Chapter 10: Storage and File Structure

■ Overview of Physical Storage Media

■ Magnetic Disks
■ RAID
■ Tertiary Storage
■ Storage Access

Chapter 10: Storage and File Structure ■ File Organization

■ Organization of Records in Files
■ Data-Dictionary Storage

Database System Concepts, 6th Ed.

©Silberschatz, Korth and Sudarshan
See www.db-book.com for conditions on re-use

Database System Concepts - 6th Edition 10.2 ©Silberschatz, Korth and Sudarshan

Classification of Physical Storage Media Physical Storage Media

■ Speed with which data can be accessed ■ Cache – fastest and most costly form of storage; volatile; managed
■ Cost per unit of data by the computer system hardware.
■ Reliability ■ Main memory:

● data loss on power failure or system crash ● fast access (10s to 100s of nanoseconds; 1 nanosecond = 10–9
seconds)
● physical failure of the storage device
● generally too small (or too expensive) to store the entire
■ Can differentiate storage into:
database
● volatile storage: loses contents when power is switched off
4 capacities of up to a few Gigabytes widely used currently
● non-volatile storage:
4 Capacities have gone up and per-byte costs have
4 Contents persist even when power is switched off. decreased steadily and rapidly (roughly factor of 2 every 2
4 Includes secondary and tertiary storage, as well as batter- to 3 years)
backed up main-memory. ● Volatile — contents of main memory are usually lost if a power
failure or system crash occurs.

Database System Concepts - 6th Edition 10.3 ©Silberschatz, Korth and Sudarshan Database System Concepts - 6th Edition 10.4 ©Silberschatz, Korth and Sudarshan
 Physical Storage Media (Cont.) Physical Storage Media (Cont.)
■ Flash memory
■ Magnetic-disk
● Data survives power failure ● Data is stored on spinning disk, and read/written magnetically
● Data can be written at a location only once, but location can be ● Primary medium for the long-term storage of data; typically stores entire
erased and written to again database.
4 Can support only a limited number (10K – 1M) of write/erase ● Data must be moved from disk to main memory for access, and written
cycles. back for storage

4 Erasing of memory has to be done to an entire bank of 4 Much slower access than main memory (more on this later)
memory ● direct-access – possible to read data on disk in any order, unlike
magnetic tape
● Reads are roughly as fast as main memory
● Capacities range up to roughly 1.5 TB as of 2009
● But writes are slow (few microseconds), erase is slower
4 Much larger capacity and cost/byte than main memory/flash memory
● Widely used in embedded devices such as digital cameras, 4 Growing constantly and rapidly with technology improvements (factor
phones, and USB keys of 2 to 3 every 2 years)
● Survives power failures and system crashes
4 disk failure can destroy data, but is rare

Database System Concepts - 6th Edition 10.5 ©Silberschatz, Korth and Sudarshan Database System Concepts - 6th Edition 10.6 ©Silberschatz, Korth and Sudarshan

Physical Storage Media (Cont.) Physical Storage Media (Cont.)

■ Optical storage ■ Tape storage
● non-volatile, data is read optically from a spinning disk using ● non-volatile, used primarily for backup (to recover from disk
a laser failure), and for archival data
● CD-ROM (640 MB) and DVD (4.7 to 17 GB) most popular ● sequential-access – much slower than disk
forms
● very high capacity (40 to 300 GB tapes available)
● Blu-ray disks: 27 GB to 54 GB
● tape can be removed from drive Þ storage costs much
● Write-one, read-many (WORM) optical disks used for archival cheaper than disk, but drives are expensive
storage (CD-R, DVD-R, DVD+R)
● Tape jukeboxes available for storing massive amounts of
● Multiple write versions also available (CD-RW, DVD-RW, data
DVD+RW, and DVD-RAM)
4 hundreds of terabytes (1 terabyte = 109 bytes) to even
● Reads and writes are slower than with magnetic disk multiple petabytes (1 petabyte = 1012 bytes)
● Juke-box systems, with large numbers of removable disks, a
few drives, and a mechanism for automatic loading/unloading
of disks available for storing large volumes of data

Database System Concepts - 6th Edition 10.7 ©Silberschatz, Korth and Sudarshan Database System Concepts - 6th Edition 10.8 ©Silberschatz, Korth and Sudarshan
 Storage Hierarchy Storage Hierarchy (Cont.)

■ primary storage: Fastest media but volatile (cache, main

memory).
■ secondary storage: next level in hierarchy, non-volatile,
moderately fast access time
● also called on-line storage
● E.g. flash memory, magnetic disks
■ tertiary storage: lowest level in hierarchy, non-volatile, slow
access time
● also called off-line storage
● E.g. magnetic tape, optical storage

Database System Concepts - 6th Edition 10.9 ©Silberschatz, Korth and Sudarshan Database System Concepts - 6th Edition 10.10 ©Silberschatz, Korth and Sudarshan

Magnetic Hard Disk Mechanism Magnetic Disks

■ Read-write head
● Positioned very close to the platter surface (almost touching it)
● Reads or writes magnetically encoded information.
■ Surface of platter divided into circular tracks
● Over 50K-100K tracks per platter on typical hard disks
■ Each track is divided into sectors.
● A sector is the smallest unit of data that can be read or written.
● Sector size typically 512 bytes
Typical sectors per track: 500 to 1000 (on inner tracks) to 1000 to 2000 (on
●
outer tracks)
■ To read/write a sector
disk arm swings to position head on right track
●
platter spins continually; data is read/written as sector passes under head
●
■ Head-disk assemblies
● multiple disk platters on a single spindle (1 to 5 usually)
● one head per platter, mounted on a common arm.

NOTE: Diagram is schematic, and simplifies the structure of actual disk drives ■ Cylinder i consists of ith track of all the platters

Database System Concepts - 6th Edition 10.11 ©Silberschatz, Korth and Sudarshan Database System Concepts - 6th Edition 10.12 ©Silberschatz, Korth and Sudarshan
 Magnetic Disks (Cont.) Disk Subsystem

■ Earlier generation disks were susceptible to head-crashes

Surface of earlier generation disks had metal-oxide coatings which
●
would disintegrate on head crash and damage all data on disk
● Current generation disks are less susceptible to such disastrous
failures, although individual sectors may get corrupted
■ Disk controller – interfaces between the computer system and the disk
drive hardware.
● accepts high-level commands to read or write a sector
● initiates actions such as moving the disk arm to the right track and
actually reading or writing the data
■ Multiple disks connected to a computer system through a controller
● Computes and attaches checksums to each sector to verify that
● Controllers functionality (checksum, bad sector remapping) often
data is read back correctly carried out by individual disks; reduces load on controller
4 If data is corrupted, with very high probability stored checksum ■ Disk interface standards families
won’t match recomputed checksum
● ATA (AT adaptor) range of standards
● Ensures successful writing by reading back sector after writing it
● SATA (Serial ATA)
● Performs remapping of bad sectors
● SCSI (Small Computer System Interconnect) range of standards
● SAS (Serial Attached SCSI)
● Several variants of each standard (different speeds and capabilities)

Database System Concepts - 6th Edition 10.13 ©Silberschatz, Korth and Sudarshan Database System Concepts - 6th Edition 10.14 ©Silberschatz, Korth and Sudarshan

Disk Subsystem Performance Measures of Disks

■ Disks usually connected directly to computer system ■ Access time – the time it takes from when a read or write request is issued to
when data transfer begins. Consists of:
■ In Storage Area Networks (SAN), a large number of disks are ● Seek time – time it takes to reposition the arm over the correct track.
connected by a high-speed network to a number of servers 4 Average seek time is 1/2 the worst case seek time.
■ In Network Attached Storage (NAS) networked storage provides a – Would be 1/3 if all tracks had the same number of sectors, and we
file system interface using networked file system protocol, instead of ignore the time to start and stop arm movement
4 4 to 10 milliseconds on typical disks
providing a disk system interface
● Rotational latency – time it takes for the sector to be accessed to appear
under the head.
4 Average latency is 1/2 of the worst case latency.
4 4 to 11 milliseconds on typical disks (5400 to 15000 r.p.m.)
■ Data-transfer rate – the rate at which data can be retrieved from or stored to
the disk.
● 25 to 100 MB per second max rate, lower for inner tracks
● Multiple disks may share a controller, so rate that controller can handle is
also important
4 E.g. SATA: 150 MB/sec, SATA-II 3Gb (300 MB/sec)
4 Ultra 320 SCSI: 320 MB/s, SAS (3 to 6 Gb/sec)
4 Fiber Channel (FC2Gb or 4Gb): 256 to 512 MB/s

Database System Concepts - 6th Edition 10.15 ©Silberschatz, Korth and Sudarshan Database System Concepts - 6th Edition 10.16 ©Silberschatz, Korth and Sudarshan
 Performance Measures (Cont.) Optimization of Disk-Block Access

■ Block – a contiguous sequence of sectors from a single track

■ Mean time to failure (MTTF) – the average time the disk is
expected to run continuously without any failure. ● data is transferred between disk and main memory in blocks
● Typically 3 to 5 years ● sizes range from 512 bytes to several kilobytes
● Probability of failure of new disks is quite low, corresponding to a 4 Smaller blocks: more transfers from disk
“theoretical MTTF” of 500,000 to 1,200,000 hours for a new disk 4 Larger blocks: more space wasted due to partially filled blocks
4 E.g., an MTTF of 1,200,000 hours for a new disk means that 4 Typical block sizes today range from 4 to 16 kilobytes
given 1000 relatively new disks, on an average one will fail
■ Disk-arm-scheduling algorithms order pending accesses to tracks so
every 1200 hours
that disk arm movement is minimized
● MTTF decreases as disk ages
● elevator algorithm:

R6 R3 R1 R5 R2 R4

Inner track Outer track

Database System Concepts - 6th Edition 10.17 ©Silberschatz, Korth and Sudarshan Database System Concepts - 6th Edition 10.18 ©Silberschatz, Korth and Sudarshan

Optimization of Disk Block Access (Cont.) Optimization of Disk Block Access (Cont.)

■ File organization – optimize block access time by organizing the ■ Nonvolatile write buffers speed up disk writes by writing blocks to a non-volatile
blocks to correspond to how data will be accessed RAM buffer immediately
● E.g. Store related information on the same or nearby cylinders. ● Non-volatile RAM: battery backed up RAM or flash memory
Even if power fails, the data is safe and will be written to disk when power
● Files may get fragmented over time 4
returns
4 E.g. if data is inserted to/deleted from the file ● Controller then writes to disk whenever the disk has no other requests or
4 Or free blocks on disk are scattered, and newly created file request has been pending for some time
has its blocks scattered over the disk ● Database operations that require data to be safely stored before continuing can
4 Sequential access to a fragmented file results in increased continue without waiting for data to be written to disk
disk arm movement ● Writes can be reordered to minimize disk arm movement

● Some systems have utilities to defragment the file system, in ■ Log disk – a disk devoted to writing a sequential log of block updates
order to speed up file access ● Used exactly like nonvolatile RAM
4 Write to log disk is very fast since no seeks are required
4 No need for special hardware (NV-RAM)
■ File systems typically reorder writes to disk to improve performance
● Journaling file systems write data in safe order to NV-RAM or log disk
● Reordering without journaling: risk of corruption of file system data

Database System Concepts - 6th Edition 10.19 ©Silberschatz, Korth and Sudarshan Database System Concepts - 6th Edition 10.20 ©Silberschatz, Korth and Sudarshan
 Flash Storage RAID
■ NOR flash vs NAND flash
■ RAID: Redundant Arrays of Independent Disks
■ NAND flash
● disk organization techniques that manage a large numbers of disks,
● used widely for storage, since it is much cheaper than NOR flash providing a view of a single disk of
● requires page-at-a-time read (page: 512 bytes to 4 KB)
4 high capacity and high speed by using multiple disks in parallel,
● transfer rate around 20 MB/sec
4 high reliability by storing data redundantly, so that data can be
● solid state disks: use multiple flash storage devices to provide recovered even if a disk fails
higher transfer rate of 100 to 200 MB/sec
■ The chance that some disk out of a set of N disks will fail is much higher than
● erase is very slow (1 to 2 millisecs)
the chance that a specific single disk will fail.
4 after 100,000 to 1,000,000 erases, erase block becomes
unreliable and cannot be used ● E.g., a system with 100 disks, each with MTTF of 100,000 hours (approx.
11 years), will have a system MTTF of 1000 hours (approx. 41 days)
– wear leveling
● Techniques for using redundancy to avoid data loss are critical with large
numbers of disks
■ Originally a cost-effective alternative to large, expensive disks
● I in RAID originally stood for ``inexpensive’’
● Today RAIDs are used for their higher reliability and bandwidth.
4 The “I” is interpreted as independent
Database System Concepts - 6th Edition 10.21 ©Silberschatz, Korth and Sudarshan Database System Concepts - 6th Edition 10.22 ©Silberschatz, Korth and Sudarshan

Improvement of Reliability via Redundancy Improvement in Performance via Parallelism

■ Redundancy – store extra information that can be used to rebuild ■ Two main goals of parallelism in a disk system:
information lost in a disk failure
1. Load balance multiple small accesses to increase throughput
■ E.g., Mirroring (or shadowing)
2. Parallelize large accesses to reduce response time.
● Duplicate every disk. Logical disk consists of two physical disks.
■ Improve transfer rate by striping data across multiple disks.
● Every write is carried out on both disks
4 Reads can take place from either disk ■ Bit-level striping – split the bits of each byte across multiple disks

● If one disk in a pair fails, data still available in the other ● In an array of eight disks, write bit i of each byte to disk i.
4 Data loss would occur only if a disk fails, and its mirror disk ● Each access can read data at eight times the rate of a single disk.
also fails before the system is repaired ● But seek/access time worse than for a single disk
– Probability of combined event is very small
4 Bit level striping is not used much any more
» Except for dependent failure modes such as fire or
building collapse or electrical power surges ■ Block-level striping – with n disks, block i of a file goes to disk (i
mod n) + 1
■ Mean time to data loss depends on mean time to failure,
and mean time to repair ● Requests for different blocks can run in parallel if the blocks
reside on different disks
● E.g. MTTF of 100,000 hours, mean time to repair of 10 hours
gives mean time to data loss of 500*106 hours (or 57,000 years) ● A request for a long sequence of blocks can utilize all disks in
for a mirrored pair of disks (ignoring dependent failure modes) parallel
Database System Concepts - 6th Edition 10.23 ©Silberschatz, Korth and Sudarshan Database System Concepts - 6th Edition 10.24 ©Silberschatz, Korth and Sudarshan
 RAID Levels RAID Levels (Cont.)
■ Schemes to provide redundancy at lower cost by using disk ■ RAID Level 2: Memory-Style Error-Correcting-Codes (ECC) with bit
striping combined with parity bits striping.
● Different RAID organizations, or RAID levels, have differing ■ RAID Level 3: Bit-Interleaved Parity
cost, performance and reliability characteristics ● a single parity bit is enough for error correction, not just
■ RAID Level 0: Block striping; non-redundant. detection, since we know which disk has failed
● Used in high-performance applications where data loss is not critical. 4 When writing data, corresponding parity bits must also be
■ RAID Level 1: Mirrored disks with block striping
computed and written to a parity bit disk

● Offers best write performance. 4 To recover data in a damaged disk, compute XOR of bits
from other disks (including parity bit disk)
● Popular for applications such as storing log files in a database system.

Database System Concepts - 6th Edition 10.25 ©Silberschatz, Korth and Sudarshan Database System Concepts - 6th Edition 10.26 ©Silberschatz, Korth and Sudarshan

RAID Levels (Cont.) RAID Levels (Cont.)

■ RAID Level 3 (Cont.) ■ RAID Level 4 (Cont.)
● Faster data transfer than with a single disk, but fewer I/Os per ● Provides higher I/O rates for independent block reads than Level 3
second since every disk has to participate in every I/O. 4 block read goes to a single disk, so blocks stored on different
● Subsumes Level 2 (provides all its benefits, at lower cost). disks can be read in parallel
■ RAID Level 4: Block-Interleaved Parity; uses block-level striping, ● Provides high transfer rates for reads of multiple blocks than no-
and keeps a parity block on a separate disk for corresponding striping
blocks from N other disks. ● Before writing a block, parity data must be computed
● When writing data block, corresponding block of parity bits must 4 Can be done by using old parity block, old value of current block
also be computed and written to parity disk and new value of current block (2 block reads + 2 block writes)
● To find value of a damaged block, compute XOR of bits from 4 Or by recomputing the parity value using the new values of
corresponding blocks (including parity block) from other disks. blocks corresponding to the parity block
– More efficient for writing large amounts of data sequentially
● Parity block becomes a bottleneck for independent block writes
since every block write also writes to parity disk

Database System Concepts - 6th Edition 10.27 ©Silberschatz, Korth and Sudarshan Database System Concepts - 6th Edition 10.28 ©Silberschatz, Korth and Sudarshan
 RAID Levels (Cont.) RAID Levels (Cont.)
■ RAID Level 5: Block-Interleaved Distributed Parity; partitions data and ■ RAID Level 5 (Cont.)
parity among all N + 1 disks, rather than storing data in N disks and ● Higher I/O rates than Level 4.
parity in 1 disk.
4 Block writes occur in parallel if the blocks and their parity
● E.g., with 5 disks, parity block for nth set of blocks is stored on disk blocks are on different disks.
(n mod 5) + 1, with the data blocks stored on the other 4 disks.
● Subsumes Level 4: provides same benefits, but avoids bottleneck
of parity disk.
■ RAID Level 6: P+Q Redundancy scheme; similar to Level 5, but
stores extra redundant information to guard against multiple disk
failures.
● Better reliability than Level 5 at a higher cost; not used as widely.

Database System Concepts - 6th Edition 10.29 ©Silberschatz, Korth and Sudarshan Database System Concepts - 6th Edition 10.30 ©Silberschatz, Korth and Sudarshan

Choice of RAID Level Choice of RAID Level (Cont.)

■ Factors in choosing RAID level ■ Level 1 provides much better write performance than level 5
● Monetary cost ● Level 5 requires at least 2 block reads and 2 block writes to write
● Performance: Number of I/O operations per second, and a single block, whereas Level 1 only requires 2 block writes
bandwidth during normal operation
● Level 1 preferred for high update environments such as log disks
● Performance during failure
■ Level 1 had higher storage cost than level 5
● Performance during rebuild of failed disk
● disk drive capacities increasing rapidly (50%/year) whereas disk
Including time taken to rebuild failed disk
4
access times have decreased much less (x 3 in 10 years)
■ RAID 0 is used only when data safety is not important
● I/O requirements have increased greatly, e.g. for Web servers
E.g. data can be recovered quickly from other sources
●
● When enough disks have been bought to satisfy required rate of
■ Level 2 and 4 never used since they are subsumed by 3 and 5 I/O, they often have spare storage capacity
■ Level 3 is not used anymore since bit-striping forces single block
4 so there is often no extra monetary cost for Level 1!
reads to access all disks, wasting disk arm movement, which
block striping (level 5) avoids ■ Level 5 is preferred for applications with low update rate,
■ Level 6 is rarely used since levels 1 and 5 offer adequate safety and large amounts of data
for most applications ■ Level 1 is preferred for all other applications

Database System Concepts - 6th Edition 10.31 ©Silberschatz, Korth and Sudarshan Database System Concepts - 6th Edition 10.32 ©Silberschatz, Korth and Sudarshan
 Optical Disks Magnetic Tapes
■ Compact disk-read only memory (CD-ROM) ■ Hold large volumes of data and provide high transfer rates
● Removable disks, 640 MB per disk ● Few GB for DAT (Digital Audio Tape) format, 10-40 GB with DLT
Seek time about 100 msec (optical read head is heavier and slower)
● (Digital Linear Tape) format, 100 GB+ with Ultrium format, and
● Higher latency (3000 RPM) and lower data-transfer rates (3-6 MB/s) 330 GB with Ampex helical scan format
compared to magnetic disks ● Transfer rates from few to 10s of MB/s
■ Digital Video Disk (DVD) ■ Tapes are cheap, but cost of drives is very high
● DVD-5 holds 4.7 GB , and DVD-9 holds 8.5 GB ■ Very slow access time in comparison to magnetic and optical disks
● DVD-10 and DVD-18 are double sided formats with capacities of 9.4
● limited to sequential access.
GB and 17 GB
● Blu-ray DVD: 27 GB (54 GB for double sided disk) ● Some formats (Accelis) provide faster seek (10s of seconds) at
cost of lower capacity
● Slow seek time, for same reasons as CD-ROM
■ Used mainly for backup, for storage of infrequently used information,
■ Record once versions (CD-R and DVD-R) are popular
and as an off-line medium for transferring information from one
● data can only be written once, and cannot be erased. system to another.
● high capacity and long lifetime; used for archival storage ■ Tape jukeboxes used for very large capacity storage
● Multi-write versions (CD-RW, DVD-RW, DVD+RW and DVD-RAM)
also available ● Multiple petabyes (1015 bytes)

Database System Concepts - 6th Edition 10.35 ©Silberschatz, Korth and Sudarshan Database System Concepts - 6th Edition 10.36 ©Silberschatz, Korth and Sudarshan

File Organization

■ The database is stored as a collection of files. Each file is a

sequence of records. A record is a sequence of fields.
■ One approach:
●assume record size is fixed
File Organization, Record Organization ●each file has records of one particular type only

and Storage Access ●different files are used for different relations
This case is easiest to implement; will consider variable length
records later.

Database System Concepts - 6th Edition 10.37 ©Silberschatz, Korth and Sudarshan Database System Concepts - 6th Edition 10.38 ©Silberschatz, Korth and Sudarshan
 Fixed-Length Records Deleting record 3 and compacting

■ Simple approach:
● Store record i starting from byte n * (i – 1), where n is the size of
each record.
● Record access is simple but records may cross blocks
4 Modification: do not allow records to cross block boundaries

■ Deletion of record i:
alternatives:
● move records i + 1, . . ., n
to i, . . . , n – 1
● move record n to i
● do not move records, but
link all free records on a
free list

Database System Concepts - 6th Edition 10.39 ©Silberschatz, Korth and Sudarshan Database System Concepts - 6th Edition 10.40 ©Silberschatz, Korth and Sudarshan

Deleting record 3 and moving last record Free Lists

■ Store the address of the first deleted record in the file header.
■ Use this first record to store the address of the second deleted record,
and so on
■ Can think of these stored addresses as pointers since they “point” to
the location of a record.
■ More space efficient representation: reuse space for normal attributes
of free records to store pointers. (No pointers stored in in-use records.)

Database System Concepts - 6th Edition 10.41 ©Silberschatz, Korth and Sudarshan Database System Concepts - 6th Edition 10.42 ©Silberschatz, Korth and Sudarshan
 Variable-Length Records Variable-Length Records: Slotted Page Structure

■ Variable-length records arise in database systems in several ways:

● Storage of multiple record types in a file.
● Record types that allow variable lengths for one or more fields such as
strings (varchar)
● Record types that allow repeating fields (used in some older data
models).
■ Slotted page header contains:
■ Attributes are stored in order
● number of record entries
■ Variable length attributes represented by fixed size (offset, length), with
actual data stored after all fixed length attributes ● end of free space in the block

■ Null values represented by null-value bitmap

● location and size of each record
■ Records can be moved around within a page to keep them contiguous
with no empty space between them; entry in the header must be
updated.
■ Pointers should not point directly to record — instead they should
point to the entry for the record in header.

Database System Concepts - 6th Edition 10.43 ©Silberschatz, Korth and Sudarshan Database System Concepts - 6th Edition 10.44 ©Silberschatz, Korth and Sudarshan

Organization of Records in Files Sequential File Organization

■ Heap – a record can be placed anywhere in the file where there ■ Suitable for applications that require sequential processing of
is space the entire file

■ Sequential – store records in sequential order, based on the ■ The records in the file are ordered by a search-key
value of the search key of each record
■ Hashing – a hash function computed on some attribute of each
record; the result specifies in which block of the file the record
should be placed
■ Records of each relation may be stored in a separate file. In a
multitable clustering file organization records of several
different relations can be stored in the same file
● Motivation: store related records on the same block to
minimize I/O

Database System Concepts - 6th Edition 10.45 ©Silberschatz, Korth and Sudarshan Database System Concepts - 6th Edition 10.46 ©Silberschatz, Korth and Sudarshan
 Sequential File Organization (Cont.) Multitable Clustering File Organization

■ Deletion – use pointer chains

Store several relations in one file using a multitable clustering
file organization
■ Insertion –locate the position where the record is to be inserted
● if there is free space insert there department
● if no free space, insert the record in an overflow block
● In either case, pointer chain must be updated
■ Need to reorganize the file
from time to time to restore instructor
sequential order

multitable clustering
of department and
instructor

Database System Concepts - 6th Edition 10.47 ©Silberschatz, Korth and Sudarshan Database System Concepts - 6th Edition 10.48 ©Silberschatz, Korth and Sudarshan

Multitable Clustering File Organization (cont.) Data Dictionary Storage

The Data dictionary (also called system catalog) stores
■ good for queries involving department instructor, and for queries metadata; that is, data about data, such as
involving one single department and its instructors
■ Information about relations
■ bad for queries involving only department
● names of relations
■ results in variable size records
● names, types and lengths of attributes of each relation
■ Can add pointer chains to link records of a particular relation
● names and definitions of views
● integrity constraints
■ User and accounting information, including passwords
■ Statistical and descriptive data
● number of tuples in each relation
■ Physical file organization information
● How relation is stored (sequential/hash/…)
Physical location of relation
●
■ Information about indices (Chapter 11)

Database System Concepts - 6th Edition 10.49 ©Silberschatz, Korth and Sudarshan Database System Concepts - 6th Edition 10.50 ©Silberschatz, Korth and Sudarshan
 Relational Representation of System Metadata Storage Access

■ A database file is partitioned into fixed-length storage units called

■ Relational blocks. Blocks are units of both storage allocation and data
representation on transfer.
disk
■ Database system seeks to minimize the number of block transfers
■ Specialized data between the disk and memory. We can reduce the number of
structures disk accesses by keeping as many blocks as possible in main
designed for memory.
efficient access, in
■ Buffer – portion of main memory available to store copies of disk
memory
blocks.
■ Buffer manager – subsystem responsible for allocating buffer
space in main memory.

Database System Concepts - 6th Edition 10.51 ©Silberschatz, Korth and Sudarshan Database System Concepts - 6th Edition 10.52 ©Silberschatz, Korth and Sudarshan

Buffer Manager Buffer-Replacement Policies

■ Programs call on the buffer manager when they need a block ■ Most operating systems replace the block least recently used
from disk. (LRU strategy)
1. If the block is already in the buffer, buffer manager returns ■ Idea behind LRU – use past pattern of block references as a
the address of the block in main memory predictor of future references
2. If the block is not in the buffer, the buffer manager ■ Queries have well-defined access patterns (such as sequential
1. Allocates space in the buffer for the block scans), and a database system can use the information in a user’s
query to predict future references
1. Replacing (throwing out) some other block, if required,
to make space for the new block. ● LRU can be a bad strategy for certain access patterns involving
repeated scans of data
2. Replaced block written back to disk only if it was
modified since the most recent time that it was written 4 For example: when computing the join of 2 relations r and s
to/fetched from the disk. by a nested loops
for each tuple tr of r do
2. Reads the block from the disk to the buffer, and returns for each tuple ts of s do
the address of the block in main memory to requester. if the tuples tr and ts match …
● Mixed strategy with hints on replacement strategy provided
by the query optimizer is preferable

Database System Concepts - 6th Edition 10.53 ©Silberschatz, Korth and Sudarshan Database System Concepts - 6th Edition 10.54 ©Silberschatz, Korth and Sudarshan
 Buffer-Replacement Policies (Cont.)

■ Pinned block – memory block that is not allowed to be written

back to disk.
■ Toss-immediate strategy – frees the space occupied by a block
as soon as the final tuple of that block has been processed
■ Most recently used (MRU) strategy – system must pin the
block currently being processed. After the final tuple of that block
has been processed, the block is unpinned, and it becomes the End of Chapter 10
most recently used block.
■ Buffer manager can use statistical information regarding the
probability that a request will reference a particular relation
● E.g., the data dictionary is frequently accessed. Heuristic:
keep data-dictionary blocks in main memory buffer
■ Buffer managers also support forced output of blocks for the
purpose of recovery (more in Chapter 16)

Database System Concepts, 6th Ed.

©Silberschatz, Korth and Sudarshan
See www.db-book.com for conditions on re-use

Database System Concepts - 6th Edition 10.55 ©Silberschatz, Korth and Sudarshan

Figure 10.03 Figure 10.18

Database System Concepts - 6th Edition 10.57 ©Silberschatz, Korth and Sudarshan Database System Concepts - 6th Edition 10.58 ©Silberschatz, Korth and Sudarshan
 Figure in-10.1

Database System Concepts - 6th Edition 10.59 ©Silberschatz, Korth and Sudarshan