The Productionalization of Business Intelligence?

This is the continuation of the transcript of a Webinar hosted by InetSoft on the topic of "Is Quick Business Intelligence Better Business Intelligence?" The speaker is Mark Flaherty, CMO at InetSoft.

The next question is what's to keep productionalization of business intelligence from becoming another bottleneck indeed?

Well, it's an organizational issue at the end of the day. Productionalization is where IT needs to take responsibility to become responsive. And really what we are trying to do with speeding up the earlier stages of BI development and the design of the data warehouse is to enable IT to begin to focus on the parts that are farther down the delivery chain, in this case, listening to the end-users, becoming part of the end-user groups or working with the end-user groups that are prioritizing and are peer-reviewing particular queries or particular ways of analyzing the data and bringing that forward into the productionalization. But again, if you take all of the skills or all of the resources out of IT, it will become a bottleneck. So, it does need to be balanced. It needs to be seen part of the process.

Next question, does InetSoft work with SAP BW?

InetSoft works with quite a few data sources from file, relational databases, cubes, high-end data warehouses and so most of the time we connect to just about anything that’s out there. As it relates to SAP BW, yes, we do support connecting to BW directly.

 
 

Does virtualized data access mean the end of data warehousing?

One has to remember that the enterprise data warehouse and the architecture around it is very much a child of the 1980s, the technology of the ‘80s, the organization of the ‘80s and indeed, the business drivers of the ‘80s, and that doesn’t mean it's wrong. A lot of those things are still existing, and they are still part of the culture of many, many organizations. So I don't see it as really eliminating the data warehouse. What I see this doing is positioning the warehouse in a different place. In other words, going back to an earlier question, getting it out of the bottleneck mode, there is certainly data that needs to go into something like an enterprise data warehouse. And that will continue to be the case. But other data is going to flow around the outside, and I think that’s the way we have to deal with it.

It's going to mean extra work in terms of data management and in terms of understanding within that data. The old data warehouse architecture was very good in terms of getting to a single version of the truth, but these days I think there is more than one single version of the truth, there are multiple versions of truth, and we in IT and we the vendors who support business intelligence need to cognizant of that.

Next question, how does InetSoft handle security or data permissions?

InetSoft has really what I would call four layers of security from authentication, who are you, authorization, what reports are you allowed to see, data level security within the database what can you see or can't see and then finally, transmission security, how is that data transmitted, is it encrypted or not. So, each of those layers at organizations and apply certain levels of security from the more secure environments to fully supporting and the security that you have already defined in databases to keeping it all open, all of that is completely configurable by the organization.

Next question, what is meant by data warehouse automation?

The message here is very simple. We really need to move away from the thought that we can spend months preparing the data warehouse, preparing data marts, preparing queries or applications. I mean what we saw using InetSoft's BI application was an excellent view of how I can do this quickly, how I can make something very easy and agile for the end user. And IT has the opportunity to do this as well, and I talked a little bit about the hybrid agile BI technique in terms of building data warehouses. But I really do think that we need to put in place the processes over these coming few years, as opposed to automate various parts of the stack that are new to us. For example, when we look at big data as a source, and we have got stuff from another distributed processing system, we are going to have to automate that as well and make sure that there is an automated process from there into the warehouse or virtualizing it access across the warehouse and other parts of the environment.

Next question, what sources does InetSoft connect to?

It’s quite a long list of data sources from files, Excel, Access, text files to relational databases like SQL Server, Oracle, DB2, as well as high-end data warehouses like MSAS and Netweaver, and newer databases like Hadoop are all supported by InetSoft. And in lot of cases if we don’t have the database as a native driver, you can always use the ODBC driver and get to any of the other data sources that you might have in the organization.

Here is another question. Does everyone I want to share analysis with need to have a copy of InetSoft?

No, not at all. And in fact there are a couple of ways of sharing your analysis. One way is you can export a view as a PDF or a PowerPoint presentation and then e-mail it. The more common way is people will share their analysis through the Style Intelligence server. Just provide the URL or the folder name of the analysis, and people are able to consume that analysis through a Web browser, an iPad or even a mobile phone.