DBMS Stores Data on Hard Disks
- Why not store everything id main memoery?
- Costs too much- Main memory is volatile to power off.
- Typical storage hierarchy:
- Main memory for currently used data.
- Disk for main database
- Tapses for archiving older versions
- We consider main memory and disk.
Disks and Files
- DBMS stores information on hard disks
- This has major implications for DBMS design!
- READ: transfer data from disk to main memory
- WRITE: transfer data from main memory to disk
- Called I/O operations: Both are high-cost operations due to mechanical components, relative to CPU operations, so must be planned carefully.
Disks
- Main advantge over tapes: random access vs. sequential
- Data is stored and retrieved in disk pages of fixed size, the smallest unit for data retrieval.
- Time to retrieve a disk page depends upon location on disk.
- Relative placement on disk has a major impact on performance!
Components of a Disk
block size == page size
Accessing a Disk page
- Time to access (read/write) a disk page:
- seek time (moving arms to position disk head on track)
- rotational delay (waiting for page to rotate under head)
- transfer time (actually moving data to/from disk surface)
- Seek time and rotational delay dominate
- Seek time varies from about 1 to 20msec
- Rotational delay varies from 0 to 10msec
- Transfer rate is about 1 msec per 4KB page
- Reduce seek/rotational delays through careful arrangement of data pages on disk
