|
Cluster aware
This is related to the level of flexibility, in terms of how a tool can be installed in a cluster. There are some tools, where any module can be installed in a server cluster (Crystal) OR where only one specific layer can be done. For a medium load application, this is not really a critical aspect, but for large scale implementations, it becomes important.
Enterprise Reporting Server Load Balancing
This is a bit self explanatory. A reporting tool has to have a good load balancing module which can manage the load on the servers related to various layers.
Cache
Should allow cache to be created by default. Cache is similar to ‘Cache’ files used by the internet explorer whereby it stores the files in the local hard-disk so that it does not have to go to the network every time it needs the same file. Similarly, the reports once generated are retained in the cache so that every time your need a report, you don’t have to process it again, and you can use the one existing in the cache. Some of the features are:
- Cache should be able to switch off and on: This enables you to cache, when you want to do. For example, when you don’t want a report multiple time, you may not need cache the same. Caching with all its advantage do take disc space and some processing power.
- The frequency of resetting the cache: This allows you to set the time over which the cache will be reset (deleted) for example, you may set-up the cache to be deleted after every 7 days, OR you could state the cache reset time for every different report.
- Provide page by page cache: With this feature, you should be able to get the cache in small increments. For example, if I am accessing a report of 1000 pages from the cache, it allows me the retrieve it page by page (OR at group level- for example retrieve a sales report by location and not to get the whole report of hundreds of location in one go). This reduces the retrieval time and memory requirements.
Logging
- Enterprise Reporting server execution log- This logs the programs OR objects executed by a server (Front-end layer server, process management layer server OR/and processing layer server)
- Trace log for debugging- For specific debugging purpose, one should be able to turn on the trace log, which helps you to look at each function step execution.
- Performance log- This the logs on how much time/resources were spent in completing the given number of tasks.
- Report generation log- This is similar to the report server execution log, but limited to the reports and queries and is more end-user friendly. It logs the reports, when they were generated, their destination and formats etc.
|
