We have a BI 7.0 system and the queries were running slow. We made a lot of tunning and the queries are running well now in the development environment. We will be using BI Accelerator in the Q and P environments going forward. I have the question in this regard.
1. We are using BI Accelerator and it will cut down any database access and improves the data read and hence increase the query data read time by almost 80% so the average running of any query will virtually run in about 20% of the time.
2. We have minimized the query result set by introducing mandatory selections, removing result rows, zero-suppress and no of free charcters.
3. We also have used Cache setup as Memory with swap etc and some persistance on application server. But the Persistance Accross Application Server or one server will insist on using Flat/Cluster tables, which need to be created in the application server(s). This will help only if the same result set is extracted again and again. Plus there will be a single file or table will be used per server (as it looks). My question is
What will happen when the same query is executed by different users at the same time or in close proximity with different selection criteria?