Disc Scheduling

1

CS414: Operating Systems

Agenda

•Last time (This morning)�I/O systems + Mass storage structure�PA#4 back

•This time (Thurs night / Friday afternoon)�Mass storage structure�Security introduction

•Next time (Tuesday Nov 28)�More security/protection�PA#7 out (look for it early)

•Remember, no class next week •Remember, PA#6 due NEXT Friday at noon! DO it soon!!!


ScheduleSun Tues Thurs Fri

No class – martyat conference

File systemsPA#5 duePA#6 out 7 9

14 16

21 23

protection,security

28 30

protection, securityPA#7 out

5

Wrap-upPA#7 due

Mass storage

protection, security

protection,security 10

No class – martyat conference

No class –Thanksgiving

No class –Thanksgiving

14Final exam0900-noon

PA#6 due(noon)


Before we start: PA#6 (Main)

•Datafile replaced today at noon or so!


Optimizing Disk Accesses

•In timesharing systems, disk driver may have many different requests pending

•Approach: minimize disk seek time•Some strategies are implemented in the disk driver, while others are implemented directly in the disk control ler

•We'll look at �FCFS, SSTF, SCAN, C-SCAN, LOOK, and C-LOOK

•Example: range of request: 0-199; Head pointer curr ently at 53�The request sequence: 98, 183, 37, 122, 14, 124, 65, 67


Disk Scheduling: FCFS


Disk Scheduling: SSTF

•Potential problem with SSTF?

2


Disk Scheduling: SCAN (Elevator Algorithm)


Disk Scheduling: C-SCAN


Disk Scheduling: LOOK


Disk Scheduling: C-LOOK


Selecting a Disk-Scheduling Algorithm

• SSTF is common and has a natural appeal• SCAN and C-SCAN

� perform better for systems that place a heavy load on the disk.

• Performance depends on the number and types of requests.

• Requests for disk service can be influenced by the file-allocation method.

• Either SSTF or LOOK is a reasonable choice for the default algorithm.


DEMO

•http://www.sysinternals.com/utilities/diskmon.html

3


RAID


RAID (cont.)

•Level 0 (“striping”)�Good: large requests�Bad: small requests; MTTF (no redundancy)

•Level 1 (“mirroring”)�Good: fault tolerance�Bad: Double the cost of storage

•Level 2�Bad: must be rotationally synchronized; hamming code

•Level 3�Good: fastest for large-file transfer �Bad: what happens if parity disk fails?

•Level 4•Level 5

�Good: Higher I/O rate for writing data; no dedicated parity disk�Bad: Not as fast as RAID 3

•Most popular are 0,1,3,5


Disk Management

• Low-level formatting, or physical formatting — Dividing a disk into sectors that the disk controller can read and write.

• To use a disk to hold files, the operating system s till needs to record its own data structures on the disk .

� Partition the disk into one or more groups of cylinders.� Logical formatting or “making a file system”.

• Boot block initializes system.� The bootstrap is stored in ROM.� Bootstrap loader program.

• Methods such as sector sparing used to handle bad blocks.


Swap-Space Management

• Swap-space — Virtual memory uses disk space as an extension of main memory.

• Swap-space can be carved out of the normal file system,or, more commonly, it can be in a separate d isk partition.

• Swap-space management� 4.3BSD allocates swap space when process starts; holds text

segment (the program) and data segment.� Kernel uses swap maps to track swap-space use.� Solaris 2 allocates swap space only when a page is forced out of

physical memory, not when the virtual memory page is first created.


Power Management

•ENIAC�18000 vacuum tubes; 140,000 watts

•Typical desktop PC�200 watt power supply (loses 15% to heat)�100 million PCs � 20,000 Megawatts (= 20 nuclear power plants)

•Two general approaches for reducing power consumpti on�OS turns off components not in use�App program uses less energy (at degraded user experience)M


Power Consumption in Laptops

4


Reducing Power Consumption

•Display�Dimming

•Hard Disk�Spinning down/up�Increase disk cache in RAM?�Have OS inform app (e.g., auto-save in emacs?)

•CPU�Voltage/frequency scaling �Intel “SpeedStep”: 750MHz � 600MHz; 1.6V � 1.35V (related: “Quickstart”)

•Memory�Turn off cache (lines)?; writing main memory to disk?

•Wireless


Crypto 101: Basic terminology

• Authentication : Assurance of identity of person or originator of data� Password, biometrics, challenge/response (e.g., certificates)

• Authorization : Rights to perform some action• Integrity : Maintaining data consistency

� Checksums, MD5 (hash)

• Confidentiality : Protection from disclosure to unauthorized persons� Encryption: cleartext � ciphertext, ciphertext � cleartext

• Non-repudiation : Originator of communications can't deny it later

• Availability : Resources available for authorized parties


More definitions

• Threat : external to entity under consideration/protection

• Vulnerability : internal to entity under consideration/protection

• Example: sender S wants to send a message to receiver R; what can an “man-in-the-middle” do?

1. Capture message and keep it2. Capture message, read it, and relay it on3. Capture message, modify it and send it on4. Capture message, keep it, send it on, and then send it on again

(replay) – e.g., financial transactionNote: These are threats – whether or not the original message is

vulnerable to these threats is orthogonal!

Documents

Disc Scheduling