12 Storage

12.1 Overview of Physical Storage Media

  1. Storage mediums are classified by:
    • speed at which data is accessed
    • cost per unit of data
    • reliability

\[\\[.1cm]\] b. Available Mediums: - cache - main memory - flash memory - magnetic disk storage - optical storage - tape storage

\[\\[.1cm]\]

  1. Cache - fastest & most costly; relatively small & managed by computer systems hardware

  2. Main Memory - storage medium for data that is readily availabe to be operated on

    • generally too small (too expensive) for entire large databases
    • can be lost when system crashes or in event of power failure

\[\\[.1cm]\]

  1. Flash memory - lower cost per byte than main memory but higher cost per byte than magnetic disks;
    • widely used in cameras, cellphones, & USB flash drives
    • non-volatile medium
    • popular substitute for magnetic disks in PCs
    • SOLIF STATE DRIVE (SSD) uses flash memory but its interface is also block-oriented like magnetic disk
    • block-oriented interface - allows data to be stored & retrieve in units of a block

\[\\[.1cm]\]

  1. Magnetic disk storage - primary medium for long term online storage of data
    • aka hard disk drive (HDD)
    • non-volatile but disks may fail & destroy data though rarely happens
    • to access data, systems moves data from disk to main memory & modified data must be written back to disk

\[\\[.1cm]\]

  1. optial storage - disks like DVDs where data is written & read back using a laser light source

  2. tape storage - used primary for back ups & archival data

    • cheaper than disks & can store data for many years
    • data must be read from beginning of tape, thus is called a sequential access storage

\[\\[.1cm]\]

  1. SEQUENTIAL ACCESS - read data from beginning

  1. DIRECT-ACCESS - read data from any location on medium

  2. PRIMARY STORAGE - fastest storage media but also most expensive

    • cache
    • main memory

\[\\[.1cm]\]

  1. SECONDARY/ONLINE STORAGE - flash memory & magnetic disks

  2. OFFLINE/TERTIARY STORAGE - magnetic tapes & disk jukeboxes

Storage Hierarchy

\[\\[.1cm]\]

12.2 Storage Interfaces

  1. disk interface standard types:

    • SATA (serial ATA) - supports data transfers speeds of up to 6 gigabits/sec
    • SAS (serial attached SCSI) - supports 12 gigabits/sec
    • NVMe (non-volatile memory express interface) - works w/PCIe connectors to support lower latency & higher transfer rates
    • support data transfer rate of up to 24 gigabits/sec

  2. Disks usually connect directly to computer

  3. STORAGE AREA NETWORKS - a large # of disks are connected by a high speed network to a number servers

  4. NETWORK ATTACHED STORAGE (NAS) - instead of networked storage appearing to be a large disk, it provides a file system interface using networked file system protocols such as NFS or CIFs

  5. CLOUD STORAGE - data is stored in the cloud and accessed via an API

12.3 Magnetic Disks

  1. Surface of platter divided into circular tracks
    • over 50K-100K tracks per platter on typical hard disks
  2. Each track is divided into sectors
    • A sector is the smallest unit of data that can be read or written
    • typically sector size is 512 bytes
    • Typical sectors per track: 500 to 1000 (on inner tracks) to 1000 to 2000 (on outer tracks)
  3. 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

\[\\[.1cm]\]

  1. Head-disk assemblies
    • multiple disk platters on a single spindle (1 to 5 usually)
    • one head per platter, mounted on a common arm

\[\\[.1cm]\]

  1. Cylinder i consists of ith track of all the platters

  2. 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

12.4 Performance measurement of disks

  1. 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

\[\\[.1cm]\]

  1. 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

\[\\[.1cm]\]

  1. 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

\[\\[.1cm]\]

  1. 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

\[\\[.1cm]\]

  1. 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

      Ex. 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

12.5 Flash Storage

  1. two main types: NOR flash vs NAND flash

  2. 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 and must be erased to allow rewrite

\[\\[.1cm]\]

  1. 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

\[\\[.1cm]\] d. Erase happens in units of erase block - Takes 2 to 5millisecs - Erase block typically 256 KB to 1 MB (128 to 256 pages)

\[\\[.1cm]\]

  1. Remapping of logical page addresses to physical page addresses avoids waiting for erase

  2. Flash translation table tracks mapping also stored in a label field of flash page

    • remapping carried out by flash translation layer

\[\\[.1cm]\]

  1. After 100,000 to 1,000,000 erases, erase block becomes unreliable and cannot be used

12.6 Optimization of Disk-block Access

  1. A database file is partitioned into fixed-length storage units called blocks. Blocks are units of both storage allocation and data transfer.

  2. Database system seeks to minimize the number of block transfers between the disk and memory

    • this can be done by keeping as many blocks as possible in main memory

\[\\[.1cm]\]

  1. BUFFER – portion of main memory available to store copies of disk blocks.

  2. BUFFERING- temporarily storing data read from blocks in an in-memory buffer, to satisfy future requests

    • done by both the operating system and the database system

\[\\[.1cm]\]

  1. READ-AHEAD - Read extra blocks from a track in anticipation that they will be requested soon

  2. Disk-arm-scheduling algorithms re-order block requests so that disk arm movement is minimized

  3. FILE ORGANIZATION- organize blocks in a way that corresponds closely to the way we expect data to be accessed

      Ex. If we expect a file to be accessed sequentially, then we should ideally keep 
  all the blocks of the file sequentially on adjacent cylinders.
    • disks hide the exact block location from the operating system but use a logical numbers blocks in consecutive order