![]() | |
![]() |
| | Thread Tools | Display Modes |
#1
| |||
| |||
|
#2
| |||
| |||
|
|
We are currently testing AS2K on a 64-bit 4 CPU Win2003 machine with 16GB of RAM. A complete process of one of our databases takes 140 minutes almost identical to the time taken on the 32-bit production server. An incremental update of the database takes 45 minutes on the 64 bit machine compared to 23 minutes on the 32-bit production server. The (32-bit) machine that launches the incremental process on the 64-bit machine is at a remote site connected by a fast pipe. We have turned off logging of the OLAP process. Any ideas why an incremental OLAP database update should be slower on a 64-bit server than a 32-bit server when a complete process takes the same time? TIA P Taylor |
#3
| |||
| |||
|
|
Have you taken the SQL queries executed from each and execute them interactively? That would be my first test. -- Dave Wickert [MSFT] dwickert (AT) online (DOT) microsoft.com Program Manager BI Systems Team SQL BI Product Unit (Analysis Services) -- This posting is provided "AS IS" with no warranties, and confers no rights. "ptaylor" <ptaylor (AT) discussions (DOT) microsoft.com> wrote in message news:7B785D45-A0D4-492A-8F6F-0CC675EA664D (AT) microsoft (DOT) com... We are currently testing AS2K on a 64-bit 4 CPU Win2003 machine with 16GB of RAM. A complete process of one of our databases takes 140 minutes almost identical to the time taken on the 32-bit production server. An incremental update of the database takes 45 minutes on the 64 bit machine compared to 23 minutes on the 32-bit production server. The (32-bit) machine that launches the incremental process on the 64-bit machine is at a remote site connected by a fast pipe. We have turned off logging of the OLAP process. Any ideas why an incremental OLAP database update should be slower on a 64-bit server than a 32-bit server when a complete process takes the same time? TIA P Taylor |
#4
| |||
| |||
|
|
Thank you Dave. On further examination it appears that there is a 30 second delay in connecting from the administrating server (which is at a remote site) to the 64-bit server. In the case of the incremental OLAP DTS package, there are around 40 separate steps each with its own connection and that seems to account for the time difference. In the case of the 32-bit production machine, the OLAP DTS package runs on a local machine so there is no connection delay. Hence the 32-bit production set up appears faster for incremental updates. "Dave Wickert [MSFT]" wrote: Have you taken the SQL queries executed from each and execute them interactively? That would be my first test. -- Dave Wickert [MSFT] dwickert (AT) online (DOT) microsoft.com Program Manager BI Systems Team SQL BI Product Unit (Analysis Services) -- This posting is provided "AS IS" with no warranties, and confers no rights. "ptaylor" <ptaylor (AT) discussions (DOT) microsoft.com> wrote in message news:7B785D45-A0D4-492A-8F6F-0CC675EA664D (AT) microsoft (DOT) com... We are currently testing AS2K on a 64-bit 4 CPU Win2003 machine with 16GB of RAM. A complete process of one of our databases takes 140 minutes almost identical to the time taken on the 32-bit production server. An incremental update of the database takes 45 minutes on the 64 bit machine compared to 23 minutes on the 32-bit production server. The (32-bit) machine that launches the incremental process on the 64-bit machine is at a remote site connected by a fast pipe. We have turned off logging of the OLAP process. Any ideas why an incremental OLAP database update should be slower on a 64-bit server than a 32-bit server when a complete process takes the same time? TIA P Taylor |
![]() |
| Thread Tools | |
| Display Modes | |
| |