Error updating db file on disk older man dating younger women tips
February 14, 2011 - am UTC The cause of a db file scattered read is simply that "we are doing a multi-block IO".The first sentence you have there "is the cause..." is one that I'm not able to fully parse and understand. If you have to do typical physical IO's you will either way for db file SCATTERED (multi-block) or SEQUENTIAL (single-block) IO.
These usually happen as a result of using an index.
Then it will usually be a read from a SORT (TEMPORARY) segment as multiblock reads for table scans (and index fast full scans) usually show up as waiting on "db file scattered read" A db file scattered read is the same type of event as "db file sequential read", except that Oracle will read multiple data blocks.
Multi-block reads are typically used on full table scans.
The more blocks we read in a single IO call - the more efficient it would be. Assume there are 100,000 rows in the table (100 rows/block).
Assume you are going to read 500 rows out of that table. Assume the data patterns are such that these 500 rows are on 500 different database blocks - they are not "clumped" together - they are spread out all over the table.
Meaning - if we see that on your system - we average "12" blocks in a multi-block read - we'll cost the queries using the value of 12 for dbfmbrc, but when we actually go to do the read - we max it out to the maximum read size on your system in the hopes we can get more read in.