Indexes overview
- 1. © Copyright 2015. Apps Associates LLC. 1 Performance Tuning Overview March 30, 2015
- 2. © Copyright 2015. Apps Associates LLC. 2 Indexing
- 3. © Copyright 2015. Apps Associates LLC. 3 0..50 51..100 101..150 …. 10000.. 10050 0..10 11..19 20..25 …. 47.. 50 51..58 59..63 64..75 …. 98.. 100 10000.. 10009 10010.. 10020 10021..10028 … 10046..10050 0,rowid 0,rowid 1,rowid …. 10,rowid 11,rowid 11,rowid 12,rowid …. 19,rowid 10046,rowid 10048,rowid 10048,rowid …. 10050,rowid …. …. 10021,rowid 10022,rowid 10023,rowid …. 10028,rowid …. Create index I on T(numColumn)
- 4. © Copyright 2015. Apps Associates LLC. 4 B*Tree When to use • CLUSTERING FACTOR – A measure of how sorted the table is by the key in the index – It measures how many IO’s it would take to read the entire table via the index – row after row after row – If table is sorted by key, clustering factor near number of blocks in the table. – If table is not sorted by key, clustering factor nearer number of ROWS in table. – Please ask yourself, how many ways can the table be sorted on disk?
- 5. © Copyright 2015. Apps Associates LLC. 5 1,Alice 2,Bob 3,Candy 4,Doug 5,Ellen 6,Frank 7,George 8,Hank …. …. …. …. …. … … … … … …. 0..50 51..100 101..150 …. 10000.. 100500..10 11..19 20..25 …. 47.. 50 51..58 59..63 64..75 …. 98.. 100 10000.. 10009 10010.. 10020 10021..10028 … 10046..10050 …. Create index nm_idx on name) Select * from t where pk between 1 and 8 Pk Name 1 Alice 2 Sue 3 Victor 4 Will Pk Name 5 Irene 6 Kelly 7 Melanie 8 Oliver Pk Name 9 George 10 Candy 11 Uwe 12 Wally Pk Name 13 Ellen 14 Tom 15 Rick 16 Paul Pk Name 17 Doug 18 Irene 19 Lance 20 Jack Pk Name 21 Hank 22 Frank 23 Nicole 24 Bob
- 6. © Copyright 2015. Apps Associates LLC. 6 1,Alice 2,Bob 3,Candy 4,Doug 5,Ellen 6,Frank 7,George 8,Hank …. …. …. …. …. … … … … … …. 0..50 51..100 101..150 …. 10000.. 100500..10 11..19 20..25 …. 47.. 50 51..58 59..63 64..75 …. 98.. 100 10000.. 10009 10010.. 10020 10021..10028 … 10046..10050 …. Create index nm_idx on name) Select * from t where Name between ‘Alice’ and ‘Hank’ Pk Name 1 Alice 2 Sue 3 Victor 4 Will Pk Name 5 Irene 6 Kelly 7 Melanie 8 Oliver Pk Name 9 George 10 Candy 11 Uwe 12 Wally Pk Name 13 Ellen 14 Tom 15 Rick 16 Paul Pk Name 17 Doug 18 Irene 19 Lance 20 Jack Pk Name 21 Hank 22 Frank 23 Nicole 24 Bob
- 7. Access paths – Getting the data Access Path Explanation Full table scan Reads all rows from table & filters out those that do not meet the where clause predicates. Used when no index, DOP set etc Table access by Rowid Rowid specifies the datafile & data block containing the row and the location of the row in that block. Used if rowid supplied by index or in where clause Index unique scan Only one row will be returned. Used when stmt contains a UNIQUE or a PRIMARY KEY constraint that guarantees that only a single row is accessed Index range scan Accesses adjacent index entries returns ROWID values Used with equality on non-unique indexes or range predicate on unique index (<.>, between etc) Index skip scan Skips the leading edge of the index & uses the rest Advantageous if there are few distinct values in the leading column and many distinct values in the non-leading column Full index scan Processes all leaf blocks of an index, but only enough branch blocks to find 1st leaf block. Used when all necessary columns are in index & order by clause matches index struct or if sort merge join is done Fast full index scan Scans all blocks in index used to replace a FTS when all necessary columns are in the index. Using multi-block IO & can going parallel Index joins Hash join of several indexes that together contain all the table columns that are referenced in the query. Wont eliminate a sort operation Bitmap indexes uses a bitmap for key values and a mapping function that converts each bit position to a rowid. Can efficiently merge indexes that correspond to several conditions in a WHERE clause
- 8. © Copyright 2015. Apps Associates LLC. 8 Reverse Key Index • Physically stores the bytes of the keys in reverse order • In general, typically prevents range scanning – 5 is not stored next to 6 is not stored next to 7… • Used to avoid “hot block syndrome” • Can also be used to avoid having to re-organize a ‘sweeping’ index
- 9. © Copyright 2015. Apps Associates LLC. 9 Reverse Key Index - performance • It can help with buffer busy waits – On monotonically increasing values – Eg: sequence populated fields – Eg: date/timestamp fields • Especially in a clustered environment, this can reduce latency • At the cost of additional CPU – Evident mostly in single user mode – When you go multi-user, because you are not spinning on a buffer wait, you see less overall CPU
- 10. © Copyright 2015. Apps Associates LLC. 10 Reverse Key Index - performance • PL/SQL still more efficient • Reverse Key index has measurable impact for PL/SQL Single User PL/SQL Pro*C Reverse No Reverse Reverse No Reverse Transaction/second 38.24 43.45 17.35 19.08 CPU time (seconds) 25 22 33 31 Buffer busy waits number/time 0/0 0/0 0/0 0/0 Elapsed time (minutes) 0.42 0.37 0.92 0.83 Log file sync number/time 6/0 1,940/7 1,940/7
- 11. © Copyright 2015. Apps Associates LLC. 11 Reverse Key Index - performance Transaction/second 46.59 49.03 20.07 20.29 CPU time (seconds) 77 73 104 101 Buffer busy waits number/time 4,267/2 133,644/2 3,286/0 23,688/1 Elapsed time (minutes) 0.68 0.65 1.58 1.57 Log file sync number/time 19/0 18/0 3,273/29 2,132/29 PL/SQL Pro*C Reverse No Reverse Reverse No Reverse • PL/SQL still more efficient • Reverse Key index has measurable impact for PL/SQL in this case Two Users
- 12. © Copyright 2015. Apps Associates LLC. 12 Reverse Key Index - performance PL/SQL still more efficient Reverse Key index has impressively measurable effect As we scale up – this will have more of an impact At low concurrency, little impact Ten Users PL/SQL Pro*C Reverse No Reverse Reverse No Reverse Transaction/second 45.90 35.38 17.88 16.05 CPU time (seconds) 781 789 1,256 1,384 Buffer busy waits number/time 26k/279 456k/1,382 25k/134 364k/1702 Elapsed time (minutes) 3.47 4.50 8.90 9.92 Log file sync number/time 2,602/72 11k/196 12k/141
- 13. What basic statistics to collect • By default the following basic table & column statistic are collected – Number of Rows – Number of blocks – Average row length – Number of distinct values – Number of nulls in column
- 14. What basic statistics to collect • Index statistics are automatically gathered during creation and maintained by GATHER_TABLE_STATS and include – Number of leaf blocks – Branch Levels – Clustering factor • 12c – Tables have basic statistics gathered during loads into empty segments too
- 15. What basic statistics to collect • Histograms tell Optimizer about the data distribution in a Column • Creation controlled by METHOD_OPT parameter • Default create histogram on any column that has been used in the WHERE clause or GROUP BY of a statement AND has a data skew • Relies on column usage information gathered at compilation time and stored in SYS.COL_USAGE$ Histograms
- 16. What basic statistics to collect • 11g and before - Two types of histograms – Frequency – Height-balanced • 12c and after – two more – Top Frequency – Hybrid Histograms
- 17. Histograms • A frequency histogram is only created if the number of distinct values in a column (NDV) is less than 254 values Frequency histograms (FREQUENCY) Frequency histogram
- 18. Histograms • A height balanced histogram is created if the number of distinct values in a column (NDV) is greater than 254 values Height balanced histograms (HEIGHT BALANCED) Height balanced histogram
- 19. Histograms • Traditionally a frequency histogram is only created if NDV < 254 • But if a small number of values occupies most of the rows (>99% rows) • Creating a frequency histograms on that small set of values is very useful even though NDV is greater than 254 • Ignores the unpopular values to create a better quality histogram for popular values • Built using the same technique used for frequency histograms • Only created with AUTO_SAMPLE_SIZE Top Frequency (TOP-FREQUENCY) 12c
- 20. Histograms • Similar to height balanced histogram as created if the NDV >254 • Store the actual frequencies of bucket endpoints in histograms • No values are allowed to spill over multiple buckets • More endpoint values can be squeezed in a histogram • Achieves the same effect as increasing the # of buckets • Only created with AUTO_SAMPLE_SIZE Hybrid Histograms (HYBRID) 12c
- 21. © Copyright 2015. Apps Associates LLC. 21
- 22. Thank You!