1. Storage and File structure COP 4720 Lecture 20 Lecture Notes

2. Outline RAID levels Storage access File Organization Sec 10.3, 10.5, 10.6 Chapter 10: Storage and File Structures

3. RAID Levels Schemes to provide redundancy at lower cost by using disk striping combined with parity bits. Different RAID organizations (RAID levels) have differing cost, performances and reliability characteristics. Level 0: Striping at the level of blocks; non-redundant. Used in high performance applications where data loss is not critical. B1 B2B3B4 B5

4. RAID Levels Level 1: Mirrored disks; offers the best write performance. Popular for applications such as storing log files in a database system. Mirrored Disks

5. RAID Levels Level 2: Memory style Error-Correcting-Codes with bit stripping ECC bits Data disks bit striped Not used in practice

6. RAID Levels Level 3 : Bit-Interleaved Parity; a single parity bit can be used for error correction, not just detection. Parity bit disk Use the fact that the disk unit can detect error sectors

7. RAID Levels Level 4: Block interleaved parity. Uses block-level striping and keeps a parity block on a separate disk for corresponding blocks from N other disks. Block parity disk

8. RAID Levels Level 5: Block-Interleaved Distributed Parity; partitions data and parity among all N+1 disks. Data and Parity blocks Level 6: similar to level 5 but with ECC schemes

9. Storage Access A database file is partitioned into fixed-length storage units called blocks Database systems seek to minimize the number of block transfers between disk and memory. Buffer – portion of main memory available to store copies of disk blocks. Buffer manager – subsystem responsible for allocating buffer space in main memory.

10. Buffer Replacement Policies Most operating systems use LRU Not always the best strategy for queries join involves a repeated scan a mixed strategy with help from the query optimizer is often used pinned and toss-immediate

11. File Organization The database is stored as a collection of files one approach assume fixed record size each file has records of one particular type only different files are used for different relations

12. Fixed Length Records record i is stored starting at byte (i – 1) * n simple but may cross block boundaries deletion of a record i fill in the empty space with a shift left move last record to ith location keep a free space list.

13. Free Space List File header info about file and ptr to 1 st free location each free location points to the next free location

14. Variable Length Records Variable length records arise in database systems by: storage of multiple record type is a file record types that allow variable length fields record types that allow repeating fields can be represented by a byte string use special character to represent EOR difficulty with deletion difficulty with expansion

15. Variable Length Records: slotted page structure Header contains # of record entries end of free space in the block (after header) location & size of each record records can be moved around within the block update header pointers point to the entry position in the header

16. Variable Length Records: fixed length representation two techniques 1. reserved space -know the maximum length -short records are right filled with special character -best when most records have length near max

17. Variable Length Records: fixed length representation 2. Pointer method -pointers chain together block entries belonging to the same record -implemented with anchor block and an overflow block.