IBM System z -- Customer/Business partner interview at the
Information on Demand Conference - 2009
Interviewee: Jo Coutuer, Numius - Business Intelligence on System z
Numius is a service organization, so we deliver consultancy services on the Cognos platform and now also on the IBM Information Server platform.
Many of our clients are facing challenges with degrading performance in the data warehouse environment . . . because the data volumes keep growing and they keep growing at an ever faster pace and the infrastructure often hasn't kept up with the growth of the data volumes. Another big problem or big challenge we are seeing is the challenge to communicate with business intelligence data to a large group of users.
So this is really a challenge to find a solution to keep growing, keep good performance, reach out to more people and still reduce cost of ownership and management cost.
Direct access to operational data as it happens, as the data is created, is really becoming vital and System z is really a dream platform for us to deliver that all.
The client we were dealing with is collecting large amounts of data from the retail space in Europe, collecting a lot of data from about 10,000 other companies and distributing and dealing with that data together with about 200 B2B partners. So they really have a challenge of data flows, large storage volumes and also communication. collecting large amounts of data from the retail space in Europe, . . . and distributing and dealing with that data together with about 200 B2B partners.
They were about to renew the entire infrastructure so they were really facing the decision of are we just going to do a copy paste from the old situation with more modern computers but keep the same architecture . . . or are we going to think out of the box for a minute and do things right.
We really took the idea to them because we realized that in the market it's not very well known today that Cognos which was always associated with distributed platform really is now an opportunity . . . to run that on System z so for them that was a new insight. We could construct a much simpler architecture with Cognos running on top of the WebSphere application server and the DB2 on z/OS.
And they were very skeptical and so we took up the challenge and said okay we will prove it, we will take your application, we will port it to System z and we will show you that your users can't even tell the difference.
Well they gave us the green light basically to convince them because they were very skeptical. You never love what you don't know. So they said we don't know what this is, we don't trust it.
We started by taking their data from a very heterogeneous environment running on Oracle, HP Unix, Windows, SQL Server so a real spaghetti of solutions and technologies and the first challenge was to take the data from another database Oracle and bring it to the DB2 platform. So we didn't just take up the challenge of moving Cognos to a different platform we said we will take the entire environment and bring it to a System z altogether so not just the front end layer of the BI but also the backend, the databases. So we did a database conversion that went flawlessly, thanks to a number of tools we could, IBM tools that we used. And we brought the data to the DB2 platform on System z. And then we took the Cognos environment and because Cognos is really not aware of the platform it runs on the software really doesn't care about the platform it runs on it turned out we didn't have to do any redevelopment at all.
What we really measured is at what point would we get the system to a degrading level of performance. And I am happy to say we never got to that point
We ran . . . a workload of 400 reports in the same time that it takes at the client to run one report.
They were really enthusiastic and surprised to see that their reports hadn't changed, their business logic hadn't changed, their data hadn't changed. It just went much smoother and better and in a stable way.
And most important they now have the opportunity to reach out to their clients to their B2B partners and to do really to fulfill the ambitions they have in terms of communication and sending BI, sharing BI with about 10,000 other parties, something that was totally impossible on their distributed platform.
We are dedicating and investing in terms of knowledge and people uniquely in the IBM stack because we believe in the strategic possibilities it can give to our clients.
Loading...
4:15
Introduction to Oracle Retail Merchandising Analyticsby oracleretail1,362 views
Link to this comment:
All Comments (0)