Data modeling and database design for PepsiCo.
Data modeling and database design for PepsiCo.
We were using ERwin before we used ER/Studio. I've been doing data modeling for well over 20 years. Converting from one to the other at the time, ER/Studio was more sophisticated, gave us better functionality, and was more reliable.
The ability to visually represent what the database tables are going to look like and their relationships. Also, the ability to document the definitions of the tables and the columns that are in a table so that we can communicate what the data is and how it should be used.
The Repository, as well. We have a database design group of about 15 people. The Repository enables multiple people to work on the same model at the same time. We have fully integrated models, so we have one model for the enterprise data warehouse, one model for MDM, with a lot of sub-models in these things. People work on different parts of the model at different times. So the Repository enables us to share the models, and keep track of what everybody is working on.
What I would like to see improved is the reliability of the releases, releases that are a little less buggy.
Also, I'd like the ability to debug the errors ourselves instead of having to call them. There are certain types of errors that, I wouldn't say they come up regularly, but when you have an error it is very often the same type of error. Knowing that it's a Type III or Type I, it would be nice to have some kind of debugging facility for us to use, to find out where the problem that threw that error occurs. That would be a really cool feature.
It is mostly stable. We do throw errors every once in a while. We don't have a tremendous problem with stability, but we would always like to see Repository performance improve, checking in and checking out.
We haven't had any issues with scalability because the product works on individual workstations. We have shared licenses but everybody works on their own workstation. It's only when it goes back into the Repository that everything comes together. So, working individually is never a problem. You just need enough memory on your machine.
Tech support has been good. I have no complaints about it.
We haven't needed to use tech support in quite a while. But when we do, we have to go through whatever their process is, and if we haven't been able to resolve it through the regular process, then we get an individual person that we talk with directly to work with us to solve the problem. They have gotten onto our machines with us. They've been very responsive. They have their standard routine of what they have to do, but if that doesn't solve something, then they will get on with us individually or they ask us to send them the model and they figure it out.
We were using ERwin back in the early 2000s. Whenever we got a release, it was buggy. I didn't love ERwin at the time and we were looking for another solution. We had ER/Studio and Embarcadero come and talk to us. We got really excited about the tool, ER/Studio. It was going in a direction that sounded like it would be better for us at the time. It had more features that impressed us at the time and it was easy to convert.
It was 15 years ago. I think they came in and helped us when we did the setup. The setup of the tool itself is easy. The issues that we had were related to our environment. The tool is easy to set up but it depends on what your enterprise's architecture is. It's really about the Repository. The standalone workstations are easy. But if you're dealing with the Repository and enterprise architecture that makes it a little bit more complex. Then you have issues and you need their support.
I am not involved in the pricing. We have one person who does that. so I have no knowledge of what it is.
We have some individual licenses and then we have a concurrent license. And it's been working fine.
Try to take advantage of as many of the features as you can. It has a wealth of time-savers and features that are extremely useful. We haven't been able to use all of them because we already had so much done. To try and use some of them by going back over thousands and thousands of tables and columns would have been impossible. If you are is just starting with it, if you can take advantage of the Data Lineage and some of the other interesting features that it has, it is definitely worthwhile.
My rating comes from a purely selfish perspective, in the way I use it. For what I do, I give it an eight out of 10. The reason is - and this is partly our fault because we haven't finished, we haven't upgraded to newest release yet - that there are some buggy things in it that we still have to get resolved. For us, upgrading is very difficult because we have so many people, and we have to coordinate with our database people. Our services are now outsourced, so to coordinate things and get everybody on and get an upgrade done is not easy. Obviously, we try to avoid it as long as we can.