In Dual BI Architectures: The Time Has Come, Wayne Eckerson suggests that:

As an industry, it’s time we acknowledge the obvious: our traditional data warehousing architectures are excellent for managing reports and dashboards against standard corporate data, but they are suboptimal for managing ad hoc requests against heterogeneous data. We need dual BI architectures: one geared to casual users that supports standard, interactive reports and dashboards and lightweight analyses; and another tailored to hardcore business analysts that supports complex queries against large volumes of data.

This despite the additional cost of maintaining two environments. I agree that dual environments are a must, but use with caution, because not every traditional DWH has the kind of power users that require an analytic sandbox.

What is Analytics?

July 22nd, 2010

Sybase’s Phil Bowermaster is trying to shed some light on the question of What is Analytics?

The vital distinction is this: advanced analytics involves more than just slicing and dicing of the data. […] Ultimately, it’s this reliance on models that sets advanced analytics apart from other types of BI analysis. When a business takes a look at data to try to improve decisions and performance, that’s business intelligence. When a business compares incoming data with a model in order to achieve deeper understanding, deal with human behavior in real time, or predict what’s going to happen next, that’s advanced analytics.

Oracle is inviting for its Oracle Business Intelligence 11g Launch on 7. July. Charles Phillips, President, and Thomas Kurian, Executive Vice President, Product Development are going to be at the London launch event, and will be broadcasted to local events throughout Europe. I may go to the Zurich launch event.

Cloudera are looking at Considerations for Hadoop and BI in a little series of two articles (2nd part). BI tools traditionally were designed for small volumes of structured data where Hadoop generally stores data in complex formats at scale and processes data on read using MapReduce, so that can be quite a problem, and it’s good to see some guidance around this, because I know it’ll be one of the first questions when we look at it here too. Even though I guess our main interest would first be in storing large amount of non-relational data and query it with custom tools (i.e. MapReduce jobs), and BI only as an afterthought. BI tools is still how people think about this kind of problem.