algorithms and methods for distributed storage networks
play

Algorithms and Methods for Distributed Storage Networks 2. Hard - PowerPoint PPT Presentation

Algorithms and Methods for Distributed Storage Networks 2. Hard Disks Christian Schindelhauer University of Freiburg Technical Faculty Computer Networks and Telematics Winter Semester 2011/12 Mittwoch, 26. Oktober 11 Hard Disks History


  1. Algorithms and Methods for Distributed Storage Networks 2. Hard Disks Christian Schindelhauer University of Freiburg Technical Faculty Computer Networks and Telematics Winter Semester 2011/12 Mittwoch, 26. Oktober 11

  2. Hard Disks ‣ History ‣ Interfaces • Capacity and Access Speed • ATA, SATA • Prices • SCSI, SAS • Form factors • Fibre-Channel ‣ Construction and Operation • eSATA ‣ Lifetime and Disk Failures • Mechanics • Storage technology • Error Management and Recovery ‣ Low-Level Data Structures • Types • Encoding, Decoding • S.M.A.R.T. • Counter methods • Tracks , Cylinders ‣ Special Issues • LBA • Sound avoidance • Data security Distributed Storage Networks Computer Networks and Telematics 2 and Computer Forensics University of Freiburg Winter 2011/12 Christian Schindelhauer Mittwoch, 26. Oktober 11

  3. Hard Disks History Distributed Storage Networks Computer Networks and Telematics 3 and Computer Forensics University of Freiburg Winter 2011/12 Christian Schindelhauer Mittwoch, 26. Oktober 11

  4. Evolution of Hard Disk Capacity Distributed Storage Networks Computer Networks and Telematics 4 and Computer Forensics University of Freiburg Winter 2011/12 Christian Schindelhauer Mittwoch, 26. Oktober 11

  5. History ‣ 1956 IBM invents 305 RAMAC ‣ 1981 Sony ships first 3.25 in floppy (Random Access Method of drive Accounting and Control) ‣ 1983 Rodime produces 3.25 in disk drive • 5 MBytes, 24 in ‣ 1961 IBM invents air bearing heads ‣ 1986 Conner introduces first 3.25 in voice coil actuators ‣ 1970 IBM invents 8 in floppy disk drives ‣ 1997 Seagate introduces 7,200 RPM Ultra hard disk ‣ 1973 IBM ships 3340 Winchester sealed hard drives ‣ 1996 Fujitsu introduse aero dynamic design for lower flighing heads • 30 MBytes ‣ 1999 IBM develops the smallest hard ‣ 1980 Seagate introduces 5.25 in hard disk of the World 1in (340 MB) disk drive ‣ 2007 Hitachi introduces 1 TB hard • 5 MBytes disk Distributed Storage Networks Computer Networks and Telematics 5 and Computer Forensics University of Freiburg Winter 2011/12 Christian Schindelhauer Mittwoch, 26. Oktober 11

  6. History of Hard Disk Prices Figure 6 Cost of storage at the disk drive and system level 1000 SYSTEMS W/LARGE SYSTEMS W/SMALL FF DRIVES FF DRIVES 100 10 PRICE/MBYTE, DOLLARS 1 INDUSTRY PROJECTION FROM 0.1 MULTIPLE SOURCES HIGH 0.01 PERFOR- MANCE, SERVER HDD DESKTOP Technological impact of magnetic 0.001 HDD hard disk drives on storage systems, SMALL FF HDD STORAGE SYSTEM Grochowski, R. D. Halem 0.0001 1980 1985 1990 1995 2000 2005 2010 IBM SYSTEMS JOURNAL, VOL 42, NO 2, 2003 PRODUCTION YEAR Figure 7 Cost of storage for disk drive, paper, film, and semiconductor memory Distributed Storage Networks Computer Networks and Telematics 6 and Computer Forensics University of Freiburg Winter 2011/12 Christian Schindelhauer Mittwoch, 26. Oktober 11

  7. Hard Disks Construction and Operation Distributed Storage Networks Computer Networks and Telematics 7 and Computer Forensics University of Freiburg Winter 2011/12 Christian Schindelhauer Mittwoch, 26. Oktober 11

  8. Construction of a Hard Disk (c) Western Digital Corporation Distributed Storage Networks Computer Networks and Telematics 8 and Computer Forensics University of Freiburg Winter 2011/12 Christian Schindelhauer Mittwoch, 26. Oktober 11

  9. Construction of a Hard Disk (c) Seagate Technology Distributed Storage Networks Computer Networks and Telematics 9 and Computer Forensics University of Freiburg Winter 2011/12 Christian Schindelhauer Mittwoch, 26. Oktober 11

  10. Physical Components ✴ servo system dynamicall ‣ Platters positions the heads directly • round flat disks with special material over the data tracks to store magnetic patterns - Head arms • stacked onto a spindle ✴ are moved by the actuator to • rotate at high speed choose the tracks ‣ Read/Write Devices - Head sliders • usually two per platter ✴ are responsible to keep the • Actuator heads in a small defined - old: stepper motor distance above the platter ✴ mechanic adjusts to discrete ✴ heads „fly“ over the platter on positions an air cushion ✴ low track density - Read/write heads mounted on top ✴ still used in floppy disks of arms - now: voice coil actuator Distributed Storage Networks Computer Networks and Telematics 10 and Computer Forensics University of Freiburg Winter 2011/12 Christian Schindelhauer Mittwoch, 26. Oktober 11

  11. Slider Figure 6. Illustration of suspension and slider. Left: schematic. Right: photograph. (Source: Tom Albrecht, IBM) Proceedings of the American Control Conference ,Arlington, VA June 25-27, 2001 A Tutorial on Controls for Disk Drives William Messner , Rick Ehrlich Distributed Storage Networks Computer Networks and Telematics 11 and Computer Forensics University of Freiburg coil Winter 2011/12 Christian Schindelhauer MR element / inductive head Mittwoch, 26. Oktober 11 Figure 7. Schematic of readwrite transducer. (Source: Tom Albrecht, IBM) 4 1 6

  12. Magnetization Techniques ‣ Longitudinal recording • magnetic moments in the direction of rotation • problem: super-paramagnetic effect • 100-200 Gigabit per square inch ‣ Perpendicular • magnetic moments are orthogonal to the rotation direction • increases the data density • 1 Terabit per square inch ‣ HAMR (Heat Assisted Magnetic Recording) • upcoming technology • Laser heats up area to keep the necessary magnetic field as small as possible Distributed Storage Networks Computer Networks and Telematics 12 and Computer Forensics University of Freiburg Winter 2011/12 Christian Schindelhauer Mittwoch, 26. Oktober 11

  13. Electronic Components ‣ Magnetized Surface on platter ‣ Read/Write-Head ‣ Embedded controller ‣ Disk buffer (disk cache) • store bits going to and from the platter • read-ahead/read-behind • speed matching • write acceleration • command queueing ‣ Interface Distributed Storage Networks Computer Networks and Telematics 13 and Computer Forensics University of Freiburg Winter 2011/12 Christian Schindelhauer Mittwoch, 26. Oktober 11

  14. Hard Disks Low Level Data Structure Distributed Storage Networks Computer Networks and Telematics 14 and Computer Forensics University of Freiburg Winter 2011/12 Christian Schindelhauer Mittwoch, 26. Oktober 11

  15. Tracks and Cylinders ‣ Tracks • is a circle with data on a platter ‣ Cylinder • is the set of tracks on all platters that are simultaneously accessed by the heads ‣ Sector • basic unit of data storage • angular section of a circle (c) Quantum Corporation Distributed Storage Networks Computer Networks and Telematics 15 and Computer Forensics University of Freiburg Winter 2011/12 Christian Schindelhauer Mittwoch, 26. Oktober 11

  16. Addressing ‣ CHS (cylinder, head, sector) • each logical unit is addressed by the cylinder - set of corresponding tracks on both sides of the platters • head • sector (angular section) • old system ‣ LBA (Logical Block Addressing) • simpler system all logical blocks are number • the translation to CHS is Distributed Storage Networks Computer Networks and Telematics 16 and Computer Forensics University of Freiburg Winter 2011/12 Christian Schindelhauer Mittwoch, 26. Oktober 11

  17. Adapting Sectors ‣ Zoned bit recording • adapt the sector size to the bit density • different number of sectors depending from the distance from the center ‣ Sector interleaving • for cylinder switch • when the arm moves then the disk continues spinning • to avoid waiting times the numbering of the sectors has an offset http://www.storagereview.com/guide2000/ref/hdd/geom/ tracksZBR.htm l Distributed Storage Networks Computer Networks and Telematics 17 and Computer Forensics University of Freiburg Winter 2011/12 Christian Schindelhauer Mittwoch, 26. Oktober 11

  18. Sector Format ‣ A sector is the atomic data unit of an hard disk ‣ No absolute position • must be identified from its contents ‣ Contents • ID Information (number and location) • Synchronization fields • Data • ECC: Error correcting codes • Gaps ‣ Specific contents varies from hard disk type Distributed Storage Networks Computer Networks and Telematics 18 and Computer Forensics University of Freiburg Winter 2011/12 Christian Schindelhauer Mittwoch, 26. Oktober 11

  19. Formatting ‣ Low-level formatting • creates the physical structures (tracks, sectors, control information) - starts from empty platter - map out bad sectors ‣ Partitioning • divides the disk into logical pieces (i.e. hard disk volumes) ‣ High-level formatting • logical structures for the operating-system level components Distributed Storage Networks Computer Networks and Telematics 19 and Computer Forensics University of Freiburg Winter 2011/12 Christian Schindelhauer Mittwoch, 26. Oktober 11

  20. Encoding ‣ Problem • Only the difference of orientation can be measured • Because of the para-magnetic effect orientation changes need a minimum distance • Long sequences of same orientation lead to errors ‣ Encoding • must have long, but not too long flux reversals Distributed Storage Networks Computer Networks and Telematics 20 and Computer Forensics University of Freiburg Winter 2011/12 Christian Schindelhauer Mittwoch, 26. Oktober 11

Recommend


More recommend