Database News You Can UseJuly 2001
A monthly newsletter for Oracle users
from Database Specialists, a consulting firm
specializing in Oracle technology
In this issue:
Don’t miss the Northern California Oracle Users Group (NoCOUG) Summer Conference on August 23 at Chevron in San Ramon. Attendees are in for a treat as one of the session speakers will be the highly regarded Oracle performance tuning expert Gaja Krishna Vaidyanatha. He’ll be giving a two hour presentation titled “Oracle Performance Management – A Radical Approach.”
Gaja is the Director of Storage Management Products at Quest Software, and speaks regularly at IOUG-A Live! and Oracle OpenWorld where he is consistently ranked among the most popular conference speakers. He also has a new book published by Oracle Press entitled “Oracle Performance Tuning 101.”
Don’t miss out on this opportunity to see a great speaker at a very low cost. The meeting is free for NoCOUG members and only $40 for non-members. That even includes lunch!
Check out the full agenda for the day and register at http://www.nocoug.org/next.htm. We’ll see you there!
Oracle first introduced the multi-threaded server architecture in Oracle 7 as a way to support large numbers of concurrent users without huge memory requirements. Ever since then, DBAs have been asking tough questions. “Is MTS stable?” “Does it really reduce memory usage?” “Does MTS drag down performance?”
In his latest white paper, Database Specialists consultant Brian Keating presents an unbiased report of his experience deploying MTS in a production Oracle 8i environment.
Have you ever been downloading a huge file, just to have it stop halfway through and force you to start all over from scratch? Well, with NcFTP Client, you may be on your way to hassle-free downloading.
At Database Specialists’ last in-house DBA Roundtable, consultant Jay Stanley shared his positive experiences using NcFTP when he has large files to FTP. NcFTP Client (also known as just NcFTP) is a set of FREE application programs implementing the File Transfer Protocol (FTP). NcFTP will remember where your download stopped and hold your place while it keeps trying to reconnect to the file server and resume the transfer.
The NcFTP homepage states that “the program has been in service on UNIX systems since 1991 and is a popular alternative to the standard FTP program. NcFTP offers many ease-of-use and performance enhancements over the stock FTP client, and runs on a wide variety of UNIX platforms as well as operating systems such as Microsoft Windows and Apple Mac OS X.”
See for yourself at http://www.ncftpd.com.
There is a lot more to monitoring Oracle database performance than looking at cache hit ratios. But when you do want to measure a cache hit ratio, make sure you’ve got it right. Tune into Ixora’s free tips to find out more about “Calculating the Cache Hit and Miss Rates.” Be sure to read the paragraph regarding the twist in performing direct reads. You’ll find this tip at http://www.ixora.com.au/tips/tuning/cache_miss.htm.
Part of it is quoted here:
“Now for a twist. You have probably assumed that if your cache hit rate is 85%, then your cache miss rate is 15%. Not so. Oracle actually performs direct reads for certain operations. So it is possible to have an 85% cache hit rate and a 1% cache miss rate, with the remaining 14% being accounted for by direct reads. Direct reads are performed for parallel scans, and reads from temporary tablespaces. Blocks are read directly into private buffers in the PGA, rather than into the database buffer cache in the SGA. There are no cache hits, because blocks are not searched for in the cache before being read. And there are no subsequent cache hits, because the blocks are just discarded after use, rather than cached. However, this is no great loss. The possibility of getting enough cache hits in equivalent cached operations to actually save disk reads is almost negligible, while the possibility of losing cache hits because of the additional load on the cache is quite significant. So, direct reads actually improve the cache hit rate. They also improve block access concurrency by removing a significant load from the buffer cache latches. Incidentally, direct reads can also be obtained for serial scans using the _serial_direct_read parameter in 8.1, or event 10355 in earlier releases.”