 | |
Hot Tables
sybase.public.ase.performance+tuning sybase.public.ase.performance+tuning
 |
| | Thread Tools | Display Modes |
#1
 
| |||
| |||
|
Hot Tables -
06-20-2006
, 03:51 AM
#2
| |||
| |||
|
|
1. How would one identify hot tables/indexes in a named data cache or in the default data cache. |
syslocks. Not very practical though. The MDA tables
might help too (although I've never looked at them
for this).
Quote:
|
2. I have a database which does fewer updates/inserts on history tables than select on the same tables. The tables are large...with more than 10million records (Largest being 50million). Would table partioning help to improve performance. What other perfomance tuning would you recommend. (NB Appropriate indexes have been defined on these tables). |
accessing these tables? Is their access distributed
or are they all trying to access the same data?
-am © MMVI
#3
| |||
| |||
|
|
Hi all, 1. How would one identify hot tables/indexes in a named data cache or in the default data cache. |
catch is that the information is since boot time and you might be interested
in a narrow time frame.) System table index usage does not seem to be
stored in the MDA tables.
Quote:
|
2. I have a database which does fewer updates/inserts on history tables than select on the same tables. The tables are large...with more than 10million records (Largest being 50million). Would table partioning help to improve performance. What other perfomance tuning would you recommend. (NB Appropriate indexes have been defined on these tables). Kind regards, Martin |
#4
| |||
| |||
|
|
Martin N Njeru> wrote in message news:4497b72a.207.1681692777 (AT) sybase (DOT) com... Hi all, 1. How would one identify hot tables/indexes in a named data cache or in the default data cache. 2. I have a database which does fewer updates/inserts on history tables than select on the same tables. The tables are large...with more than 10million records (Largest being 50million). Would table partioning help to improve performance. What other perfomance tuning would you recommend. (NB Appropriate indexes have been defined on these tables). 1. think of update statistics if you are not doing this periodically 2. parallism improves performance (worker process and parallel degree need to be configures) remember this requires additional memory / worker process 3. re-create clustered indexes atleast min once in 2 months (it will defragement) 4. analyse the locking schemes and think of going in for allpages since inserts are low. accurate queries use isolation levels to avoid dirty reads 5. if user-wide records are fetched depending on the where clause i would prefer to go in for views and granting required access.. this might help you. 6. go thru sp_cacheconfig and try to create cache and binding these few tables to a named cache 7. run sp_sysmon and try to figure out any bottleneck regards KRV |
|
« Previous Thread
|
Next Thread »
| Thread Tools | |
| Display Modes | |
Linear Mode
| |
Powered by vBulletin Version 3.5.3
Copyright ©2000 - 2013, Jelsoft Enterprises Ltd.
Copyright ©2000 - 2013, Jelsoft Enterprises Ltd.