Unformatted text preview:

The Big Picture Where are We Now Today s Topic I O Systems CS152 Computer Architecture and Engineering Lecture 24 Network Bus Processor Processor Input Input Control I O Systems II Control Memory Datapath Memory Output Datapath Output November 24 1999 John Kubiatowicz http cs berkeley edu kubitron lecture slides http www inst eecs berkeley edu cs152 11 24 99 UCB Fall 1999 CS152 Kubiatowicz Lec24 1 Outline of Today s Lecture 11 24 99 CS152 Kubiatowicz Lec24 2 UCB Fall 1999 Organization of a Hard Magnetic Disk Basic behavior of disks Platters Queueing theory Interfacing between processor and I O devices Track RAID disk arrays Summary Sector Typical numbers depending on the disk size 500 to 2 000 tracks per surface 32 to 128 sectors per track A sector is the smallest unit that can be read or written 11 24 99 UCB Fall 1999 CS152 Kubiatowicz Lec24 3 Traditionally all tracks have the same number of sectors Constant bit density record more sectors on the outer tracks Recently relaxed constant bit size speed varies with track location CS152 Kubiatowicz 11 24 99 UCB Fall 1999 Lec24 4 Magnetic Disk Characteristic Typical Numbers of a Magnetic Disk Track Sector Cylinder all the tacks under the head at a given point on all surface Cylinder Read write data is a three stage process Seek time position the arm over the proper track Rotational latency wait for the desired sector to rotate under the read write head Transfer time transfer a block of bits sector under the read write head Head Platter Average seek time as reported by the industry Typically in the range of 8 ms to 12 ms Sum of the time for all possible seek total of possible seeks Track Sector Rotational Latency Most disks rotate at 3 600 to 7200 RPM Approximately 16 ms to 8 ms per revolution respectively An average latency to the desired information is halfway around the disk 8 ms at 3600 RPM 4 ms at 7200 RPM Cylinder Platter Head Transfer Time is a function of Transfer size usually a sector 1 KB sector Rotation speed 3600 RPM to 10000 RPM Recording density bits per inch on a track Diameter typical diameter ranges from 2 5 to 5 25 in Typical values 2 to 40 MB per second Due to locality of disk reference actual average seek time may Only be 25 to 33 of the advertised number 11 24 99 CS152 Kubiatowicz Lec24 5 UCB Fall 1999 11 24 99 UCB Fall 1999 CS152 Kubiatowicz Lec24 6 Review Disk Device Terminology Disk I O Performance Request Rate Service Rate Disk Controller Disk Disk Controller Disk Queue Processor Queue Disk Access Time Seek time Rotational Latency Transfer time Controller Time Queueing Delay Estimating Queue Length Utilization U Request Rate Service Rate Disk Latency Queueing Time Controller time Seek Time Rotation Time Xfer Time Order of magnitude times for 4K byte transfers Average Seek 8 ms or less Mean Queue Length U 1 U As Request Rate Service Rate Mean Queue Length Infinity 11 24 99 UCB Fall 1999 Rotate 4 2 ms 7200 rpm Xfer 1 ms 7200 rpm CS152 Kubiatowicz Lec24 7 11 24 99 UCB Fall 1999 CS152 Kubiatowicz Lec24 8 Example Reliability and Availability 512 byte sector rotate at 5400 RPM advertised seeks is 12 ms transfer rate is 4 MB sec controller overhead is 1 ms queue idle so no service time Disk Access Time Seek time Rotational Latency Transfer time Controller Time Queueing Delay Two terms that are often confused Reliability Is anything broken Availability Is the system still available to the user Disk Access Time 12 ms 0 5 5400 RPM 0 5 KB 4 MB s 1 ms 0 Availability can be improved by adding hardware Example adding ECC on memory Disk Access Time 12 ms 0 5 90 RPS 0 125 1024 s 1 ms 0 Reliability can only be improved by Better environmental conditions Building more reliable components Building with fewer components Disk Access Time 12 ms 5 5 ms 0 1 ms 1 ms 0 ms Disk Access Time 18 6 ms If real seeks are 1 3 advertised seeks then its 10 6 ms with rotation delay at 50 of the time 11 24 99 UCB Fall 1999 CS152 Kubiatowicz Lec24 9 Queue 11 24 99 Server 300 Metrics Response Time Throughput 100 0 100 0 Throughput total BW Response time Begins when a task is placed in the queue Queue Proc Ends when it is completed by the server In order to minimize the response time The queue should be empty Response Time ms 200 Throughput The number of tasks completed by the server in unit time In order to get the highest possible throughput The server should never be idle The queue should never be empty 11 24 99 CS152 Kubiatowicz Lec24 10 UCB Fall 1999 Disk I O Performance Simple Producer Server Model Producer Improve availability may come at the cost of lower reliability IOC Device Response time Queue Device Service time The server will be idle UCB Fall 1999 CS152 Kubiatowicz Lec24 11 11 24 99 UCB Fall 1999 CS152 Kubiatowicz Lec24 12 Response Time vs Productivity Response Time Productivity conventional 0 3s Interactive environments Each interaction or transaction has 3 parts Entry Time time for user to enter command conventional 1 0s System Response Time time between user entry system replies graphics 0 3s Think Time Time from response until user begins next command 1st transaction entry resp think graphics 1 0s 2nd transaction 0 00 What happens to transaction time as shrink system response time from 1 0 sec to 0 3 sec With Keyboard 4 0 sec entry 9 4 sec think time With Graphics 0 25 sec entry 1 6 sec think time 5 00 10 00 15 00 Time 0 7sec off response saves 4 9 sec 34 and 2 0 sec 70 total time per transaction greater productivity Another study everyone gets more done with faster response but novice with fast response expert with slow 11 24 99 UCB Fall 1999 CS152 Kubiatowicz Lec24 13 CS152 Kubiatowicz Lec24 14 UCB Fall 1999 Computers in the News Electronic Ink Administrivia Get your project entered on web page Electronic Ink I will remove the ability to enter new projects soon won t be able to submit final report Please take time to enter reasonable title no trademarked names Should have two or three sentence description Pending schedule 11 24 99 Little capsules with charged balls that are half black half white Placing an electronic charge of one polarity makes dot black and the other polarity makes it white Flexible cheap paper like displays Monday 11 29 no class work on projects Monday 11 29 Update on design due to TAs Wednesday 12 1 Last class wrap up evaluations etc Monday 12 6 Oral reports 10 12am and 2 4pm Signup sheet will be on my office door next week Project reports must be submitted via web by 5pm on 12 6 Friday 12 10 Grades ready


View Full Document

Berkeley COMPSCI 152 - Lecture 24 I/O Systems II

Documents in this Course
Quiz 5

Quiz 5

9 pages

Memory

Memory

29 pages

Quiz 5

Quiz 5

15 pages

Memory

Memory

29 pages

Memory

Memory

35 pages

Memory

Memory

15 pages

Quiz

Quiz

6 pages

Midterm 1

Midterm 1

20 pages

Quiz

Quiz

12 pages

Memory

Memory

33 pages

Quiz

Quiz

6 pages

Homework

Homework

19 pages

Quiz

Quiz

5 pages

Memory

Memory

15 pages

Load more
Loading Unlocking...
Login

Join to view Lecture 24 I/O Systems II and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view Lecture 24 I/O Systems II and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?