Disks and FilesWhy Not Store Everything in Main Memory?Components of a DiskAccessing a Disk PageState of the Art: Barracuda 180Disk Performance1 inch disk drive!Disk Characteristics in 2000Slide 9Slide 10Slide 111COSC 3480: DisksDisks and Files •DBMS stores information on (“hard”) disks.•This has major implications for DBMS design!»READ: transfer data from disk to main memory (RAM).»WRITE: transfer data from RAM to disk.»Both are high-cost operations, relative to in-memory operations, so must be planned carefully!2COSC 3480: DisksWhy Not Store Everything in Main Memory?•Costs too much. $100 will buy you either 512MB of RAM or 50GB of disk today --- that is disk storage 100 times cheaper (but it is approx. 10000 times slower).•Main memory is volatile. We want data to be saved between runs. (Obviously!)•Typical storage hierarchy:»Main memory (RAM) for currently used data.»Disk for the main database (secondary storage).»Tapes for archiving older versions of the data (tertiary storage).3COSC 3480: DisksComponents of a Disk Platters The platters spin (say, 90rps).Spindle The arm assembly is moved in or out to position a head on a desired track. Tracks under heads make a cylinder (imaginary!).Disk headArm movementArm assembly Only one head reads/writes at any one time.TracksSector Block size is a multiple of sector size (which is fixed).4COSC 3480: DisksAccessing a Disk Page•Time to access (read/write) a disk block:»seek time (moving arms to position disk head on track)»rotational delay (waiting for block to rotate under head)»transfer time (actually moving data to/from disk surface)•Seek time and rotational delay dominate.»Seek time varies from about 1 to 20msec»Rotational delay varies from 0 to 10msec»Transfer rate is about 1msec per 32KB page5COSC 3480: DisksState of the Art: Barracuda 180»181.6 GB, 3.5 inch disk»12 platters, 24 surfaces»24,247 cylinders»7,200 RPM; (4.2 ms avg. latency)»7.4/8.2 ms avg. seek (r/w)»64 to 35 MB/s (internal)»0.1 ms controller time»10.3 watts (idle)source: www.seagate.comLatency = Queuing Time + Controller time +Seek Time + Rotation Time + Size / Bandwidthper accessper byte{+SectorTrackCylinderHeadPlatterArmTrack Buffer6COSC 3480: DisksDisk Performance•Calculate time to read 64 KB (128 sectors) for Barracuda 180 X using advertised performance; sector is on outer trackDisk latency = average seek time + average rotational delay + transfer time + controller overhead = 7.4 ms + 0.5 * 1/(7200 RPM) + 64 KB / (65 MB/s) + 0.1 ms = 7.4 ms + 0.5 /(7200 RPM/(60000ms/M)) + 64 KB / (65 KB/ms) + 0.1 ms = 7.4 + 4.2 + 1.0 + 0.1 ms = 12.7 ms7COSC 3480: Disks1 inch disk drive!•2000 IBM MicroDrive:» 1.7” x 1.4” x 0.2” »1 GB, 3600 RPM, 5 MB/s, 15 ms seek»Digital camera, PalmPC?•2006 MicroDrive?•9 GB, 50 MB/s! »Assuming it finds a niche in a successful product»Assuming past trends continue8COSC 3480: DisksDisk Characteristics in 2000SeagateCheetahST173404LCUltra160 SCSIIBMTravelstar32GH DJSA -232 ATA-4IBM 1GBMicrodriveDSCM-11000Disk diameter(inches)3.5 2.5 1.0Formatted datacapacity (GB)73.4 32.0 1.0Cylinders14,100 21,664 7,167Disks12 4 1RecordingSurfaces (Heads)24 8 2Bytes per sector512 to 4096 512 512Avg Sectors pertrack (512 byte)~ 424 ~ 360 ~ 140Max. arealdensity(Gbit/sq.in.)6.0 14.0 15.2$447 $435$8289COSC 3480: DisksDisk Characteristics in 2000SeagateCheetahST173404LCUltra160 SCSIIBMTravelstar32GH DJSA -232 ATA-4IBM 1GBMicrodriveDSCM-11000Rotation speed(RPM)10033 5411 3600Avg. seek ms(read/write)5.6/6.2 12.0 12.0Minimum seekms (read/write)0.6/0.9 2.5 1.0Max. seek ms14.0/15.0 23.0 19.0Data transferrate MB/second27 to 40 11 to 21 2.6 to 4.2Link speed tobuffer MB/s160 67 13Poweridle/operatingWatts16.4 / 23.5 2.0 / 2.6 0.5 / 0.810COSC 3480: DisksDisk Characteristics in 2000SeagateCheetahST173404LCUltra160 SCSIIBMTravelstar32GH DJSA -232 ATA-4IBM 1GBMicrodriveDSCM-11000Buffer size in MB4.0 2.0 0.125Size: height xwidth x depthinches1.6 x 4.0 x5.80.5 x 2.7 x3.90.2 x 1.4 x1.7Weight pounds2.00 0.34 0.035Rated MTTF inpowered-on hours1,200,000 (300,000?) (20K/5 yrlife?)% of POH permonth100% 45% 20%% of POHseeking, reading,writing90% 20% 20%11COSC 3480: DisksDisk Characteristics in 2000SeagateCheetahST173404LCUltra160 SCSIIBM Travelstar32GH DJSA -232 ATA-4IBM 1GB MicrodriveDSCM-11000Load/Unloadcycles (diskpowered on/off)250 per year 300,000 300,000Nonrecoverableread errors perbits read<1 per 1015< 1 per 1013< 1 per 1013Seek errors<1 per 107not available not availableShock tolerance:Operating, Notoperating10 G, 175 G 150 G, 700 G 175 G, 1500 GVibrationtolerance:Operating, Notoperating (sineswept, 0 to peak)5-400 Hz @0.5G, 22-400Hz @ 2.0G5-500 Hz @1.0G, 2.5-500Hz @ 5.0G5-500 Hz @ 1G, 10-500 Hz @