0% found this document useful (0 votes)
140 views

Physical Storage

The document discusses different types of physical storage media used in database management systems. It describes volatile and non-volatile storage, and the storage hierarchy from primary to secondary to tertiary storage. Magnetic disks and flash storage are discussed in detail, including their mechanisms, performance characteristics like access time, transfer rates, and reliability metrics like mean time to failure. Solid state drives which use flash storage are also covered.

Uploaded by

Arman Makhani
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
140 views

Physical Storage

The document discusses different types of physical storage media used in database management systems. It describes volatile and non-volatile storage, and the storage hierarchy from primary to secondary to tertiary storage. Magnetic disks and flash storage are discussed in detail, including their mechanisms, performance characteristics like access time, transfer rates, and reliability metrics like mean time to failure. Solid state drives which use flash storage are also covered.

Uploaded by

Arman Makhani
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 14

MC212 – Data Base Management System

Dr. Manish Khare

Physical Storage Systems


Classification of Physical Storage Media
 Can differentiate storage into:
 volatile storage: loses contents when power is switched off
 non-volatile storage:
 Contents persist even when power is switched off.
 Includes secondary and tertiary storage, as well as batter-
backed up main-memory.
 Factors affecting choice of storage media include
 Speed with which data can be accessed
 Cost per unit of data
 Reliability

Slide 2
Storage Hierarchy

Slide 3
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 and used for archival storage
 e.g., magnetic tape, optical storage
 Magnetic tape
 Sequential access, 1 to 12 TB capacity
 A few drives with many tapes
 Juke boxes with petabytes (1000’s of TB) of storage

Slide 4
Storage Interfaces
 Disk interface standards families
 SATA (Serial ATA)
 SATA 3 supports data transfer speeds of up to 6 gigabits/sec
 SAS (Serial Attached SCSI)
 SAS Version 3 supports 12 gigabits/sec
 NVMe (Non-Volatile Memory Express) interface
 Works with PCIe connectors to support lower latency and higher transfer rates
 Supports data transfer rates of up to 24 gigabits/sec
 Disks usually connected directly to computer system
 In Storage Area Networks (SAN), a large number of disks are connected by a high-
speed network to a number of servers
 In Network Attached Storage (NAS) networked storage provides a file system
interface using networked file system protocol, instead of providing a disk system
interface

Slide 5
Magnetic Hard Disk Mechanism

Schematic diagram of magnetic disk drive Photo of magnetic disk drive

Slide 6
Magnetic Disks
 Read-write head
 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.
 Cylinder i consists of ith track of all the platters

Slide 7
Magnetic Disks (Cont.)
 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
 Computes and attaches checksums to each sector to verify that data is
read back correctly
 If data is corrupted, with very high probability stored checksum
won’t match recomputed checksum
 Ensures successful writing by reading back sector after writing it
 Performs remapping of bad sectors

Slide 8
Performance Measures of Disks
 Access time – the time it takes from when a read or write request is issued to when data
transfer begins. Consists of:
 Seek time – time it takes to reposition the arm over the correct track.
 Average seek time is 1/2 the worst case seek time.
 Would be 1/3 if all tracks had the same number of sectors, and we ignore the
time to start and stop arm movement
 4 to 10 milliseconds on typical disks
 Rotational latency – time it takes for the sector to be accessed to appear under the head.
 4 to 11 milliseconds on typical disks (5400 to 15000 r.p.m.)
 Average latency is 1/2 of the above latency.
 Overall latency is 5 to 20 msec depending on disk model
 Data-transfer rate – the rate at which data can be retrieved from or stored to the disk.
 25 to 200 MB per second max rate, lower for inner tracks

Slide 9
Performance Measures (Cont.)
 Disk block is a logical unit for storage allocation and retrieval
 4 to 16 kilobytes typically
 Smaller blocks: more transfers from disk
 Larger blocks: more space wasted due to partially filled blocks
 Sequential access pattern
 Successive requests are for successive disk blocks
 Disk seek required only for first block
 Random access pattern
 Successive requests are for blocks that can be anywhere on disk
 Each access requires a seek
 Transfer rates are low since a lot of time is wasted in seeks
 I/O operations per second (IOPS)
 Number of random block reads that a disk can support per second
 50 to 200 IOPS on current generation magnetic disks

Slide 10
Performance Measures (Cont.)

 Mean time to failure (MTTF) – the average time the disk is expected to
run continuously without any failure.
 Typically 3 to 5 years
 Probability of failure of new disks is quite low, corresponding to a
“theoretical MTTF” of 500,000 to 1,200,000 hours for a new disk
 E.g., an MTTF of 1,200,000 hours for a new disk means that
given 1000 relatively new disks, on an average one will fail
every 1200 hours
 MTTF decreases as disk ages

Slide 11
Flash Storage

 NOR flash vs NAND flash


 NAND flash
 used widely for storage, cheaper than NOR flash
 requires page-at-a-time read (page: 512 bytes to 4 KB)
 20 to 100 microseconds for a page read
 Not much difference between sequential and random read
 Page can only be written once
 Must be erased to allow rewrite
 Solid state disks
 Use standard block-oriented disk interfaces, but store data on multiple
flash storage devices internally
 Transfer rate of up to 500 MB/sec using SATA, and
up to 3 GB/sec using NVMe PCIe

Slide 12
Flash Storage (Cont.)
 Erase happens in units of erase block
 Takes 2 to 5 millisecs
 Erase block typically 256 KB to 1 MB (128 to 256 pages)
 Remapping of logical page addresses to physical page addresses avoids waiting for
erase
 Flash translation table tracks mapping
 also stored in a label field of flash page
 remapping carried out by flash translation layer

 After 100,000 to 1,000,000 erases, erase block becomes unreliable and cannot be used
 wear leveling

Slide 13
SSD Performance Metrics

 Random reads/writes per second


 Typical 4 KB reads: 10,000 reads per second (10,000 IOPS)
 Typical 4KB writes: 40,000 IOPS
 SSDs support parallel reads
 Typical 4KB reads:
 100,000 IOPS with 32 requests in parallel (QD-32) on SATA
 350,000 IOPS with QD-32 on NVMe PCIe
 Typical 4KB writes:
 100,000 IOPS with QD-32, even higher on some models
 Data transfer rate for sequential reads/writes
 400 MB/sec for SATA3, 2 to 3 GB/sec using NVMe PCIe
 Hybrid disks: combine small amount of flash cache with larger magnetic disk

Slide 14

You might also like