2010. How I learned to stop worrying and love Flash endurance. In Proceedings of the 2nd USENIX Conference on Hot Topics in Storage and File Systems (HotStorage’10). USENIX Association, Berkeley, CA, 8--13. Google Scholar
Digital Library
- NetApp. 2015. NetApp Data Ontap 8.3: Flash optimization features. Retrieved July 5, 2015 from http://www.netapp.com/us/products/platform-os/flashpool.aspx.Google Scholar
- nimblestorage 2015. Driving storage lifecycle efficiency with data science. Retrieved July 5, 2015 from www.nimblestorage.com/products/architecture.php.Google Scholar
- R. Panabaker. 2006. Hybrid Hard Disk and Ready-Drive Technology: Improving Performance and Power for Windows Vista Mobile PCs. Retrieved July 6, 2015 from http://www.microsoft.com/whdc/winhec/pres06.mspx.Google Scholar
- E. Pinheiro and R. Bianchini. 2004. Energy conservation techniques for disk array-based servers. In Proceedings of the 18th International Conference on Supercomputing (ICS’04). ACM, New York, NY, 68--78. Google Scholar
Digital Library
- E. Pinheiro, W. Weber, and L. A. Barroso. 2007. Failure trends in a large disk drive population. In Proceedings of the 4th UNENIX Conference on File and Storage Technologies (FAST’07). USENIX Association, San Jose, CA, 17--28. Google Scholar
Digital Library
- SNIA Iotta Repository. 2011. FIU SRCMap Trace Repository. Retrieved July 5, 2015 from http://iotta. snia.org/traces/414.Google Scholar
- G. Soundararajan, V. Prabhakaran, M. Balakrishnan, and T. Wobber. 2010. Extending SSD lifetimes with disk-based write caches. In Proceedings of the 8th USENIX Conference on File and Storage Technologies (FAST’10). USENIX Association, Berkeley, CA, 8--20. Google Scholar
Digital Library
- Sourceforge. 2015. btier beta. Retrieved July 5, 2015 from http://sourceforge.net/projects/tier/.Google Scholar
- M. W. Storer, K. M. Greenan, E. L. Miller, and K. Voruganti. 2008. Pergamum: Replacing tape with energy efficient, reliable, disk-based archival storage. In Proceedings of the 5th UNENIX Conference on File and Storage Technologies (FAST’08). USENIX Association, San Jose, CA, 1:1--1:16. Google Scholar
Digital Library
- J. D. Strunk. 2012. Hybrid aggregates: Combining SSDs and HDDs in a single storage pool. SIGOPS Operating Systems Review 46, 3, 50--56. Google Scholar
Digital Library
- V. Tarasov, S. Bhanage, E. Zadok, and M. Seltzer. 2011. Benchmarking file system benchmarking: It *IS* rocket science. In Proceedings of HotOS XIII: The 13th USENIX Workshop on Hot Topics in Operating Systems. USENIX Association, Napa, CA, 8--13. Google Scholar
Digital Library
- Tintri. 2013. Tintri VMStore. Retrieved July 5, 2015 from www.tintri.com/resources/videos/introduction-to- tintri/.Google Scholar
- UMassTraceRepository. 2009. UMass Trace Repository. http://traces.cs.umass.edu.Google Scholar
- A. Verma, R. Koller, L. Useche, and R. Rangaswami. 2010. SRCMap: Energy proportional storage using dynamic consolidation. In Proceedings of the 8th USENIX Conference on File and Storage Technologies (FAST’10). USENIX Association, Berkeley, CA, USA, 8--20. Google Scholar
Digital Library
- vFlash 2012. Virtual Flash Tech Preview. Retrieved July 5, 2015 from http://blogs.vmware.com/vsphere/2012/12/virtual-flash-vflash-te ch-preview.html.Google Scholar
- Watts up. 2010. Smart Circuit Controllers. Retrieved July 5, 2015 from www.wattsupmeters.com/secure/products.php.Google Scholar
- C. Weddle, M. Oldham, J. Qian, A. A. Wang, P. Reiher, and G. Kuenning. 2007. PARAID: A gear-shifting power-aware RAID, See FAST 2007 {2007}, 245--260. Google Scholar
Digital Library
- Western Digital. 2013. WD Blue. Retrieved July 5, 2015 from http://www.wd.com/en/products/products.aspx?id=800#tab11.Google Scholar
- M. Wei, L. M. Grupp, F. E. Spada, and S. Swanson. 2011. Reliably erasing data from flash-based solid state drives. In Proceedings of the 9th USENIX Conference on File and Stroage Technologies (FAST’11). USENIX Association, Berkeley, CA, USA, 8--22. Google Scholar
Digital Library
- Wikipedia. 2014. ZCAV. Retrieved July 5, 2015 from https://en.wikipedia.org/wiki/Zone_bit_recording.Google Scholar
- Wikipedia. 2015a. Device Mapper. Retrieved July 5, 2015 from http://en.wikipedia.org/wiki/Device_mapper.Google Scholar
- Wikipedia 2015b. Fusion Drive. Retrieved July 5, 2015 from http://en.wikipedia.org/wiki/Fusion_Drive.Google Scholar
- Wikipedia. 2015c. Gamma distribution. Retrieved July 5, 2015 from http://en.wikipedia.org/wiki/Gamma_distribution.Google Scholar
- A. W. Wilson. 2008. Operation and implementation of random variables in Filebench. White Paper, Sun Microsystems.Google Scholar
- C. P. Wright, N. Joukov, D. Kulkarni, Y. Miretskiy, and E. Zadok. 2005. Auto-pilot: A platform for system software benchmarking. In Proceedings of the Annual USENIX Technical Conference, FREENIX Track. USENIX Association, Anaheim, CA, 175--187. Google Scholar
Digital Library
- T. Xie and Y. Sun. 2008. PEARL: Performance, energy, and reliability balanced dynamic data redistribution for next generation disk arrays. In IEEE MASCOTS. IEEE Computer Society, Washington, DC, 8--16.Google Scholar
- Q. Zhu, Z. Chen, L. Tan, Y. Zhou, K. Keeton, and J. Wilkes. 2005. Hibernator: Helping disk arrays sleep through the winter. In Proceedings of the 20th ACM Symposium on Operating Systems Principles (SOSP’05). ACM Press, Brighton, UK, 177--190. Google Scholar
Digital Library
- X. Zhu, M. Uysal, Z. Wang, S. Singhal, A. Merchant, P. Padala, and K. Shin. 2009. What does control theory bring to systems research? SIGOPS Operating Systems Review 43, 1, 62--69. Google Scholar
Digital Library
- Y. Zhu and F. Mueller. 2004. Feedback EDF scheduling exploiting dynamic voltage scaling. In Proceedings of the 10th IEEE Real-Time and Embedded Technology and Applications Symposium (RTAS’04). IEEE Computer Society, Washington, DC, 33--63. Google Scholar
Digital Library
Index Terms
On the Trade-Offs among Performance, Energy, and Endurance in a Versatile Hybrid Drive
Recommendations
Read-Performance Optimization for Deduplication-Based Storage Systems in the Cloud
Data deduplication has been demonstrated to be an effective technique in reducing the total data transferred over the network and the storage space in cloud backup, archiving, and primary storage systems, such as VM (virtual machine) platforms. However, ...
Efficient journaling writeback schemes for reliable and high-performance storage systems
We propose a efficient writeback scheme that enables guaranteeing throughput in high-performance storage systems. The proposed scheme, called de-fragmented writeback (DFW), reduces positioning time of storage devices in writing workloads, and thus ...
Enhancing hybrid parallel file system through performance and space-aware data layout
Hybrid parallel file systems PFSs, which consist of solid-state drive servers SServer and hard disk drive servers HServer, have recently attracted growing attention. Compared to a traditional HServer, an SServer consistently provides improved storage ...






Comments