Skip to Content
avatar image
Former Member

How to increase Memory in Sybase and which parameter should affect ?

Hello All,

I want to increase Memory in ASE Sybase 15.7

Which parameter should require to reconfigure ?

Right now we have 48 GB Memory in Server and 21 GB allocated to Sybase DB but we are facing too slow in pick hours. planning to 16 GB increase and want to allocate 12 GB for Sybase but don't know how to reconfigure

Current Configuration

---------------------------------

OS - Windows Server 2008 R2

SAP - ECC6 EHP6

SAP Users - 900

Max Memory - 11320000 (2K)

Total Logical Memory - 3158101 (2K)

Add comment
10|10000 characters needed characters exceeded

  • Get RSS Feed

1 Answer

  • Oct 27, 2015 at 09:24 PM

    sp_configure 'max memory' will show you the maximum amount of memory ASE can (attempt to) request from the OS.

    To increase 'max memory' just bump up the current value (represented as a number of 2KB pages).

    Once 'max memory' has been bumped up you can add memory to various other configurations (eg, proc cache, statement cache, descriptor counts, data caches, etc).

    Add comment
    10|10000 characters needed characters exceeded

    • How to configure memory within ASE (eg, proc cache, statement cache, data caches, number of locks/connections/descriptors/etc) depends on your environment and where the memory will be of best use.

      One approach would be to bump up the big memory users (data caches, proc/statement cache) by the same percentage that you're bumping up 'max memory' (ie, from 21GB to 33GB is about 50%, so bump up your big memory users by 50%).

      If sp_sysmon is showing no performance issues with proc/statement cache then perhaps just putting all of the memory into your data cache(s) should be sufficient.

      If you've done some thorough analysis of MDA/OpenObjectActivity metrics perhaps you want to put some of that extra/new memory into a named data cache and bind some specific tables/indexes to this new data cache.

      The possibilities go on and on and on ... and they all depend on your specific environment and the associated performance metrics.