Pivot Table Using 1.6Gigs of RAM When Connected To Access
I'm having performance issues with Excel Pivot Tables connected to a large Access table.
I manage two Access databases of ledger details that are used by 15 people both in the US and India. The first database processes the data (Extract, Transform, Load), and then outputs a final table to a second standalone database for users to connect to with Excel Pivot Tables over the network. The final table currently has 508,000 records with 42 columns, and with about 50,000 records added each month, I have 2 more months to go in this fiscal year (3/31 year end).
When the Pivot Table connects to the data, it scans the entire table and loads it into the Pivot Cache. Looking at the memory usage for the Excel Instance in Task Manager, it is exceeding a whopping 1.6 gigabytes.
To reduce the size of the Pivot Cache, i'm considering:
1. Creating queries based on the Output Table so users can connect to smaller data sets, or
2. Allow users to choose dimension elements in either a Form, List Box, or Combo Box that would then use VBA to limit the data returned to the Pivot Cache.
How should I handle it?
I manage two Access databases of ledger details that are used by 15 people both in the US and India. The first database processes the data (Extract, Transform, Load), and then outputs a final table to a second standalone database for users to connect to with Excel Pivot Tables over the network. The final table currently has 508,000 records with 42 columns, and with about 50,000 records added each month, I have 2 more months to go in this fiscal year (3/31 year end).
When the Pivot Table connects to the data, it scans the entire table and loads it into the Pivot Cache. Looking at the memory usage for the Excel Instance in Task Manager, it is exceeding a whopping 1.6 gigabytes.
To reduce the size of the Pivot Cache, i'm considering:
1. Creating queries based on the Output Table so users can connect to smaller data sets, or
2. Allow users to choose dimension elements in either a Form, List Box, or Combo Box that would then use VBA to limit the data returned to the Pivot Cache.
How should I handle it?
ASKER CERTIFIED SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
ASKER
Access pivot tables only allow up to 3 row fields which is wholly inadequate for my purposes.
Everyone uses Excel exclusively to review, manage, report, and share information. I'll use Access for its efficient query engine and data storage, but all in the service of extending Excel's abilities. We've no functional reason to use it as a front-end.
I parsed the data into queries that return smaller, more manageable record sets and instructed people to use the queries instead. Its an easy fix. I am also going to pursue a strategy of using a control sheet in our Excel workbooks to allow users to choose parameters to filter the data on the workbook level, and I'll use code to insert the parameters into the connection string.
Everyone uses Excel exclusively to review, manage, report, and share information. I'll use Access for its efficient query engine and data storage, but all in the service of extending Excel's abilities. We've no functional reason to use it as a front-end.
I parsed the data into queries that return smaller, more manageable record sets and instructed people to use the queries instead. Its an easy fix. I am also going to pursue a strategy of using a control sheet in our Excel workbooks to allow users to choose parameters to filter the data on the workbook level, and I'll use code to insert the parameters into the connection string.
ASKER
Actually I mis-spoke.
For simplicity's sake, I should always consider the existing tools in an application before developing a custom solution. I need to look at the functioning of Access pivot tables more closely.
Thanks for the tip Boaq2000.
For simplicity's sake, I should always consider the existing tools in an application before developing a custom solution. I need to look at the functioning of Access pivot tables more closely.
Thanks for the tip Boaq2000.
ASKER
Thanks for your replies.
I currently use Excel pivot tables for the flexible and fast drill-down feature, essentially producing a cube that behaves like Essbase. The Pivot Tables drive all of the user's reports, which they update by refreshing each month.
So, for example, an accrual team scans ledger balances firm-wide to determine if accruals are missing or wrong. They may need to investigate the details of 20 balances in 20 minutes. This tool makes it happen.
Some people write commentary on material month-to-month variances, wherever they appear. Often, the delta can only be understood by reviewing the transaction details. Others look at only individual businesses or account areas.
If I was just managing the process for myself, I'd just filter the data as needed when creating the pivot table. An ideal fix would be to detect the parameters selected in the Pivot Table and shape the pivot cache to contain only that data. So for example, if the Pivot Table had [Region] = "US" as the page field, an event would capture that choice and cache only those records.
Another idea i had was to see if a pivot table can use an unbound recordset that uses ADO to feed the cache every time the parameters of the Pivot Table are chosen. I want to make it simple if I can for the non-techies, but, not being a professional programmer, I lack the experience to know if this or that will work well.