calculate package cache hit ratio

[xixi]

the formula for package cache hit ratio is

1 - (package cache inserts/package cache lookups), what the result
would be a effective ratio? currently we have cache inserts=25, cache
lookups=35, so i get ratio =0.28, is this a good number?

[Pierre Saint-Jacques]

No it is not.
The formula can be used as: (1 -Inserts/Lookups) * 100
In your case, you get 28.57%. You aim is to get => 80%

Be careful. If the cache is empty, you'll obviously get a low ratio.
You should snapshot when the db has been activated, users have logged on
and have started using their application.

Normally a low ratio is an indication of a Package Cache too small.

HTH, Pierre
xixi wrote:

the formula for package cache hit ratio is

1 - (package cache inserts/package cache lookups), what the result
would be a effective ratio? currently we have cache inserts=25, cache
lookups=35, so i get ratio =0.28, is this a good number?

--
Pierre Saint-Jacques - Reply to: sesconsjunk at attglobaljunk dot com
Reconstruct address: Remove the two junk and replace at and dot by
their symbols.
IBM DB2 Cerified Solutions Expert - Administration
SES Consultants Inc.

[Pierre Saint-Jacques]

No it is not.
The formula can be used as: (1 -Inserts/Lookups) * 100
In your case, you get 28.57%. You aim is to get => 80%

Be careful. If the cache is empty, you'll obviously get a low ratio.
You should snapshot when the db has been activated, users have logged on
and have started using their application.

Normally a low ratio is an indication of a Package Cache too small.

HTH, Pierre
xixi wrote:

the formula for package cache hit ratio is

1 - (package cache inserts/package cache lookups), what the result
would be a effective ratio? currently we have cache inserts=25, cache
lookups=35, so i get ratio =0.28, is this a good number?

--
Pierre Saint-Jacques - Reply to: sesconsjunk at attglobaljunk dot com
Reconstruct address: Remove the two junk and replace at and dot by
their symbols.
IBM DB2 Cerified Solutions Expert - Administration
SES Consultants Inc.

[Tomas Hallin]

Not really - but on the other hand your sample is so small that it's
not particularly meaningful. I have these values:

Package cache lookups = 6368651
Package cache inserts = 396292

which comes out to around .94. I'd suggest you hammer your database
for a while before taking the snapshot so you get a better feel for
what's going on under load. The packages are cached (inserted) the
first time they're used, and the next time they're needed they will be
retrieved from the cache if they're still there (lookup).

/T

da****@yahoo.com (xixi) wrote in message news:<c0**************************@posting.google. com>...

the formula for package cache hit ratio is

1 - (package cache inserts/package cache lookups), what the result
would be a effective ratio? currently we have cache inserts=25, cache
lookups=35, so i get ratio =0.28, is this a good number?

[Tomas Hallin]

Not really - but on the other hand your sample is so small that it's
not particularly meaningful. I have these values:

Package cache lookups = 6368651
Package cache inserts = 396292

which comes out to around .94. I'd suggest you hammer your database
for a while before taking the snapshot so you get a better feel for
what's going on under load. The packages are cached (inserted) the
first time they're used, and the next time they're needed they will be
retrieved from the cache if they're still there (lookup).

/T

da****@yahoo.com (xixi) wrote in message news:<c0**************************@posting.google. com>...

the formula for package cache hit ratio is

1 - (package cache inserts/package cache lookups), what the result
would be a effective ratio? currently we have cache inserts=25, cache
lookups=35, so i get ratio =0.28, is this a good number?