| Author |
A Performance Tuning question
|
|
| sid456 2002-03-09, 11:55 pm |
| Which two statements about row migration are true? (Choose two)
a. Row migration is caused by a PCTREE value set too low.
b. Row migration can be resolved using the ANALYZE command.
c. Row migration can be reduced by choosing a larger block size.
d. Row migration means that row pieces are stored in different blocks.
e. Queries that use an index to select migrated rows perform additional I/O
A is the first answer. Not sure about the
second answer, C may be?
Please help! | |
|
| Definately A for the first, and I'd be tempted to say E for the second but not 100% sure. | |
|
| i think A and D area the correct answers | |
|
| I'd disagree on answer D as row migration only places a 'pointer' to the new location of the data in the original block and doesn't split the row across multiple blocks.
Only row chaining splits the data across multiple blocks. | |
| vlegaspi 2002-03-15, 1:48 am |
| I'm gonna go with A&E. | |
| zbhana 2002-03-19, 8:40 am |
| A & E Definately. More I/O as a result of having to access to seperate rows. One to the pointer and the second to the actual data. High PCTFREE value helps avoid row migration as more space available for updates.
Row chaining needs larger block size. Row chaining involves rows being split up to accomodate larger data needs. |
|
|
|