David A. Patterson
Garth Gibson
Randy H. Katz
Abstract
Increasing performance of CPUs and memories will be squandered if not matched by a similar performance increase in I/O. While the capacity of Single Large Expensive Disks (SLED) has grown rapidly, the performance improvement of SLED has been modest. Redundant Arrays of Inexpensive Disks (RAID), based on the magnetic disk technology developed for personal computers, offers an attractive alternative to SLED, promising improvements of an order of magnitude in performance, reliability, power consumption, and scalability. This paper introduces five levels of RAIDs, giving their relative cost/performance, and compares RAID to an IBM 3380 and a Fujitsu Super Eagle.
- Bell 84 C G Bell, "The Mml and Micro Industries,' IEEE Computer Vol 17 No 10(October 1984),pp 14-30 Google Scholar
Digital Library
- Joy 85 B Jo~ prc~entauon at ISSCC '85 panel session, Feb 1985Google Scholar
- Siewiorek 82 D P Ssea~orek, C G Bell, and A Ne~ell, Computer Strm tl~rc ~ Prm, lple~ and Example~, p 46Google Scholar
- Moore 75 G E Moore, "Progress sn Digital Integrated Electromcs," Proc IEEE Dtgttal Integrated Electrom6 Device Meeting, (1975), p 11Google Scholar
- Myers 86 G J Myers A Y C Yu, and D L House, "Microprocessor Technology Trend~ " Proc IEEE, Vol 74, no 12, (December 1986), pp 1605-1622Google ScholarCross Ref
- Garcia 84 H Garc~a Mohna, R Culhngford, P Honeyman, R Lipton, "The Case for Massive Memory," Techmeal Report 326, Dept of EE and CS, Pnnceton Umv, May 1984Google Scholar
- Myers 86 W Myers, "The Compettuveness of the Umted States Disk Industry," IEEE Computer, Vol 19, No 11 (January 1986), pp 85-90Google Scholar
- Frank 87 P D Frank, "Advances m Head Technology," presentaUon at Challenges m Dzsk Technology She, t Course, Institute for Informauon Storage Technology, Santa Clara Untvers~ty, Santa Clara, Cahfomxa, ~mber 15-17, 1987Google Scholar
- Stevens 81 L D Stevens, "The Evoluuon of Magnetic Storage," IBM Journal of Research and Development, Voi 25, No 5, Sept 1981, pp 663-675Google ScholarDigital Library
- Harker 81 J M Harker et al, "A Quarter Century of Disk File Innovataon," tb~d, pp 677-689 Google ScholarDigital Library
- Amdahl 67 G M Amdahl, "Vahdaty of the single processor approach to achieving large scale computing eapabd~ttes," Procee&ngs AFIPS 1967 Spnng Joint Computer Conference Vol 30 (Aflantte City, New Jersey April 1967), pp 483-485Google Scholar
- Boral 83 H Boral and D J DeWltt, "Database Machines An Ideas Whose Ttme Has Passed9 A Cnttque of the Future of Database Machines," Proc International Conf on Database Mactunes, Edited by H -O Lelheh and M Mtsskoff, Spnnger-Verlag, Berhn, 1983Google Scholar
- IBM 87 "IBM 3380 Dtreet Access Storage Introdueuon," IBM GC 26-4491-0, September 1987Google Scholar
- Gawlick 87 D Gawhek, private eommumeatton, Nov, 1987Google Scholar
- Fujitsu 87 "M2361A Mm~-Dssk Drive Englneenng Speelfleauons," (revased) Feb, 1987, B03P-4825-0001AGoogle Scholar
- Adaptec 87 AIC-6250, IC Product Grade, Adaptec, stock # DB0003-00 rev B, 1987, p 46Google Scholar
- Livny 87 Llvny, M, S Khoshahan, H Boral, "Multi-disk management algorithms," ~roc ofACM SIGMETRICS, May 1987 Google ScholarDigital Library
- Kim 86 M Y Kim, "Synehromzed desk interleaving," IEEE Trans on Computers, vol C-35, no 11, Nov 1986 Google ScholarDigital Library
- Salem 86 K Salem and Garcm-Mohna, H, "Dssk Striping," IEEE 1986 lnt Con/on Data Engmeenng, 1986 Google ScholarDigital Library
- Bitton 88 D Bitton and J Gray, "Desk Shadowing," m press, 1988Google Scholar
- Kurzweil 88 F Kurzwell, "Small Dask Arrays - The Emerging Approach to H~gh Performance," presentauon at Spnng COMPCON 88, March 1, 1988, San Francisco, CAGoogle Scholar
- Hamming 50 R W Hamming, "Error Detecting and Correcting Codes," The Bell System Techmcal Journal, Vol XXVI, No 2 (April 1950), pp 147-160Google ScholarCross Ref
- Hillis 87 D Htlhs, private eommumeauon, October, 1987Google Scholar
- Park 86 A Park and K Balasubramaman, "Provldmg Fault Tolerance m Parallel Secondary Storage Systems," Department of Computer Science, Prmeeton Omversaty, CS.TR-057-86, Nov 7, 1986Google Scholar
- Maginnis 87 N B Magmms, "Store More, Spend Less Meal-range Opttons Abound,"Computerworld, Nov 16, 1987, p 71Google Scholar
- Denning 78 Pj Denning and D P Slutz, "Generahzed Workang Sets for Segment Reference Strings," CACM, vol 21, no 9, (Sept. 1978) pp 750-759 Google ScholarDigital Library
- Bell 85 Bell, C G, "Mult~s a new class of muitzproeessor computers,"Sc~ence, voL 228 (Alm126, 1985) 462-467Google Scholar
Index Terms
- A case for redundant arrays of inexpensive disks (RAID)
Recommendations
A case for redundant arrays of inexpensive disks (RAID)
SIGMOD '88: Proceedings of the 1988 ACM SIGMOD international conference on Management of dataIncreasing performance of CPUs and memories will be squandered if not matched by a similar performance increase in I/O. While the capacity of Single Large Expensive Disks (SLED) has grown rapidly, the performance improvement of SLED has been modest. ...
Redundant array of inexpensive disks (RAID)
Encyclopedia of Computer ScienceRAID (Redundant Array of Inexpensive [or Independent] Disks) is an architectural concept developed to turn relatively slow and inexpensive hard disks into fast, large-capacity, and more reliable storage systems. The RAID concept was introduced by a team ...
Comments