One thing to look at is how clustered are the full blocks.  Things look
great for access but if the blocks are clustered then the puts into the
cluster can be really slow (serial read to find free record for new entry).

-------------------------------------------------
Larry Simonsen                Phone: 801-489-2450
Flowserve Corporation     Fax: 801-491-1750
PO Box 2200                    http://www.Flowserve.com
Springville, UT 84663      e-mail: [log in to unmask]
-------------------------------------------------
All opinions expressed herein are my own and reflect, in no way, those of my
employer.

 -----Original Message-----
From:   Dave Darnell [mailto:[log in to unmask]]
Sent:   Wednesday, January 03, 2001 10:47 AM
To:     [log in to unmask]
Subject:        Re: Transact database program getting slower

I don't think that this is a high enough percent secondaries to really cause
a performance degradation (anyone who disagrees please respond.)  Average
chain length and it's variability look very nice.  I don't see any problems
here!  What am I missing everyone?

-dtd

> -----Original Message-----
> From: Chris Schofield [mailto:[log in to unmask]]
> Sent: Wednesday, January 03, 2001 10:17 AM
> To: [log in to unmask]
> Subject: Re: Transact database program getting slower
>
>
> Hello Everyone
>
> Thanks for the very many replies.
>
> Unfortunately I do not have adager or robelle software,
> therefore I have not
> got howmessy.  I have emailed Robelle about getting a copy.
>
> I do however have a listing produced by DBLOADSX from Beechglen.
>
> The program was run remotely and the info faxed to me.  I
> will list all the
> info
>
> Capacity 6000011
> Num Entries 3010823
> Load Factor 50.1
> Secondaries 22.1
> Max Pgs 0
> Block Factor 13
> Max Chain 10
> Avg Chain 1.28
> Std Dev 0.56
> Expected Blocks 1.00
> Avg Blocks 1.00
> Elongation 1.00
> Number of Primaries 2335771
> Number of Secondaries 675052
> Number of Chains 558256
> External Pointers 0
> Contiguous Pointers 675052
> Total Blocks 461540
> Full Blocks 15204
>
> I think I got everything !
>
> Is this any use ? Does this give anybody any clues ?
>
> Extra info.  This database started at 1 million capacity  and has been
> increased several times since then.  Could this be affecting
> the hashing
> algorithm ?  I read somewhere that the algorithm uses the
> capacity of the
> set.  Is this the original capacity perhaps ?
>
> Hopefully I will get a copy of Howmessy from Robelle
>
> Thanks again for all you ideas
>
> Chris
>