In classic OLAP solutions, which usually store data on a hard disk, a key bottleneck is data access speed. Storing data in memory speeds up data access dramatically is a first step to performing real time OLAP analysis. In in memory business intelligence solutions, data access can be further accelerated, due to its in memory storage and smart data compression.
Another difference between in memory business intelligence solutions and classic OLAP solutions is that pre-calculations are not stored in the in memory data store, which means in memory solutions are an order of magnitude faster than the classic OALP solutions. The in memory solutions typically use queries to access the data store and run the required analyses, and send the data back to the user.
Beyond the speed advantage, in memory engine is not limited by external data formats, which allows for easier integration of different data sources. Additional configurations that optimize the data access and analytics processes to meet the challenges of real time processing include:
- Storage schema application
- Data compression
- Parallel processing
- Smart query algorithms
We have developed the ActivePivot system, which is an in memory business intelligence solution that uses bitmap indexing to map and access data and integrates different data compression and grouping options, configurable query algorithms, parallel processing and other capabilities that enable much faster analysis cycles. Additionally, ActivePivot’s independence from data sources enables data enrichment by calculating additional object properties and aggregating them within the hypercube.
This enrichment increases the depth of analysis possible with in memory analytics. For example, ActivePivot can analyze an event as it unfolds, analyzing the impact on the business as it changes in real time. Beyond analysis of a single event, ActivePivot can perform calculations on the aggregated values, after integrating numerous data sources.
To try out ActivePivot’s real time OLAP solution, see our live demo.