Lecture 6: Vector - People @ EECS at UC Berkeley

Lecture 6: Vector - People @ EECS at UC Berkeley

Storage Devices and RAID Professor David A. Patterson Computer Science 252 Fall 2000 DAP Fall .00 UCB 1 Outline Disk Basics Disk History Disk options in 2000 Disk fallacies and performance Tapes RAID DAP Fall .00 UCB 2 Disk Device Terminology Arm Head Actuator Inner Outer

Sector Track Track Platter Several platters, with information recorded magnetically on both surfaces (usually) Bits recorded in tracks, which in turn divided into sectors (e.g., 512 Bytes) Actuator moves head (end of arm,1/surface) over track (seek), select surface, wait for sector rotate under head, then read or write Cylinder: all tracks under heads DAP Fall .00 UCB 3 Photo of Disk Head, Arm, Actuator Spindle Arm { Actuator Head Platters (12) DAP Fall .00 UCB 4 Disk Device Performance

Outer Track Platter Inner Sector Head Arm Controller Spindle Track Actuator Disk Latency = Seek Time + Rotation Time + Transfer Time + Controller Overhead Seek Time? depends no. tracks move arm, seek speed of disk Rotation Time? depends on speed disk rotates, how far sector is from head Transfer Time? depends on data rate (bandwidth) of disk (bit density), size of request DAP Fall .00 UCB 5 Disk Device Performance Average distance sector from head? 1/2 time of a rotation 7200 Revolutions Per Minute 120 Rev/sec 1 revolution = 1/120 sec 8.33 milliseconds 1/2 rotation (revolution) 4.16 ms Average no. tracks move arm? Sum all possible seek distances from all possible tracks / # possible

Assumes average seek distance is random Disk industry standard benchmark DAP Fall .00 UCB 6 Data Rate: Inner vs. Outer Tracks To keep things simple, orginally kept same number of sectors per track Since outer track longer, lower bits per inch Competition decided to keep BPI the same for all tracks (constant bit density) More capacity per disk More of sectors per track towards edge Since disk spins at constant speed, outer tracks have faster data rate Bandwidth outer track 1.7X inner track! DAP Fall .00 UCB 7 Devices: Magnetic Disks Purpose: Track Sector

Long-term, nonvolatile storage Large, inexpensive, slow level in the storage hierarchy Characteristics: Cylinder Seek Time (~8 ms avg) Transfer rate 10-30 MByte/sec Blocks Capacity Head positional latency rotational latency

Platter 7200 RPM = 120 RPS => 8 ms per rev ave rot. latency = 4 ms 128 sectors per track => 0.25 ms per sector 1 KB per sector => 16 MB / s Response time Gigabytes = Queue + Controller + Seek + Rot + Xfer Quadruples every 3 years Service time (aerodynamics) DAP Fall .00 UCB 8 Historical Perspective 1956 IBM Ramac early 1970s Winchester Developed for mainframe computers, proprietary interfaces Steady shrink in form factor: 27 in. to 14 in. 1970s developments 5.25 inch floppy disk formfactor (microcode into mainframe) early emergence of industry standard disk interfaces ST506, SASI, SMD, ESDI Early 1980s PCs and first generation workstations Mid 1980s

Client/server computing Centralized storage on file server accelerates disk downsizing: 8 inch to 5.25 inch Mass market disk drives become a reality industry standards: SCSI, IPI, IDE 5.25 inch drives for standalone PCs, End of proprietary interfaces DAP Fall .00 UCB 9 Disk History Data density Mbit/sq. in. Capacity of Unit Shown Megabytes 1973: 1. 7 Mbit/sq. in 140 MBytes 1979: 7. 7 Mbit/sq. in 2,300 MBytes source: New York Times, 2/23/98, page C3, Makers of disk drives crowd even mroe data into even smaller spaces DAP Fall .00 UCB 10

Historical Perspective Late 1980s/Early 1990s: Laptops, notebooks, (palmtops) 3.5 inch, 2.5 inch, (1.8 inch formfactors) Formfactor plus capacity drives market, not so much performance Recently Bandwidth improving at 40%/ year Challenged by DRAM, flash RAM in PCMCIA cards still expensive, Intel promises but doesnt deliver unattractive MBytes per cubic inch Optical disk fails on performace but finds niche (CD ROM) DAP Fall .00 UCB 11 Disk History 1989: 63 Mbit/sq. in 60,000 MBytes 1997: 1450 Mbit/sq. in 2300 MBytes 1997: 3090 Mbit/sq. in 8100 MBytes

source: New York Times, 2/23/98, page C3, Makers of disk drives crowd even mroe data into even smaller spaces DAP Fall .00 UCB 12 Disk Performance Model /Trends Capacity + 100%/year (2X / 1.0 yrs) Transfer rate (BW) + 40%/year (2X / 2.0 yrs) Rotation + Seek time 8%/ year (1/2 in 10 yrs) MB/$ > 100%/year (2X / <1.5 yrs) Fewer chips + areal density DAP Fall .00 UCB 13 State of the Art: Ultrastar 72ZX Track Sector Cylinder Track Arm Platter Head

Buffer Latency = Queuing Time + Controller time + per access Seek Time + + Rotation Time + per byte Size / Bandwidth { source: www.ibm.com; www.pricewatch.com; 2/14/00 73.4 GB, 3.5 inch disk 2/MB 10,000 RPM; 3 ms = 1/2 rotation 11 platters, 22 surfaces 15,110 cylinders 7 Gbit/sq. in. areal den 17 watts (idle) 0.1 ms controller time 5.3 ms avg. seek 50 to 29 MB/s(internal) DAP Fall .00 UCB 14 Disk Performance Example (will fix later)

Calculate time to read 1 sector (512B) for UltraStar 72 using advertised performance; sector is on outer track Disk latency = average seek time + average rotational delay + transfer time + controller overhead = 5.3 ms + 0.5 * 1/(10000 RPM) + 0.5 KB / (50 MB/s) + 0.15 ms = 5.3 ms + 0.5 /(10000 RPM/(60000ms/M)) + 0.5 KB / (50 KB/ms) + 0.15 ms = 5.3 + 3.0 + 0.10 + 0.15 ms = 8.55 ms DAP Fall .00 UCB 15 Areal Density Bits recorded along a track Metric is Bits Per Inch (BPI) Number of tracks per surface Metric is Tracks Per Inch (TPI) Care about bit density per unit area Metric is Bits Per Square Inch Called Areal Density Areal Density = BPI x TPI DAP Fall .00 UCB 16 Areal Density Areal Density 1973

1.7 100000 1979 7.7 10000 1989 63 1997 3090 1000 2000 17100 Areal Density Year 100 10 1 1970 1980 1990 2000 Year Areal Density = BPI x TPI

Change slope 30%/yr to 60%/yr about 1991 DAP Fall .00 UCB 17 MBits per square inch: DRAM as % of Disk over time 9 v. 22 Mb/si 40% 35% 30% 25% 20% 15% 470 v. 3000 Mb/si 10% 5% 0.2 v. 1.7 Mb/si 0% 1974 1980 1986 1992

1998 source: New York Times, 2/23/98, page C3, Makers of disk drives crowd even mroe data into even smaller spaces DAP Fall .00 UCB 18 Historical Perspective Form factor and capacity drives market, more than performance 1970s: Mainframes 14 inch diameter disks 1980s: Minicomputers, Servers 8, 5.25 diameter disks Late 1980s/Early 1990s: Pizzabox PCs 3.5 inch diameter disks Laptops, notebooks 2.5 inch disks Palmtops didnt use disks, so 1.8 inch diameter disks didnt make it DAP Fall .00 UCB 19 1 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 continue DAP Fall .00 UCB 20 Disk Characteristics in 2000 Seagate IBM IBM 1GB Cheetah Travelstar Microdrive ST173404LC 32GH DJSA - DSCM-11000 Ultra160 SCSI 232 ATA-4 Disk diameter (inches) Formatted data capacity (GB) Cylinders 3.5 2.5 1.0 73.4

32.0 1.0 14,100 21,664 7,167 Disks 12 4 1 Recording Surfaces (Heads) Bytes per sector 24 8 2 512 to 4096

512 512 ~ 424 ~ 360 ~ 140 6.0 14.0 15.2 Avg Sectors per track (512 byte) Max. areal density(Gbit/sq.in.) DAP Fall .00 UCB 21 Disk Characteristics in 2000 Seagate IBM IBM 1GB Cheetah Travelstar

Microdrive ST173404LC 32GH DJSA - DSCM-11000 Ultra160 SCSI 232 ATA-4 Rotation speed (RPM) Avg. seek ms (read/write) Minimum seek ms (read/write) Max. seek ms Data transfer rate MB/second Link speed to buffer MB/s Power idle/operating Watts 10033 5411 3600 5.6/6.2 12.0 12.0

0.6/0.9 2.5 1.0 14.0/15.0 23.0 19.0 27 to 40 11 to 21 2.6 to 4.2 160 67 13 16.4 / 23.5 2.0 / 2.6 0.5 / 0.8

DAP Fall .00 UCB 22 Disk Characteristics in 2000 Seagate IBM IBM 1GB Cheetah Travelstar Microdrive ST173404LC 32GH DJSA - DSCM-11000 Ultra160 SCSI 232 ATA-4 Buffer size in MB 4.0 2.0 0.125 Size: height x width x depth inches Weight pounds 1.6 x 4.0 x 5.8 2.00 0.5 x 2.7 x 0.2 x 1.4 x 3.9

1.7 0.34 0.035 Rated MTTF in powered-on hours 1,200,000 % of POH per month % of POH seeking, reading, writing 100% (300,000?) (20K/5 yr life?) 45% 20% 90% 20% 20% DAP Fall .00 UCB 23 Disk Characteristics in 2000

Seagate IBM Travelstar Cheetah 32GH DJSA ST173404LC 232 ATA-4 Ultra160 SCSI IBM 1GB Microdri DSCM-11000 Load/Unload cycles (disk powered on/off) Nonrecoverable read errors per bits read Seek errors 250 per year 300,000 300,000 <1 per 1015 < 1 per 1013 < 1 per 1013

not available not available Shock tolerance: Operating, Not operating Vibration tolerance: Operating, Not operating (sine swept, 0 to peak) 10 G, 175 G 150 G, 700 G <1 per 10 7 175 G, 1500 G 5-400 Hz @ 5-500 Hz @ 5-500 Hz @ 1G, 1 0.5G, 22-400 1.0G, 2.5-500 500 Hz @ 5G Hz @ 2.0G Hz @ 5.0G DAP Fall .00 UCB 24 Technology Trends Disk Capacity now doubles

every 12 months; before 1990 every 36 motnhs Today: Processing Power Doubles Every 18 months Today: Memory Size Doubles Every 18-24 months(4X/3yr) The TheI/O I/O GAP GAP Today: Disk Capacity Doubles Every 12-18 months Disk Positioning Rate (Seek + Rotate) Doubles Every Ten Years! DAP Fall .00 UCB 25 Fallacy: Use Data Sheet Average Seek Time Manufacturers needed standard for fair comparison (benchmark) Calculate all seeks from all tracks, divide by number of seeks => average Real average would be based on how data laid out on disk, where seek in real applications, then measure performance Usually, tend to seek to tracks nearby, not to random track

Rule of Thumb: observed average seek time is typically about 1/4 to 1/3 of quoted seek time (i.e., 3X4X faster) UltraStar 72 avg. seek: 5.3 ms 1.7 ms DAP Fall .00 UCB 26 Fallacy: Use Data Sheet Transfer Rate Manufacturers quote the speed off the data rate off the surface of the disk Sectors contain an error detection and correction field (can be 20% of sector size) plus sector number as well as data There are gaps between sectors on track Rule of Thumb: disks deliver about 3/4 of internal media rate (1.3X slower) for data For example, UlstraStar 72 quotes 50 to 29 MB/s internal media rate Expect 37 to 22 MB/s user data rate DAP Fall .00 UCB 27 Disk Performance Example Calculate time to read 1 sector for UltraStar 72 again, this time using 1/3 quoted seek time, 3/4 of internal outer track bandwidth; (8.55 ms before) Disk latency = average seek time + average rotational

delay + transfer time + controller overhead = (0.33 * 5.3 ms) + 0.5 * 1/(10000 RPM) + 0.5 KB / (0.75 * 50 MB/s) + 0.15 ms = 1.77 ms + 0.5 /(10000 RPM/(60000ms/M)) + 0.5 KB / (37 KB/ms) + 0.15 ms = 1.73 + 3.0 + 0.14 + 0.15 ms = 5.02 ms DAP Fall .00 UCB 28 Future Disk Size and Performance Continued advance in capacity (60%/yr) and bandwidth (40%/yr) Slow improvement in seek, rotation (8%/yr) Time to read whole disk Year Sequentially Randomly (1 sector/seek) 1990 4 minutes 6 hours 2000 12 minutes 1 week(!) 3.5 form factor make sense in 5-7 yrs? DAP Fall .00 UCB 29 SCSI: Small Computer System Interface Clock rate: 5 MHz / 10 (fast) / 20 (ultra)- 80 MHz (Ultra3)

Width: n = 8 bits / 16 bits (wide); up to n 1 devices to communicate on a bus or string Devices can be slave (target) or master(initiator) SCSI protocol: a series of phases, during which specific actions are taken by the controller and the SCSI disks Bus Free: No device is currently accessing the bus Arbitration: When the SCSI bus goes free, multiple devices may request (arbitrate for) the bus; fixed priority by address Selection: informs the target that it will participate (Reselection if disconnected) Command: the initiator reads the SCSI command bytes from host memory and sends them to the target Data Transfer: data in or out, initiator: target Message Phase: message in or out, initiator: target (identify, save/restore data pointer, disconnect, command complete) Status Phase: target, just before command complete DAP Fall .00 UCB 30 Tape vs. Disk Longitudinal tape uses same technology as hard disk; tracks its density improvements Disk head flies above surface, tape head lies on surface Disk fixed, tape removable Inherent cost-performance based on geometries: fixed rotating platters with gaps (random access, limited area, 1 media / reader) vs. removable long strips wound on spool (sequential access, "unlimited" length, multiple / reader)

New technology trend: Helical Scan (VCR, Camcoder, DAT) Spins head at angle to tape to improve densityDAP Fall .00 UCB 31 Current Drawbacks to Tape Tape wear out: Helical 100s of passes to 1000s for longitudinal Head wear out: 2000 hours for helical Both must be accounted for in economic / reliability model Long rewind, eject, load, spin-up times; not inherent, just no need in marketplace (so far) Designed for archival DAP Fall .00 UCB 32 Automated Cartridge System STC 4400 8 feet 10 feet 6000 x 30 GB D3 tapes = 180 TBytes in 2000 Library of Congress: all information in the world; in 1992, ASCII of all books = 30 TB

DAP Fall .00 UCB 33 Library vs. Storage Getting books today as quaint as the way I learned to program punch cards, batch processing wander thru shelves, anticipatory purchasing Cost $1 per book to check out $30 for a catalogue entry 30% of all books never checked out Write only journals? Digital library can transform campuses Will have lecture on getting electronic information DAP Fall .00 UCB 34 Use Arrays of Small Disks? Katz and Patterson asked in 1987: Can smaller disks be used to close gap in performance between disks and CPUs? Conventional: 4

3.5 5.25 disk designs Low End 10 14 High End Disk Array: 1 disk design 3.5 DAP Fall .00 UCB 35 Advantages of Small Formfactor Disk Drives Low cost/MB High MB/volume High MB/watt Low cost/Actuator Cost and Environmental Efficiencies DAP Fall .00 UCB 36 Replace Small Number of Large Disks with Large Number of Small Disks! (1988 Disks) IBM 3390K IBM 3.5" 0061

x70 Capacity 20 GBytes 320 MBytes 23 GBytes 97 cu. ft. 11 cu. ft. 9X Volume 0.1 cu. ft. 3 KW 1 KW 3X Power 11 W 15 MB/s 120 MB/s 8X Data Rate 1.5 MB/s 600 I/Os/s 3900 IOs/s 6X I/O Rate 55 I/Os/s 250 KHrs ??? Hrs MTTF 50 KHrs $250K $150K Cost $2K Disk Arrays have potential for large data and I/ O rates, high MB per cu. ft., high MB per KW, but what about reliability?

DAP Fall .00 UCB 37 Array Reliability Reliability of N disks = Reliability of 1 Disk N 50,000 Hours 70 disks = 700 hours Disk system MTTF: Drops from 6 years to 1 month! Arrays (without redundancy) too unreliable to be useful! Hot Hot spares spares support support reconstruction reconstruction in in parallel parallel with with access: access: very veryhigh high media media availability availability can can be beachieved achieved DAP Fall .00 UCB 38 Redundant Arrays of (Inexpensive)

Disks Files are "striped" across multiple disks Redundancy yields high data availability Availability: service still provided to user, even if some components failed Disks will still fail Contents reconstructed from data redundantly stored in the array Capacity penalty to store redundant info Bandwidth penalty to update redundant info DAP Fall .00 UCB 39 Redundant Arrays of Inexpensive Disks RAID 1: Disk Mirroring/Shadowing recovery group Each disk is fully duplicated onto its mirror Very high availability can be achieved Bandwidth sacrifice on write: Logical write = two physical writes Reads may be optimized Most expensive solution: 100% capacity overhead (RAID 2 not interesting, so skip) DAP Fall .00 UCB 40

Redundant Array of Inexpensive Disks RAID 3: Parity Disk 10010011 11001101 10010011 ... logical record Striped physical records 1 1 0 1 1 0 0 0 P contains sum of 0 1 other disks per stripe 0 1 mod 2 (parity) 1 0 If disk fails, subtract 1 1 P from sum of other

disks to find missing information P 1 0 1 0 0 0 1 1 1 1 0 0 1 1 0 1 DAP Fall .00 UCB 41 RAID 3 Sum computed across recovery group to protect against hard disk failures, stored in P disk Logically, a single high capacity, high transfer rate disk: good for large transfers Wider arrays reduce capacity costs, but decreases availability 33% capacity cost for parity in this configuration

DAP Fall .00 UCB 42 Inspiration for RAID 4 RAID 3 relies on parity disk to discover errors on Read But every sector has an error detection field Rely on error detection field to catch errors on read, not on the parity disk Allows independent reads to different disks simultaneously DAP Fall .00 UCB 43 Redundant Arrays of Inexpensive Disks RAID 4: High I/O Rate Parity Increasing Insides Insides of of 55 disks disks Example: Example: small small read read D0 D0 & & D5, D5,

large large write write D12-D15 D12-D15 D0 D1 D2 D3 P D4 D5 D6 D7 P D8 D9

D10 D11 P D12 D13 D14 D15 P D16 D17 D18 D19 P D20 D21

D22 D23 P . . . . . . . . Disk . . . . . Columns . . Logical Disk Address

Stripe DAP Fall .00 UCB 44 Inspiration for RAID 5 RAID 4 works well for small reads Small writes (write to one disk): Option 1: read other data disks, create new sum and write to Parity Disk Option 2: since P has old sum, compare old data to new data, add the difference to P Small writes are limited by Parity Disk: Write to D0, D5 both also write to P disk D0 D1 D2 D3 P D4 D5 D6

D7 P DAP Fall .00 UCB 45 Redundant Arrays of Inexpensive Disks RAID 5: High I/O Rate Interleaved Parity Independent Independent writes writes possible possible because because of of interleaved interleaved parity parity Example: write to D0, D5 uses disks 0, 1, 3, 4 D0

D1 D2 D3 P D4 D5 D6 P D7 D8 D9 P D10 D11 D12

P D13 D14 D15 P D16 D17 D18 D19 D20 D21 D22 D23 . . .

. . . P . . . . . . . Disk Columns . . Increasing Logical Disk Addresses DAP Fall .00 UCB 46 Problems of Disk Arrays: Small Writes RAID-5: Small Write Algorithm 1 Logical Write = 2 Physical Reads + 2 Physical Writes D0' new

data D0 D1 D2 D3 old data (1. Read) P old (2. Read) parity + XOR + XOR (3. Write) D0' D1 (4. Write) D2 D3

P' DAP Fall .00 UCB 47 System Availability: Orthogonal RAIDs Array Controller String Controller . . . String Controller . . . String Controller . . . String Controller . . .

String Controller . . . String Controller . . . Data Recovery Group: unit of data redundancy Redundant Support Components: fans, power supplies, controller, cables End to End Data Integrity: internal parity protected data paths DAP Fall .00 UCB 48 System-Level Availability host host Fully dual redundant I/O Controller Array Controller I/O Controller Array Controller

... ... ... ... Goal: Goal:No NoSingle Single Points Pointsof of Failure Failure ... Recovery Group . . . with duplicated paths, higher performance can be obtained when there are no failures DAP Fall .00 UCB 49

Summary: Redundant Arrays of Disks (RAID) Techniques Disk Mirroring, Shadowing (RAID 1) Each disk is fully duplicated onto its "shadow" Logical write = two physical writes 100% capacity overhead Parity Data Bandwidth Array (RAID 3) Parity computed horizontally Logically a single high data bw disk High I/O Rate Parity Array (RAID 5) 1 0 0 1 0 0 1 1 1 0 0 1 0 0 1 1 1

0 0 1 0 0 1 1 1 1 0 0 1 1 0 1 1 0 0 1 0 0 1 1 0 0 1 1 0

0 1 0 Interleaved parity blocks Independent reads and writes Logical write = 2 reads + 2 writes Parity + Reed-Solomon codes DAP Fall .00 UCB 50 Berkeley History: RAID-I RAID-I (1989) Consisted of a Sun 4/280 workstation with 128 MB of DRAM, four dual-string SCSI controllers, 28 5.25-inch SCSI disks and specialized disk striping software Today RAID is $19 billion dollar industry, 80% nonPC disks sold in RAIDs DAP Fall .00 UCB 51

Recently Viewed Presentations

  • Chapter 4: The Sounds of American English

    Chapter 4: The Sounds of American English

    Plosives =sounds you make by blocking off the breath stream entirely for a short period of time, just long enough to build up some air pressure behind your articulators. You then "explode" this air to produce the sound. E.g. "pet"...
  • Introduction to World Regional Geography

    Introduction to World Regional Geography

    Regional Geography Study of Geography Evolution of Geography Contributions of the Greeks/Romans geo (the earth) graphos (to write about/describe) Herodotus-"Father of Geography" who explained the physical and human geography of his day
  • CMO PREPARAR Y GESTIONAR LAS EVALUACIONES INTERMEDIAS DE

    CMO PREPARAR Y GESTIONAR LAS EVALUACIONES INTERMEDIAS DE

    La MTE debe realizarse en la mitad del período de implementación (presentación del informe de MTE dentro de 36 meses de la asignación) No es necesario que la MTE coincida con los plazos estacionales de la línea de base o...
  • Chapter 12 Physical Geography of Europe: The Peninsula of ...

    Chapter 12 Physical Geography of Europe: The Peninsula of ...

    Chapter 12Physical Geography of Europe: The Peninsula of Peninsulas. Because of its unique geography and weather patterns, Europe's landscapes, waterways, and climates vary greatly.
  • MEKARN Programme 2001-2011

    MEKARN Programme 2001-2011

    All research done in home country on practical, farmer-related issues. Thesis(130 credits) normally includes 4 published or submitted articles. Achievements? 61 students from the 4 countries received MSc degree from SLU! 32% Cambodia. 24% Lao PDR.
  • Vaccins - الموقع الأول للدراسة في ...

    Vaccins - الموقع الأول للدراسة في ...

    Les vecteurs virus de la vaccine, les adenovirus 2 et 5 et virus des insectes comme les Baculovirus. Les protéines recombinantes ont l'avantage d'êtres produites avec la majorité des modifications post-traductionnelles nécessaires à leur rôle fonctionnel.
  • Un mondo … matematico - San Giuseppe Lugo

    Un mondo … matematico - San Giuseppe Lugo

    Leonardo Pisano detto Leonardo Fibonacci, perché filius del Bonacci (Pisa, 1170 - 1240) fu un matematico italiano. LEONARDO FIBONACCI. L'intento di Fibonacci era quello di trovare una legge matematica che potesse descrivere . la crescita di una popolazione di conigli
  • You Spoke, We Listened Pella provides contractors the

    You Spoke, We Listened Pella provides contractors the

    Lead-Gen Sources. Digital Quotes - PQM. Simplified Price Books. Most Competitive Consumer Financing Options. Putting Tools and Programs to Work to Find and Close More Sales " Pella is really upping their game where it comes to supporting their certified...