Chapter 14: Mass-Storage Systems

1
Chapter 14 Mass-Storage Systems
2
Chapter 14 Mass-Storage Systems

• Overview of Mass Storage Structure
• Disk Structure
• Disk Scheduling
• Disk Management
• Swap-Space Management
• RAID Structure
• Tertiary Storage Devices
• Operating System Issues
• Performance Issues

3
Objectives

• Describe the physical structure of secondary and
  tertiary storage devices and the resulting
  effects on the uses of the devices
• Explain the performance characteristics of
  mass-storage devices
• Discuss operating-system services provided for
  mass storage, including RAID and HSM

4
Overview of Mass Storage Structure

• Magnetic tape
• early secondary-storage medium
• Relatively permanent and holds large quantities
  of data
• Access time slow
• Random access 1000 times slower than disk
• Mainly used for backup, storage of
  infrequently-used data, transfer medium between
  systems
• Kept in spool and wound or rewound past
  read-write head
• Once data under head, transfer rates comparable
  to disk
• 20-200GB typical storage
• Common technologies are 4mm, 8mm, 19mm, LTO-2 and
  SDLT

5
Overview of Mass Storage Structure

• Magnetic disks provide bulk of secondary storage
  of modern computers
• Drives rotate up to 7200 times per minute
• Transfer rate is rate at which data flow between
  drive and computer
• 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)
• Head crash results from disk head making contact
  with the disk surface
• Disks can be removable
• Drive attached to computer via I/O bus
• Busses vary, including EIDE, ATA, SATA, USB,
  Fibre Channel, SCSI
• Host controller in computer uses bus to talk to
  disk controller built into drive or storage array

6
Moving-head Disk Machanism
7
Physical Disk Structure

• Disk are made of thin metallic platters with a
  read/write head flying over it.
• To read from disk, we must specify
• cylinder
• surface
• sector
• transfer size
• memory address
• Transfer time includes seek, latency, and
  transfer time

ReadWrite heads
Platters
Spindle
Track
Sector
Rot Delay
Seek Time
8
Disk Structure

• Disk drives are addressed as large 1-dimensional
  arrays of logical blocks, where the logical block
  is the smallest unit of transfer.
• The 1-dimensional array of logical blocks is
  mapped into the sectors of the disk sequentially.
• Sector 0 is the first sector of the first track
  on the outermost cylinder.
• 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.

9
Some Typical Numbers

• Sector size 512 bytes
• Sectors per track 32
• Cylinders per disk (tracks per platter) 1000
• Platters 4-20
• Rotational speed 7200 RPM
• Storage size 100 750 GB
• Seek time 5-100ms
• Latency 010ms(average 4ms)
• Transfer rate 100/300 Mbytes/sec

10
Disk Attachment

• Host-attached storage accessed through I/O ports
  talking to I/O busses
• SCSI itself is a bus, up to 16 devices on one
  cable, SCSI initiator requests operation and SCSI
  targets perform tasks
• Each target can have up to 8 logical units (disks
  attached to device controller
• FC is high-speed serial architecture
• 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
• Can be arbitrated loop (FC-AL) of 126 devices

11
Network-Attached Storage

• Network-attached storage (NAS) is storage made
  available over a network rather than over a local
  connection (such as a bus)
• NFS and CIFS are common protocols
• Implemented via remote procedure calls (RPCs)
  between host and storage
• New iSCSI protocol uses IP network to carry the
  SCSI protocol

12
Storage Area Network

• Common in large storage environments (and
  becoming more common)
• Multiple hosts attached to multiple storage
  arrays - flexible

13
NAS vs SAN

• A SAN is a dedicated storage area network that is
  interconnected through a Fiber Channel protocol
• SAN uses either 1-gigabit or 2-gigabit Fiber
  Channel switches and Fiber Channel host bus
  adaptors.
• Devices such as file servers connect directly to
  the SAN through the Fiber Channel protocol.
• NAS connects directly to the network using TCP/IP
  over Ethernet.
• In most cases, no changes to the existing network
  infrastructure need to be made in order to
  install a NAS solution.
• The network-attached storage device is attached
  to the local area network (typically, an
  Ethernet) and assigned an IP address just like
  any other network device.

14
NAS vs SAN

• Storage solutions vendors try to be all things
  to all people,
• SAN and NAS technologies are beginning to
  converge and blur the line even more.
• CROSSOVER FEATURESFeatures that were once only
  available in SAN solutions are now starting to
  become available in NAS products,
• Ex. iSCSI allows parts of SAN infrastructures to
  act as NAS filers.
• iSCSI (Internet small computer system interface)
  is an Internet protocol-based storage networking
  standard for linking data storage facilities.
• By carrying SCSI commands over IP networks,
  iSCSI is used to facilitate data transfers over
  intranets and to manage storage over long
  distances, loaning SAN a few of the NAS
  capabilities.

15
NAS vs SAN

• SANs are highly redundant through the
  implementation of multipathing and the ability to
  create fully redundant fiber meshes so there is
  no single point of failure.
• SANs feature block-level transfers instead of NAS
  file-level transfers. This is critical if you
  have database applications that read and write
  data at the block level.
• SAN products run on Fiber Channel protocol and
  are entirely isolated from the IP network, so
  there is no contention over the TCP/IP offload
  engine (which optimizes throughput).
• NAS products run over your existing TCP/IP
  network, and, as such, are prone to latency and
  broadcast storms, and compete for bandwidth with
  users and other network devices.

16

NAS vs SAN

• You can leverage an existing SAN to act like a
  NAS with an iSCSI switch, which saves money.
• With a SAN, there is higher security because SAN
  uses zoning and logical unit number (LUN)
  security. NAS security is typically implemented
  at the file-system level through traditional
  operating system access-control lists.
• There is more flexibility for redundant array of
  independent disks (RAID) levels. While NAS
  products do support standard RAID levels, such as
  0,1,5, you typically do not get the flexibility
  to mix RAID levels within the same device.

17
NAS vs SAN

• LOWER COSTS WITH NASFile servers see SAN
  attached volumes as locally attached disks,
  whereas a NAS presents them as remote network
  file system (NFS) or new technology file system
  (NTFS) file shares.
• NTFS is the file system that the operating system
  uses for storing and retrieving files on a hard
  disk.
• NFS is a client/server application that lets a
  computer user view and optionally store and
  update files on a remote computer as though they
  were on the users own computer.
• Some applications do not support remote drives
  and can only use a volume that is local to the
  OS.
• NAS is cheaper than SAN. The initial investment
  in a SAN is expensive due to the high cost of
  Fiber Channel switches and host bus adaptors.

18
NAS vs SAN

• If you need just simple file storage, NAS is the
  way to go.
• A NAS product simply plugs into your existing IP
  network as any other device and looks like a
  normal file share on the network.
• So, a NAS can be dropped right into your
  existing IP network without any additional costs
  or infrastructure changes.
• Ethernet is a stable and mature protocol and any
  IT administrator already knows Ethernet and
  TCP/IP,
• there are no steep learning curves compared with
  learning and understanding the Fiber Channel
  protocol.
• After carefully weighing all the benefits,
  ProfitLine chose to go with a Hitachi 9200 SAN
  infrastructure with Compaq/HP Proliant Servers
  because of its better support for databases and
  higher security.
• Since ProfitLine is processing tens of thousands
  of invoices monthly on behalf of its clients, and
  the number is only going up with the addition of
  new clients, we needed the more powerful and
  scalable SAN technology.

19
Disk Scheduling

• The operating system is responsible for using
  hardware efficiently
• For the disk drives, this means having a fast
  access time and disk bandwidth.
• Access time has two major components
• Seek time is the time for the disk are to move
  the heads to the cylinder containing the desired
  sector.
• Rotational latency is the additional time waiting
  for the disk to rotate the desired sector to the
  disk head.
• Minimize seek time
• Seek time ? seek distance
• 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.

20
Disk Structure

• There is no structure to a disk except cylinders
  and sectors, anything else is up to the OS.
• The OS imposes some structure on disks.
• Each disk contains
• 1. data e.g., user files
• 2. meta-data OS info describing the disk
  structure
• For example, the free list is a data structure
  indicating which disk blocks are free. It is
  stored on disk (usually) as a bit map each bit
  corresponds to one disk block.
• The OS may keep the free list bit map in memory
  and write it back to disk from time to time.

21
??? ????

• ??? ????? ???
• ?? ???? FCFS ??
• Seek ???? ? ??
• latency ???? ???? ??
• PC? ?? n-way disk interleaving ??
• Disk ???? ??? ???? ??
• ???
• ?? ?? ??
• ??? ????? ??

22
Disk Scheduling

• Because disks are slow and seeks are long and
  depend on distance, we can schedule disk
  accesses, e.g.
• FCFS (do nothing)
• ok when load is low
• long waiting times for long request queue
• SSTF (shortest seek time first)
• always minimize arm movement. maximize
  throughput.
• favors middle blocks
• SCAN (elevator) -- continue in same direction
  until done, then reverse direction and service
  in that order
• C-SCAN -- like scan, but go back to 0 at end
• In general, unless there are request queues, it
  doesnt matter
• explains why some single user systems do badly
  under heavy load.
• The OS (or database system) may locate files
  strategically for performance reasons.

23
Disk Scheduling (Cont.)

• Several algorithms exist to schedule the
  servicing of disk I/O requests.
• We illustrate them with a request queue (0-199).
• 98, 183, 37, 122, 14, 124, 65, 67
• Head pointer 53

24
??? ???? - Seek ???

• FCFS (First-Come-First-Served) ????
• ??? ???? ???? ??
• Disk ??(load)? ?? ? ??

25
FCFS
Illustration shows total head movement of 640
cylinders.
26
??? ???? - Seek ???

• SSTF(Shortest Seek Time First) ????
• Seek Distance? ?? ??? ?? ??
• ??? track? ?? ??? ?? Starvation??
• ?? ??? ??? ?

27
SSTF

• Selects the request with the minimum seek time
  from the current head position.
• SSTF scheduling is a form of SJF scheduling may
  cause starvation of some requests.
• Illustration shows total head movement of 236
  cylinders.

28
SSTF (Cont.)
29
??? ???? - Seek ???

• SCAN???? (Elevator Algorithm)
• Head? ?? ???? ?? ?? ??? ??? ??
• ???? ?? ???? ????

30
SCAN

• The disk arm starts at one end of the disk, and
  moves toward the other end, servicing requests
  until it gets to the other end of the disk, where
  the head movement is reversed and servicing
  continues.
• Sometimes called the elevator algorithm.
• Illustration shows total head movement of 208
  cylinders.

31
SCAN (Cont.)
32
??? ???? - Seek ???

• C-SCAN
• SCAN? ?? -gt Head? ?? ????? ??

33
C-SCAN

• Provides a more uniform wait time than SCAN.
• The head moves from one end of the disk to the
  other. servicing requests as it goes. When it
  reaches the other end, however, it immediately
  returns to the beginning of the disk, without
  servicing any requests on the return trip.
• Treats the cylinders as a circular list that
  wraps around from the last cylinder to the first
  one.

34
C-SCAN (Cont.)
35
C-LOOK

• Version of C-SCAN
• Arm only goes as far as the last request in each
  direction, then reverses direction immediately,
  without first going all the way to the end of the
  disk.

36
C-LOOK (Cont.)
37
??? ???? - Seek ???

• N?? SCAN????
• ?? ???? ??? ??? ? ???? ??? ??
• ???? ??? ??? ?????? ??? ??

38
??? ???? - Seek ???

• ???? ??
• ?? ?? ???? ??
• seek latency ???
• C-scan with rotational optimization? ?? ??

39
Basic disk scheduling policies
40
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.
• The disk-scheduling algorithm should be written
  as a separate module of the operating system,
  allowing it to be replaced with a different
  algorithm if necessary.
• Either SSTF or LOOK is a reasonable choice for
  the default algorithm.

41
??? ???? - ?? ???

• ???? ?? ???? ?? ???? ?? ??
• SLTF(Shortest-latency-time-first)
• ? ??? ?? SLTF? ???
• Sector queueing???? ??

42
??? ????- ??? ????

• ?? ??? ????? ????
• ???? ????? ??? ??
• ????? ?? ?? ???? ?? ??? ??? ?? ??? ????? ???
  ????? ????
• ?? ?????? ??
• ?? ????? ??? ???? ??? ??? ??? ????? ??
• ??? ????? ??? ??? ?? ????? ??? ???? ???
• ?????? ?? ??? ???? ?????? ??? ???, ??? ???? ???
  ??? ????? ????

43
??? ????- ??? ????

• ?? ??? ?? ???
• ??? ??? ?? ??? ? ???? ?
• ?) ????? ???? ?? ??? ?? ??
• ??? ??? ?? ?? RPS(Rotational positional
  sensing)??
• RPS rotational delay?? ??? free?
• ???(nonuniform) ?? ??
• ?? ?? ??? ??? ??? ???? ?? ??? ??
• ???? ??? ??? ??
• ??? ????? ? ?? ??

44
Disk Caching

• The idea is that data you (or someone) accessed
  recently is likely to be data that you (or
  someone) needs again.

Excel
PPT
The FileSystem
The Buffer Cache
The Disk Driver
45
Disk Locality Works Quite Well

• Most studies have shown that a disk cache on
  the order of 2-4 megabytes captures 90 of the
  read traffic for many kinds of applications
• your mileage will definitely vary
• but the impact can be enormous
• 90 hit rate
• hit time of .1 ms
• miss time of 10ms
• average access time
• (.9 .1) (.1 10) 1.09 ms
• No cache increases disk access time by almost
  1000.

90
H I T R a t e
a reasonable operating range.
50
1MB 2MB 4MB 8MB 16MB
46
The Buffer Cache

• The buffer cache has only a finite number of
  blocks in it.
• On a cache miss, one block must be discarded in
  favor of the fresh block that is being brought
  in.
• Typical systems operate in an LRU order
• the least recently used block is the one that
  should be discarded next.
• favors busier portions of the filesystem
• can hurt the quiet client.
• Some workloads LRU is really bad for
• sequential scans
• video, audio
• random access
• large database

47
Read Ahead

• Many file systems will implement read ahead
  whereby the file system will try to predict what
  the application is going to need next, go to
  disk, and read it in.
• The goal is to beat the application to the punch.
• For sequentially accessed files, this can be a
  big win.
• as long as reads are spaced in time, the OS can
  schedule the next IO during the current gap.
• enabling read ahead can give applications a 100
  speedup.
• suppose that it takes 10ms to read a block.
• the application requests the next block in
  sequence every 10 ms.
• with read ahead, wait time is 0 ms per block and
  execution time is
• 10 number of blocks read.
• without read ahead, wait time is 20 ms per block
  and execution time is
• 20 number of blocks read.

48
Caching works for reads, what about writes?

• On write, it is necessary to ensure that the data
  makes it through the buffer cache and onto the
  disk.
• Consequently, writes, even with caching, can be
  slow.
• Systems do several things to compensate for this
• write-behind
• maintain a queue of uncommitted blocks
• periodically flush the queue to disk
• unreliable
• battery backed up RAM
• as with write-behind, but maintain the queue in
  battery backed up memory
• expensive
• log structured filed system
• always write the next block on disk the one past
  where the last block was written
• complicated

49
? ???

• RAM?? ???? ?????? ??
• ??? ??? ??? ?? ??? ?? ??? ???? ????? ??, ?? ?? ??
  ?? ??? ??? ????
• ??? ??? ??
• ??
• ??
• ???? volatile

50
RAM Disk

• Dedicate a large chunk of memory to act as a file
  system.
• Note that most disks now have a small amount of
  built in RAM already on them.
• Extremely simple and at times effective.
• Not so good for reliability
• used to maintain temporary, or expendable
  files.
• Considering hit rates of caches, not clear that
  this is a great win.
• FLASH RAM is one kind of RAM disk that does not
  have a problem with power outage.
• faster than disk, slower than memory
• 10K writes, and you throw it away.

51
? ???

• ??? ?? ??
• ?? ??? ??? ?? ?? ?? ??? ??(WORM)
• ??? ?? ??? ??? ???? ??
• ??? ???? ??
• ?? ???? ?? ? ???? 1021 ?? ??? ??

52
CD-ROMS, WORMS ? ??-??? ???

• ??? ?(600MB)
• ??? ?? ??? ??? ??
• ??? ???? ??
• CD-ROM ? ??? ?? ??? ??
• ??? ??? ??? ???? ?? ??
• WORM? ??
• ?? ?? ? ?? ????? ?? ?
• local seek? ?? ????? ???? ?? ???? ????? ?? ??.

53
?? ?? ??

• ISAM ? ????? ?? ??? ?? ?? ??
• ??? ??? ???? ???? ?? ?? ? ??(seek) ??
• master index
• cylinder index
• ?? ???

54
??? ??

• Delayed write write? ??? ??(??)? ??
• sync system call ?? ??? ???? ???? ??

55
?? ?? ?? ??

• ???? ???(fragmentation)?
• ?? OS? ??? ??? ???? ??
• ?? ??? ????? ????? ???
• ?? ????? ???? ?? ????? ???? ?? ??
• ?? ??? ?? ???? ???? ?? ??? ??(read-only? ??)
• ?? ??? ????? ?? ???? ???? ?? ?? ????? ???? ??? ??

56
?? ?? ?? ??

• ??? ??? ?? ? ???? ?? ???? ???? ???? ???? ?? ??
• ???? ????? ???? ??
• ?? ??? ?? ?? ??? ?? ??? ??
• ??? ?? ?? ??
• ????, latency ??, ???? ??
• ??? ?? ??

57
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
  still 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.

58
Booting from a Disk in Windows 2000
59
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
  disk 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.

60
Data Structures for Swapping on Linux Systems
61
RAID

• Caching, RAM disks deal with the latency issue.
• DISKS can also be used in PARALLEL
• This is the idea behind RAIDs
• Redundant Array of Inexpensive Disks
• Many RAID levels (Raid0-5)

Data bits
0
But we have an increased reliability problem. If
any one disk fails, all 8 are effectively useless.
1
0
1
1
1
0
1
Parity bits
A redundant array of inexpensive disks.
An array of inexpensive disks (Can read 8 tracks
at once)
62
RAID (cont)

• Several improvements in disk-use techniques
  involve the use of multiple disks working
  cooperatively.
• Disk striping uses a group of disks as one
  storage unit.
• RAID schemes improve performance and improve the
  reliability of the storage system by storing
  redundant data.
• Mirroring or shadowing keeps duplicate of each
  disk.
• Block interleaved parity uses much less
  redundancy.

63
Different RAID levels

• Raid 0 Striping
• reduce disk queue length, good for random access.
  no reliability
• Raid 1 Mirroring
• write data to both disks, expensive
• Raid 2 ECC
• stripe at bit level, multiple parity bits to
  determine failed bit. expensive (eg, 10 data
  disks require 4 ECC disks), read/write 1 bit,
  means read/write all in a parity stripe!
• Raid 3 PARITY only.
• disks say they break. Only need to reconstruct.
  Byte stripe. Single parity drive. All disks
  involved in a single read. Bottleneck writes.
• Raid 4 PARITY w/larger stripe size
• can do multiple reads in parallel. Single parity
  drive. Bottleneck writes.
• Raid 5 Floating parity
• assign parity bits for different stripes to
  different disks. Still have write bottleneck, but
  distributed. One write requires a read/write.

64
RAID Levels
65
Data mapping for RAID Level 0 Array
Physical Disk 0
Physical Disk 1
Physical Disk 2
Physical Disk 3
strip 0
strip 1
strip 2
strip 3
strip 4
strip 5
strip 6
strip 7
strip 8
strip 9
strip 10
strip11
strip 12
strip 13
strip 14
strip 15
66
RAID Level 0 Array

• ??? ??? ?? ?? ??? ?? ??? ???? ?? ????? ???? ???
• ?? ???? ?? ?? ?? ??? ??? ?? ???? I/O? ??? ? ??

67
RAID 1 (mirrored)
68
RAID 1 (mirrored)

• ?? ???? ?? ???
• ? ?? ??? ?? ?? ??? ??
• ??? ???? ??
• ?? ??? ??? ??? ?? ??
• ?? ? ???? ?? ???? ??? ? ?? ??? ??
• OLTP(On Line Transaction System)? ??

69
RAID 2 (redundancy through Hamming code)
f2(b)
f1(b)
f0(b)
b2
b1
b0
b2
(7, 4) system 7 number of total data
bits(includes parity bits) 4 number of actual
data bits(excludes parity bits) (c, d), p c-d ?
? d p 1 lt 2p ???? ?
70
RAID 2 (redundancy through Hamming code)

• ???? ?? ??? ???? ? ? Hamming code ??
• ? ?? ??? ?? ??
• ?? ??? ?? ?? ??
• ?? ?? ??? ?? ?? ??? ??

71
RAID 3 (byte-interleaved parity)
P(b)
b2
b1
b0
b2
72
RAID 3 (byte-interleaved parity)

• ??? ??? ???? 1?? ?? parity disk
• ??? ???? ?? ??? ?? ??? ?? ??
• ??/??? ??? ? ????? ?? ??? ??? ??
• ?? ??? ??? ? ????? ??? ??

73
RAID 4 (block-level parity)
block 0
block 1
block 2
block 3
P(0-3)
block 4
block 5
block 6
block 7
P(4-7)
block 9
block 10
block 11
block 8
P(8-11)
block 13
block 15
block 12
block 14
P(12-15)
74
RAID 4 (block-level parity)

• RAID 3? ?? ??? ???? ?? ????? ??? ???
• ?? ??? ???? 1?? ?? parity disk
• ?? ? ? ?? ??? ??? ??

75
RAID 5 (block-level distributed parity)
block 0
block 1
block 2
block 3
P(0-3)
block 5
block 4
block 6
P(4-7)
block 7
block 9
block 10
block 11
P(8-11)
block 8
block 12
P(12-15)
block 13
block 14
block 15
P(16-19)
block 16
block 17
block 18
block 19
76
RAID 5 (block-level distributed parity)

• RAID 4?? ??? ???? ??? ?
• ???? ? ???? ??

77

• RAID 6
• ??? ???? ??? ?? ???? ???? ???? ???? RAID ??
• ???? ???? ??? ?? 2??? ???? ??
• ? ?? ???? ?? Orthogonal ?? ??? ? ?? ???? ?? ????
  ??? ?? ??

78
RAID (0 1) and (1 0)
79
RAID (0 1) and (1 0)
80
Why is RAID 10 better than RAID 01?

• RAID 01 configuration where multiple disks are
  striped together into sets (sets A B in the
  diagram, each set being as large as the resulting
  final volume), and then two or more sets are
  mirrored together.
• RAID 10 configuration where two or more drives
  are mirrored together (mirrors 1-4 in the
  diagram), and then the mirrors (as many as are
  needed to result in the desired amount of space)
  are striped together.
• In either case (01 or 10), the loss of a single
  drive does not result in failure of the RAID
  system.
• The difference comes in the chance that the loss
  of a second drive from the system will result in
  the failure of the whole system.
• In RAID 01, you have to lose one drive from each
  disk set (one drive from set A and one drive from
  set B) to result in the failure of the whole
  system..
• In RAID 10, you have to lose all drives in a
  mirror. This would be both drives in any numbered
  pair in the diagram.

81
Why is RAID 10 better than RAID 01?

• Mathematically, the difference is that the chance
  of system failure with two drive failures in a
  RAID 01 system with two sets of drives is (n/2 -
  1)/(n - 1) where n is the total number of drives
  in the system.
• The chance of system failure in a RAID 10 system
  with two drives per mirror is 1/(n - 1).
• So, using the 8 drive systems shown in the
  diagrams, the chance that losing a second drive
  would bring down the RAID system is 3/7 with a
  RAID 01 system and 1/7 with a RAID 10 system.
• The math gets more complicated when you have more
  than two elements to a mirror.

82
Why is RAID 10 better than RAID 01?

• Another difference between the two RAID
  configurations is performance when the system is
  in a degraded state, i.e. after it has lost one
  or more drives but has not lost the right
  combination of drives to completely fail.
• In a RAID 01 configuration, the loss of any
  drive in a set causes the failure of that entire
  set and the set is removed from the RAID system.
  Generally (in the two set case) this means you
  are left with a RAID 0 system made up of the
  remaining set of disks. This probably slightly
  improves write performance and slightly degrades
  read performance.
• In a RAID 10 system, you would see the same
  effect on each mirror that loses a drive, but not
  the whole system. In other words, a RAID 10
  configuration will tend to show similar, but less
  dramatic, changes in performance when in a
  degraded mode than RAID 01. However, the changes
  will likely be slight in any case.
• One more difference is the speed at which the
  RAID system recovers once the failed disk is
  replaced.
• RAID 10 only has to re-mirror one drive,
• whereas RAID 01 has to re-mirror the entire
  failed set. So RAID 10 will recover
  significantly faster.

83
Stable-Storage Implementation

• Write-ahead log scheme requires stable storage.
• To implement stable storage
• Replicate information on more than one
  nonvolatile storage media with independent
  failure modes.
• Update information in a controlled manner to
  ensure that we can recover the stable data after
  any failure during data transfer or recovery.

84
Tertiary Storage Devices

• Low cost is the defining characteristic of
  tertiary storage.
• Generally, tertiary storage is built using
  removable media
• Common examples of removable media are floppy
  disks and CD-ROMs other types are available.

85
Removable Disks

• Floppy disk thin flexible disk coated with
  magnetic material, enclosed in a protective
  plastic case.
• Most floppies hold about 1 MB similar technology
  is used for removable disks that hold more than 1
  GB.
• Removable magnetic disks can be nearly as fast as
  hard disks, but they are at a greater risk of
  damage from exposure.

86
Removable Disks (Cont.)

• A magneto-optic disk records data on a rigid
  platter coated with magnetic material.
• Laser heat is used to amplify a large, weak
  magnetic field to record a bit.
• Laser light is also used to read data (Kerr
  effect).
• The magneto-optic head flies much farther from
  the disk surface than a magnetic disk head, and
  the magnetic material is covered with a
  protective layer of plastic or glass resistant
  to head crashes.
• Optical disks do not use magnetism they employ
  special materials that are altered by laser light.

87
WORM Disks

• The data on read-write disks can be modified over
  and over.
• WORM (Write Once, Read Many Times) disks can be
  written only once.
• Thin aluminum film sandwiched between two glass
  or plastic platters.
• To write a bit, the drive uses a laser light to
  burn a small hole through the aluminum
  information can be destroyed by not altered.
• Very durable and reliable.
• Read Only disks, such ad CD-ROM and DVD, com from
  the factory with the data pre-recorded.

88
Tapes

• Compared to a disk, a tape is less expensive and
  holds more data, but random access is much
  slower.
• Tape is an economical medium for purposes that do
  not require fast random access, e.g., backup
  copies of disk data, holding huge volumes of
  data.
• Large tape installations typically use robotic
  tape changers that move tapes between tape drives
  and storage slots in a tape library.
• stacker library that holds a few tapes
• silo library that holds thousands of tapes
• A disk-resident file can be archived to tape for
  low cost storage the computer can stage it back
  into disk storage for active use.

89
Operating System Issues

• Major OS jobs are to manage physical devices and
  to present a virtual machine abstraction to
  applications
• For hard disks, the OS provides two abstraction
• Raw device an array of data blocks.
• File system the OS queues and schedules the
  interleaved requests from several applications.

90
Application Interface

• Most OSs handle removable disks almost exactly
  like fixed disks a new cartridge is formatted
  and an empty file system is generated on the
  disk.
• Tapes are presented as a raw storage medium,
  i.e., and application does not not open a file on
  the tape, it opens the whole tape drive as a raw
  device.
• Usually the tape drive is reserved for the
  exclusive use of that application.
• Since the OS does not provide file system
  services, the application must decide how to use
  the array of blocks.
• Since every application makes up its own rules
  for how to organize a tape, a tape full of data
  can generally only be used by the program that
  created it.

91
Tape Drives

• The basic operations for a tape drive differ from
  those of a disk drive.
• locate positions the tape to a specific logical
  block, not an entire track (corresponds to seek).
• The read position operation returns the logical
  block number where the tape head is.
• The space operation enables relative motion.
• Tape drives are append-only devices updating a
  block in the middle of the tape also effectively
  erases everything beyond that block.
• An EOT mark is placed after a block that is
  written.

92
File Naming

• The issue of naming files on removable media is
  especially difficult when we want to write data
  on a removable cartridge on one computer, and
  then use the cartridge in another computer.
• Contemporary OSs generally leave the name space
  problem unsolved for removable media, and depend
  on applications and users to figure out how to
  access and interpret the data.
• Some kinds of removable media (e.g., CDs) are so
  well standardized that all computers use them the
  same way.

93
Hierarchical Storage Management (HSM)

• A hierarchical storage system extends the storage
  hierarchy beyond primary memory and secondary
  storage to incorporate tertiary storage usually
  implemented as a jukebox of tapes or removable
  disks.
• Usually incorporate tertiary storage by extending
  the file system.
• Small and frequently used files remain on disk.
• Large, old, inactive files are archived to the
  jukebox.
• HSM is usually found in supercomputing centers
  and other large installations that have enormous
  volumes of data.

94
Speed

• Two aspects of speed in tertiary storage are
  bandwidth and latency.
• Bandwidth is measured in bytes per second.
• Sustained bandwidth average data rate during a
  large transfer of bytes/transfer time.Data
  rate when the data stream is actually flowing.
• Effective bandwidth average over the entire I/O
  time, including seek or locate, and cartridge
  switching.Drives overall data rate.

95
Speed (Cont.)

• Access latency amount of time needed to locate
  data.
• Access time for a disk move the arm to the
  selected cylinder and wait for the rotational
  latency lt 35 milliseconds.
• Access on tape requires winding the tape reels
  until the selected block reaches the tape head
  tens or hundreds of seconds.
• Generally say that random access within a tape
  cartridge is about a thousand times slower than
  random access on disk.
• The low cost of tertiary storage is a result of
  having many cheap cartridges share a few
  expensive drives.
• A removable library is best devoted to the
  storage of infrequently used data, because the
  library can only satisfy a relatively small
  number of I/O requests per hour.

96
Reliability

• A fixed disk drive is likely to be more reliable
  than a removable disk or tape drive.
• An optical cartridge is likely to be more
  reliable than a magnetic disk or tape.
• A head crash in a fixed hard disk generally
  destroys the data, whereas the failure of a tape
  drive or optical disk drive often leaves the data
  cartridge unharmed.

97
Cost

• Main memory is much more expensive than disk
  storage
• The cost per megabyte of hard disk storage is
  competitive with magnetic tape if only one tape
  is used per drive.
• The cheapest tape drives and the cheapest disk
  drives have had about the same storage capacity
  over the years.
• Tertiary storage gives a cost savings only when
  the number of cartridges is considerably larger
  than the number of drives.

98
Price per Megabyte of DRAM, From 1981 to 2004
99
Price per Megabyte of Magnetic Hard Disk, From
1981 to 2004
100
Price per Megabyte of a Tape Drive, From 1984-2000
101
End of Chapter 14