Nikolai Joukov
Abstract
Hard disks contain data - frequently an irreplaceable asset of high monetary and non-monetary value. At the same time, hard disks are mechanical devices that consume power, are noisy, and fragile when their platters are rotating.
In this paper we demonstrate that hard disks cause different kinds of problems for different types of computer systems and demystify several common misconceptions. We show that solutions developed to date are incapable of solving the power consumption, noise, and data reliability problems without sacrificing hard disk life-time, data reliability, or user convenience.
We considered data reliability, recovery, performance, user convenience, and hard disk-caused problems together at the enterprise scale. We have designed GreenFS: a fan-out stackable file system that offers all-time all-data run-time data protection, improves performance under typical user workloads, and allows hard disks to be kept off most of the time. As a result, GreenFS improves enterprise data protection, minimizes disk drive-related power consumption and noise and increases the chances of disk drive survivability in case of unexpected external impacts.
- N. Agrawal, W. J. Bolosky, J. R. Douceur, and J. R. Lorch. A five-year study of file-system metadata. In Proceedings of the Fifth USENIX Conference on File and Storage Technologies (FAST '07), pages 31--45, San Jose, CA, February 2007. USENIX Association. Google Scholar
Digital Library
- M. Anand and J. Flinn. PAN-on-demand: Building self-organizing PANs for better power management. Technical Report CSE-TR-524-06, Computer Science and Engineering Division, University of Michigan, August 2006.Google Scholar
- M. Anand, E. B. Nightingale, and J. Flinn. Ghosts in the machine: Interfaces for better power management. In Proceedings of MobiSys 2004, pages 23--35, Boston, MA, June 2004. ACM. Google ScholarDigital Library
- L. A. Barroso and U. Holzle. The case for energy-proportional computing. Computer, 40(12):33--37, December 2007. Google ScholarDigital Library
- T. Bisson, S. A. Brandt, and D. Long. A hybrid disk-aware spin-down algorithm with I/O subsystem support. In Proceedings of the International Performance Conference on Computers and Communication (IPCCC '07), New Orleans, April 2007.Google ScholarCross Ref
- E. V. Carrera, E. Pinheiro, and R. Bianchini. Conserving disk energy in network servers. In Proceedings of the International Conference on Supercomputers (ICS '03), San Francisco, CA, June 2003. ACM. Google ScholarDigital Library
- Noise Pollution Clearinghouse. Noise control in PCs - reduction of noise in PCs, May 2004. http://www.nonoise.org/resource/pcnoise/poweroid/poweroid.htm.Google Scholar
- D. Colarelli and D. Grunwald. Massive arrays of idle disks for storage archives. In Proceedings of the Supercomputing Conference 2002 (SC2002), pages 1--11, Baltimore, MD, November 2002. Google ScholarDigital Library
- Tokyo Electric Power Company. Service guide.Google Scholar
- F. Douglis, P. Krishnan, and B. Marsh. Thwarting the power-hungry disk. In Proceedings of the Winter USENIX Technical Conference, pages 293--306. USENIX Association, January 1994. Google ScholarDigital Library
- G. W. Evans and D. Johnson. Stress and open-office noise. Journal of Applied Psychology, 85(5):779--783, 2000.Google ScholarCross Ref
- M. A. Halcrow. eCryptfs: An Enterprise-class Encrypted Filesystem for Linux. In Proceedings of the 2005 Linux Symposium, pages 201--218, Ottawa, Canada, July 2005. Linux Symposium.Google Scholar
- Hitachi CinemaStar Hard Disk Drive Specifications Hitachi Global Storage Technologies.Google Scholar
- B. Hong, D. Plantenberg, D. D. E. Long, and M. Sivan-Zimet. Duplicate data elimination in a san file system. In Proceedings of the 12th NASA Goddard, 21st IEEE Conference on Mass Storage Systems and Technologies (MSST 2004), pages 301--314, College Park, MD, April 2004. IEEE.Google Scholar
- H. Huang, W. Hung, and K. Shin. FS2: Dynamic Data Replication in Free Disk Space for Improving Disk Performance and Energy Consumption. In Proceedings of the 20th ACM Symposium on Operating Systems Principles (SOSP '05), pages 263--276, Brighton, UK, October 2005. ACM Press. Google ScholarDigital Library
- IBM. Active Protection System whitepaper, October 2003.Google Scholar
- D. D. E. Long J. Rybczynski and A. Amer. Expecting the unexpected: Adaptation for predictive energy conservation. In Proceedings of the First ACM Workshop on Storage Security and Survivability (StorageSS 2005), pages 130--134, Fairfax, VA, November 2005. ACM. Google ScholarDigital Library
- N. Joukov, A. M. Krishnakumar, C. Patti, A. Rai, S. Satnur, A. Traeger, and Erez Zadok. Raif: Redundant array ofindependent filesystems. In Proceedings of the 24th International IEEE Symposium on Mass Storage Systems and Technologies, pages 199--212, San Diego, CA, September 2007. IEEE. Google ScholarDigital Library
- N. Joukov, A. Traeger, R. Iyer, C. P. Wright, and E. Zadok. Operating system profiling via latency analysis. In Proceedings of the 7th Symposium on Operating Systems Design and Implementation (OSDI 2006), pages 89--102, Seattle, WA, November 2006. ACM SIGOPS. Google ScholarDigital Library
- N. Joukov and E. Zadok. Adding Secure Deletion to Your Favorite File System. In Proceedings of the third international IEEE Security In Storage Workshop (SISW 2005), pages 63--70, San Francisco, CA, December 2005. Google ScholarDigital Library
- A. Kashyap, S. Patil, G. Sivathanu, and E. Zadok. I3FS: An In-Kernel Integrity Checker and Intrusion Detection File System. In Proceedings of the 18th USENIX Large Installation System Administration Conference (LISA 2004), pages 69--79, Atlanta, GA, November 2004. USENIX Association. Google ScholarDigital Library
- P. Khanna. Operation consolidation: reducing the number of servers can offer dramatic cost savings, but experts warn that trying to cram too much onto one box can backfire. Computing Canada, March 12 2004. by quoting G. Haff, senior analyst at Illuminata, Inc.Google Scholar
- J. Koomey. Estimating total power consumption by servers in the U.S. and the world. Technical Report Final Report, Lawrence Berkeley National Laboratory, February 2007.Google Scholar
- P. Kulkarni, F. Douglis, J. LaVoie, and J. M. Tracey. Redundancy elimination within large collections of files. In Proceedings of the Annual USENIX Technical Conference, pages 59--72, Boston, MA, June 2004. USENIX Association. Google ScholarDigital Library
- Rakesh Kumar. Gartner: A message from data center managers to CIOs: Floor space, power and cooling will limit our growth, August 2006.Google Scholar
- G. Laden, P. Ta-Shma, E. Yaffe, M. Factor, and S. Fienblit. Architectures for controller based CDP. In Proceedings of the Fifth USENIX Conference on File and Storage Technologies (FAST '07), pages 107--121, San Jose, CA, February 2007. USENIX Association. Google ScholarDigital Library
- J. Layton. The coming of diskless clusters. Linux Magazine, October 2005.Google Scholar
- A. Mahesri and V. Vardhan. Power consumption breakdown on a modern laptop. In Proceedings of the Workshop on Power-Aware Computer Systems (PACS 2004), Portland, OR, December 2004. IEEE Computer Society. Google ScholarDigital Library
- Y. Miretskiy, A. Das, C. P. Wright, and E. Zadok. Avfs: An On-Access Anti-Virus File System. In Proceedings of the 13th USENIX Security Symposium (Security 2004), pages 73--88, San Diego, CA, August 2004. USENIX Association. Google ScholarDigital Library
- K. Muniswamy-Reddy, C. P. Wright, A. Himmer, and E. Zadok. A Versatile and User-Oriented Versioning File System. In Proceedings of the Third USENIX Conference on File and Storage Technologies (FAST 2004), pages 115--128, San Francisco, CA, March/April 2004. USENIX Association. Google ScholarDigital Library
- E. Nightingale and J. Flinn. Energy-efficiency and storage flexibility in the blue file system. In Proceedings of the 6th Symposium on Operating Systems Design and Implementation (OSDI 2004), pages 363--378, San Francisco, CA, December 2004. ACM SIGOPS. Google ScholarDigital Library
- A. E. Papathanasiou and M. L. Scott. Energy efficient prefetching and caching. In Proceedings of the Annual USENIX Technical Conference, pages 255--268, Boston, MA, June 2004. USENIX Association. Google ScholarDigital Library
- Z. N. J. Peterson and R. C. Burns. Ext3cow: The design, Implementation, and Analysis of Metadata for a Time-Shifting File System. Technical Report HSSL-2003-03, Computer Science Department, The Johns Hopkins University, 2003. http://hssl.cs.jhu.edu/papers/peterson-ext3cow03.pdf.Google Scholar
- D. Quigley, J. Sipek, C. P. Wright, and E. Zadok. UnionFS: User- and Community-oriented Development of a Unification Filesystem. In Proceedings of the 2006 Linux Symposium, volume 2, pages 349--362, Ottawa, Canada, July 2006.Google Scholar
- D. S. H. Rosenthal. Evolving the Vnode interface. In Proceedings of the Summer USENIX Technical Conference, pages 107--118, Anaheim, CA, June 1990. USENIX Association.Google Scholar
- M. Russinovich. Inside the windows vista kernel: Part 2. Microsoft TechNet Magazine, 2007.Google Scholar
- P. Sarbanes and M. G. Oxley. Sarbanes-Oxley Act of 2002. U.S. Government Printing Office, July 2002.Google Scholar
- B. Schroeder and G. A. Gibson. Disk failures in the real world: What does an MTTF of 1,000,000 hours mean to you? In Proceedings of the Fifth USENIX Conference on File and Storage Technologies (FAST '07), pages 1--16, San Jose, CA, February 2007. USENIX Association. Google ScholarDigital Library
- SPEC. SPEC SFS97 R1 V3.0. www.spec.org/sfs97r1, September 2001.Google Scholar
- Silicon Systems. Increasing flash solid state disk reliability, April 2005.Google Scholar
- A. Traeger, K. Thangavelu, and E. Zadok. Round-trip privacy with NFSv4. In Proceedings of the Third ACM Workshop on Storage Security and Survivability (StorageSS 2007), pages 1--7, Alexandria, VA, October 2007. ACM. Google ScholarDigital Library
- M. Trainor. Overcoming disk drive access bottlenecks with intel robson technology. TechnologyυIntel Magazine, December 2006.Google Scholar
- A. A. Wang, P. Reiher, G. J. Popek, and G. H. Kuenning. Conquest: Better Performance Through A Disk/Persistent-RAM Hybrid File System. In Proceedings of the Annual USENIX Technical Conference, pages 15--28, Monterey, CA, June 2002. USENIX Association. Google ScholarDigital Library
- C. Weddle, M. Oldham, J. Qian, A. A. Wang, P. Reiher, and G. Kuenning. PARAID: A gear-shifting power-aware RAID. In Proceedings of the Fifth USENIX Conference on File and Storage Technologies (FAST '07), pages 245--260, San Jose, CA, February 2007. USENIX Association. Google ScholarDigital Library
- J. Wires and M. J. Feeley. Secure file system versioning at the block level. In Proceedings of the EuroSys 2007 Conference, pages 203--215, Lisboa, Portugal, March 2007. ACM. Google ScholarDigital Library
- C. P. Wright, N. Joukov, D. Kulkarni, Y. Miretskiy, and E. Zadok. Auto-pilot: A platform for system software benchmarking. In Proceedings of the Annual USENIXTechnical Conference, FREENIX Track, pages 175--187, Anaheim, CA, April 2005. USENIX Association. Google ScholarDigital Library
- C. P. Wright, M. Martino, and E. Zadok. NCryptfs: A secure and convenient cryptographic file system. In Proceedings of the Annual USENIX Technical Conference, pages 197--210, San Antonio, TX, June 2003. USENIX Association.Google Scholar
- X. Yao and J. Wang. RIMAC: A novel redundancy-based hierarchical cache architecture for energy efficient, high performance storage systems. In Proceedings of the EuroSys 2006 Conference, pages 249--262, Leuven, Belgium, April 2006. ACM. Google ScholarDigital Library
- E. Zadok, J. M. Anderson, I. Bădulescu, and J. Nieh. Fast Indexing: Support for size-changing algorithms in stackable file systems. In Proceedings of the Annual USENIX Technical Conference, pages 289--304, Boston, MA, June 2001. USENIX Association. Google ScholarDigital Library
- E. Zadok, R. Iyer, N. Joukov, G. Sivathanu, and C. P. Wright. On incremental file system development. ACM Transactions on Storage (TOS), 2(2):161--196, May 2006. Google ScholarDigital Library
- Q. Zhu, Z. Chen, L. Tan, Y. Zhou, K. Keeton, and J. Wilkes. Hibernator: Helping Disk Arrays Sleep through the Winter. In Proceedings of the 20th ACM Symposium on Operating Systems Principles (SOSP '05), pages 177--190, Brighton, UK, October 2005. ACM Press. Google ScholarDigital Library
Index Terms
- GreenFS: making enterprise computers greener by protecting them better
Recommendations
GreenFS: making enterprise computers greener by protecting them better
Eurosys '08: Proceedings of the 3rd ACM SIGOPS/EuroSys European Conference on Computer Systems 2008Hard disks contain data - frequently an irreplaceable asset of high monetary and non-monetary value. At the same time, hard disks are mechanical devices that consume power, are noisy, and fragile when their platters are rotating.
In this paper we ...
An economical backup strategy for floppy disks
A variety of data files are preserved on floppy disks. Document files created using a word processor as well as a personal computer are, for instance, stored on floppy disks. Nevertheless, such flies are occasionally lost due to human errors, the life ...
HerpRap: A Hybrid Array Architecture Providing Any Point-in-Time Data Tracking for Datacenter
CLUSTER '12: Proceedings of the 2012 IEEE International Conference on Cluster ComputingBoth physical disk failure and logical errors such as software error, user abuse and virus attacks may cause data lose. The risk of logical errors is far greater than physical disk failure. Moreover, existing RAID solution cannot satisfy the reliability ...
Reviews
Elliot Jaffe
Power consumption and the management of disk failures have been niche topics in computer science over the past ten years. It takes a nontrivial amount of energy to keep a disk spinning. The costs become significant when we consider the number of personal computers, servers, and notebooks currently in use. More critically, disks fail much more frequently than commonly believed. GreenFS is a system that addresses both the reliability of disks and the cost of constant operation. The GreenFS architecture begins with the notion that most enterprise computers, of whatever form factor, are networked. Joukov and Sipek introduce the concept of a reverse backup. Typical systems store backup data on the server and use the local disk for daily operations. In GreenFS, they store the backup on the local disk and use the network for daily operations. This inversion allows the local disk to be spun up only a few times a day to retrieve updates from the active network. In case of disconnected operations, the local disk is spun up and used as the main repository. To provide complete protection and uninterrupted operations, the GreenFS file system uses a local flash disk as an intermediary storage area for items that need to be retained locally, but do not warrant spinning up the local disk. Although necessary, the local flash disk complicates the architecture and its implementation. I enjoyed reading this paper. It is full of good details and presents a novel file system design. Online Computing Reviews Service
Access critical reviews of Computing literature here
Become a reviewer for Computing Reviews.
Comments