Bài giảng Hệ điều hành nâng cao - Chapter 12: Mass - Storage Systems trình bày tổng quan về cấu trúc Mass Storage, cấu trúc đĩa, quản lý lưu trữ, trao đổi và thiết bị lưu trữ,...Đây là tài liệu tham khảo thuộc chuyên ngành Công nghệ thông tin.
AMBIENT/
Chủ đề:
Nội dung Text: Bài giảng Hệ điều hành nâng cao - Chapter 12: Mass - Storage Systems
- Chapter 12: Mass-Storage Systems
Operating System Concepts – 8th 8th Edition
Operating System Concepts – Edition 12.1 Silberschatz, Galvin and Gagne ©2009
- Chapter 12: Mass-Storage Systems
s Overview of Mass Storage Structure
s Disk Structure
s Disk Attachment
s Disk Scheduling
s Disk Management
s Swap-Space Management
s RAID Structure
s Stable-Storage Implementation
s Tertiary Storage Devices
Operating System Concepts – 8th Edition 12.2 Silberschatz, Galvin and Gagne ©2009
- Objectives
s Describe the physical structure of secondary and tertiary storage devices and the resulting effects on the
uses of the devices
s Explain the performance characteristics of mass-storage devices
s Discuss operating-system services provided for mass storage, including RAID and HSM
Operating System Concepts – 8th Edition 12.3 Silberschatz, Galvin and Gagne ©2009
- Overview of Mass Storage Structure
s Magnetic disks provide bulk of secondary storage of modern computers
q Drives rotate at 60 to 250 times per second
q Transfer rate is rate at which data flow between drive and computer
q Positioning time (random-access time) is time to move disk arm to desired cylinder (seek time) and
time for desired sector to rotate under the disk head (rotational latency)
q Head crash results from disk head making contact with the disk surface
4 That’s bad
s Disks can be removable
s Drive attached to computer via I/O bus
q Busses vary, including EIDE, ATA, SATA, USB, Fibre Channel, SCSI, SAS, Firewire
q Host controller in computer uses bus to talk to disk controller built into drive or storage array
Operating System Concepts – 8th Edition 12.4 Silberschatz, Galvin and Gagne ©2009
- Magnetic Disks
s Platters range from .85” to 14” (historically)
q Commonly 3.5”, 2.5”, and 1.8”
s Range from 30GB to 3TB per drive
s Performance
q Transfer Rate – theoretical – 6 Gb/sec
q Effective Transfer Rate – real – 1Gb/sec
q Seek time from 3ms to 12ms – 9ms common for desktop
drives
q Average seek time measured or calculated based on 1/3 of
tracks
q Latency based on spindle speed
4 1/(RPM * 60)
q Average latency = ½ latency
(From Wikipedia)
Operating System Concepts – 8th Edition 12.5 Silberschatz, Galvin and Gagne ©2009
- Magnetic Disk Performance
s Access Latency = Average access time = average seek time + average latency
q For fastest disk 3ms + 2ms = 5ms
q For slow disk 9ms + 5.56ms = 14.56ms
s Average I/O time = average access time + (amount to transfer / transfer rate) + controller overhead
s For example to transfer a 4KB block on a 7200 RPM disk with a 5ms average seek time, 1Gb/sec transfer rate with
a .1ms controller overhead =
q 5ms + 4.17ms + 4KB / 1Gb/sec + 0.1ms =
q 9.27ms + 4 / 131072 sec =
q 9.27ms + .12ms = 9.39ms
Operating System Concepts – 8th Edition 12.6 Silberschatz, Galvin and Gagne ©2009
- Moving-head Disk Mechanism
Operating System Concepts – 8th Edition 12.7 Silberschatz, Galvin and Gagne ©2009
- The First Commercial Disk Drive
1956
IBM RAMDAC computer
included the IBM Model 350
disk storage system
5M (7 bit) characters
50 x 24” platters
Access time = < 1 second
Operating System Concepts – 8th Edition 12.8 Silberschatz, Galvin and Gagne ©2009
- Magnetic Tape
s Was early secondary-storage medium
q Evolved from open spools to cartridges
s Relatively permanent and holds large quantities of data
s Access time slow
s Random access ~1000 times slower than disk
s Mainly used for backup, storage of infrequently-used data, transfer medium between systems
s Kept in spool and wound or rewound past read-write head
s Once data under head, transfer rates comparable to disk
q 140MB/sec and greater
s 200GB to 1.5TB typical storage
s Common technologies are LTO-{3,4,5} and T10000
Operating System Concepts – 8th Edition 12.9 Silberschatz, Galvin and Gagne ©2009
- Disk Structure
s Disk drives are addressed as large 1-dimensional arrays of logical blocks, where the logical block is the
smallest unit of transfer
s The 1-dimensional array of logical blocks is mapped into the sectors of the disk sequentially
q Sector 0 is the first sector of the first track on the outermost cylinder
q Mapping proceeds in order through that track, then the rest of the tracks in that cylinder, and then
through the rest of the cylinders from outermost to innermost
q Logical to physical address should be easy
4 Except for bad sectors
4 Non-constant # of sectors per track via constant angular velocity
Operating System Concepts – 8th Edition 12.10 Silberschatz, Galvin and Gagne ©2009
- Disk Attachment
s Host-attached storage accessed through I/O ports talking to I/O busses
s SCSI itself is a bus, up to 16 devices on one cable, SCSI initiator requests operation and SCSI targets perform
tasks
q Each target can have up to 8 logical units (disks attached to device controller)
s FC is high-speed serial architecture
q Can be switched fabric with 24-bit address space – the basis of storage area networks (SANs) in which
many hosts attach to many storage units
s I/O directed to bus ID, device ID, logical unit (LUN)
Operating System Concepts – 8th Edition 12.11 Silberschatz, Galvin and Gagne ©2009
- Storage Array
s Can just attach disks, or arrays of disks
s Storage Array has controller(s), provides features to attached host(s)
q Ports to connect hosts to array
q Memory, controlling software (sometimes NVRAM, etc)
q A few to thousands of disks
q RAID, hot spares, hot swap (discussed later)
q Shared storage -> more efficiency
q Features found in some file systems
4 Snaphots, clones, thin provisioning, replication, deduplication, etc
Operating System Concepts – 8th Edition 12.12 Silberschatz, Galvin and Gagne ©2009
- Storage Area Network
s Common in large storage environments
s Multiple hosts attached to multiple storage arrays - flexible
Operating System Concepts – 8th Edition 12.13 Silberschatz, Galvin and Gagne ©2009
- Storage Area Network (Cont.)
s SAN is one or more storage arrays
q Connected to one or more Fibre Channel switches
s Hosts also attach to the switches
s Storage made available via LUN Masking from specific arrays to specific servers
s Easy to add or remove storage, add new host and allocate it storage
q Over low-latency Fibre Channel fabric
s Why have separate storage networks and communications networks?
q Consider iSCSI, FCOE
Operating System Concepts – 8th Edition 12.14 Silberschatz, Galvin and Gagne ©2009
- Network-Attached Storage
s Network-attached storage (NAS) is storage made available over a network rather than over a local
connection (such as a bus)
q Remotely attaching to file systems
s NFS and CIFS are common protocols
s Implemented via remote procedure calls (RPCs) between host and storage over typically TCP or UDP on
IP network
s iSCSI protocol uses IP network to carry the SCSI protocol
q Remotely attaching to devices (blocks)
Operating System Concepts – 8th Edition 12.15 Silberschatz, Galvin and Gagne ©2009
- Disk Scheduling
s The operating system is responsible for using hardware efficiently — for the disk drives, this means having
a fast access time and disk bandwidth
s Minimize seek time
s Seek time seek distance
s Disk bandwidth is the total number of bytes transferred, divided by the total time between the first request
for service and the completion of the last transfer
Operating System Concepts – 8th Edition 12.16 Silberschatz, Galvin and Gagne ©2009
- Disk Scheduling (Cont.)
s There are many sources of disk I/O request
q OS
q System processes
q Users processes
s I/O request includes input or output mode, disk address, memory address, number of sectors to transfer
s OS maintains queue of requests, per disk or device
s Idle disk can immediately work on I/O request, busy disk means work must queue
q Optimization algorithms only make sense when a queue exists
s Note that drive controllers have small buffers and can manage a queue of I/O requests (of varying “depth”)
s Several algorithms exist to schedule the servicing of disk I/O requests
s The analysis is true for one or many platters
s We illustrate scheduling algorithms with a request queue (0-199)
98, 183, 37, 122, 14, 124, 65, 67
Head pointer 53
Operating System Concepts – 8th Edition 12.17 Silberschatz, Galvin and Gagne ©2009
- FCFS
Illustration shows total head movement of 640 cylinders
Operating System Concepts – 8th Edition 12.18 Silberschatz, Galvin and Gagne ©2009
- SSTF
s Shortest Seek Time First selects the request with the minimum seek time from the current head position
s SSTF scheduling is a form of SJF scheduling; may cause starvation of some requests
s Illustration shows total head movement of 236 cylinders
Operating System Concepts – 8th Edition 12.19 Silberschatz, Galvin and Gagne ©2009
- SSTF (Cont.)
Operating System Concepts – 8th Edition 12.20 Silberschatz, Galvin and Gagne ©2009
Thêm vào BST
Báo xấu
Download
Vui lòng tải xuống để xem tài liệu đầy đủ
