An analytic platform is arguably the next phase of business intelligence and analytics, and ParAccel has emerged with a solution that offers an integrated, extensible, scalable platform for analytics. In this year-end webcast with ParAccel, plus independent Analysts Mark Madsen and John O’Brien, we found out what an analytic platform should do and why it could change the future of analytics.
Here is a quick summary of some of the major points in this webinar:
- Madsen suggests that an Analytic Platform should be designed in a way that decouples it from the data warehouse; supports analytic development and maintenance processes; and doesn’t try to force-fit "offload" and "merge" patterns.
- O’Brien predicts that future analytic platforms will have better mixed workload capabilities as well as semantic integration at and above the data layer; he also suggests that the use of the predictive analytical programming language R, and the standard Predictive Modeling Markup Language (PMML) for analytic models will expand portability across platforms.
- ParAccel is a SQL database with a columnar engine that leverages a wide range of optimizations, methods for parallel processing, and integrations.
- ParAccel developed On Demand Integration (ODI) services to bi-directionally share data and processes across sources, i.e., enterprise data warehouse, Hadoop, machine data, streaming data, event capture, RFID tags, etc.
- ParAccel plans to take the Hadoop ODI a step further by providing integration with Apache HCatalog, Hadoop’s table and storage management layer, to simplify workflow and queries.
- Developing and deploying an ODI module can be done a matter of days.
Here is the URL to the archive:
Past Briefing Room episodes are available here.