erwin Data Modeler (DM) Review

Allows us to review databases with our business and technology people and to understand data relationships in our company

What is our primary use case?

The whole purpose of the erwin tool is for the designing of databases. We use it for our conceptual, logical, and physical database modeling.

How has it helped my organization?

We've been using this product as long as I can remember at our company, so it's hard to say how it has improved things. It's existed since I've been here. But it gives everybody the ability to see the physical implementations in a visual manner.

The solution is extremely critical to driving business change and transformation in our company because we do 100 percent of our data modeling using this tool. We meet with the business to show what exists and we show them what our changes are going to be to meet new requirements. We review that with business to get its agreement to the approach. We also meet with technology to show how it's going to be transformed in the physical implementation. So it is extremely critical to our everyday process.

What is most valuable?

The most valuable feature is the physical or visual representation of the database, showing the tables, the columns, the foreign keys, and the ability to generate DDL, so you can physically implement databases.

It lets you display the actual physically implemented databases or the logical databases. That enables you to review them with business users or technology people, to understand the relationships of the data throughout the company and show how data is joined together to achieve whatever the desired business results are.

What needs improvement?

I would like to see more support for working with the big-data world. There are so many new databases evolving and it's very hard for them to keep up with all of the new technologies. It would be good if they were able to dynamically support big-data platforms, other than Hive and Teradata. There's a new release coming out this year and they're adding two more platforms in that next release. So they are striving to keep up with technology, but technology is just evolving too rapidly. There are just too many options.

For how long have I used the solution?

I've been using erwin since 1998 or 1999.

What do I think about the stability of the solution?

The solution is very stable. It continues to evolve. 

A lot of the things, a lot of the new tools that they're introducing as part of erwin, are to make it more of a data governance tool in general, beyond just the data modeling which we've traditionally used. That whole piece is rapidly evolving. I've been watching it evolve over the past two or three years. We're not ready to purchase the products yet that they're putting out because we feel things are still evolving, but in the next couple of years they'll be the leader in the entire data governance realm.

What do I think about the scalability of the solution?

Data Modeler is very scalable. It keeps evolving as new technologies come out. People put in requests for it to be able to support different database platforms, and in just about every release in the past couple of years they've come out with support for one or two additional platforms. They are trying to keep up with customers' demands. They're very good about continuing to upgrade support for their legacy stuff as well. They're evolving and they're doing a pretty good job.

How was the initial setup?

The product was sold by a company called Platinum when I first started using it. It was then sold to CA and I was involved when CA produced its first rollout. After being owned by CA, it then was spun off to its own, standalone company, as erwin. I was there for the initial deployment of that as well. So, I've done many deployments of erwin, as different releases have come out.

The setup has become more complex. That's probably related to the fact that they're doing more things on the cloud, such as licensing, which has caused problems because we have very tight security here. Access to servers outside of our firewall causes issues for people who work in certain regions around the world because we restrict access to the cloud for governance purposes. So, we have had some issues with licensing. People can't connect to the license server, because it's done over the cloud, so they have to do an off-line license, which locks the license. We have concurrent licenses. So when a license is locked, even though the person is no longer using the product, because that person is not connected through the cloud, erwin's system doesn't know to release the license. We've worked with erwin and they have tried to help mitigate that, but we still do encounter issues with licensing.

In terms of deployment, just the install of the product on somebody's machine takes about 10 minutes. It's not very long at all. There are other features, such as setting up users in Model Mart, which take longer because you have to analyze the user's needs and set up appropriate permissions. That could take longer, depending on what the user's roles are.

As for our implementation strategy for Data Modeler, we just deployed it on someone's computer. We tested it on that person's box, one that everybody had access to. We all got to try the tool to see that we wanted to use it and to understand its features. Once everybody was comfortable with the features of it, we then had to upgrade our Model Mart repository, which is where we store all of the erwin models. Everyone has to be on the same release. So, we have to QA the whole process of upgrading our Windows Server and upgrading our database server. After we do those upgrades, we can then deploy the software on the machines. 

That's probably one of the biggest issues: Everybody has to be on the exact same version and release to be able to work together, if you're using the Model Mart repository. It's not very backward-compatible.

What about the implementation team?

We did have to involve erwin consultants because of the firewall issues that we were having when we were doing testing. We had to involve their helpdesk. Their helpdesk is extremely responsive. They actually tried to help us immediately on the phone. We needed a higher level of support so they scheduled meetings where we were sharing screens with them and they were able to help us. They were very helpful. One of the best features of erwin is its helpdesk.

What was our ROI?

It's hard to know how to gauge ROI. We've been using it since I got here. With the tool, we have a very good service-oriented architecture. We know exactly where all the data is; it's very clearly documented. If we didn't have this tool, I don't know how we would manage knowing where data is or manage having a consistent business glossary or data dictionary.

Which other solutions did I evaluate?

We've experimented with other solutions, such as ER/Studio, which has had different names. We experimented with SAP PowerDesigner, but that was not as robust in performing what we wanted it to do.

The main differences between the products we evaluated and erwin would be the ease of use, between logical and physical transformed models. The logical is more on the business side, and the physical is more on the technical side. The ease of maintaining those two models together was the number-one advantage of erwin. Number-two is the ability of the tool to support many platforms while successfully creating DDL, without issues with the DDL. erwin also has the ability to do compares of the models against the database, and to synchronize differences, whether importing database changes into the model or exporting the model to the database. Those were the primary things that it did well.

But I come back to ease of use. It's a very easy tool to train somebody on and for them to use. ER/Studio is probably the second-best product, but it's not as self-explanatory, it's not as easy to use. It's a little bit more clunky. It probably performs just as well, but it's a bit more difficult to use.

What other advice do I have?

If you want good data architecture in your company, you need to have database design done. It's probably the most important factor for having things clearly modeled and documented. erwin Data Modeler is not just a modeling tool, it's also used for documentation. If you're using the tool's functions properly, analyzing the documentation, flagging fields that are NPPI data, it is invaluable for business use. You can generate data dictionaries, you can make sure people are speaking common languages, and you can enforce company standards so that people are doing things in a consistent manner. It's an invaluable tool. If you want to have good data architecture, you need to have a tool like this.

We don't currently use the collaborative web modeling capability. We just recently purchased that tool and we are planning on deploying it at the end of Q1 of this year.

We don't use the erwin data transformation for integration to a wider ecosystem. We are actually able to directly do all of the transformations that we need from erwin, so we're not required to do any transformations. It supports legacy systems like Db2, Oracle, SQL Server, and now Teradata and Hive, which were introduced in the past few years. But it can currently support all of the data modeling we need to support, so no transformations are needed.

We have different flavors of people who use the tool. We have people who are dedicated data architects, that's their full-time job. There are 15 to 20 of them in the company. And we have many people who do use it for very specific applications on more of a part-time basis, where they're doing the data modeling and reviewing it with an enterprise architect. There are about 150 people who are doing that. Overall, we have about 170 people who have access to the software.

For deployment, upgrades, and maintenance of the solution, we generally require four people. We require somebody to do a Windows upgrade; we require somebody to do a database upgrade, and that's for the Mart repository portion; and we have two people who do the testing for the erwin tool: somebody who installs the upgrades of erwin on the local machines, and somebody who's testing it. When it comes to the installs and the upgrades, each person who's using the tool is expected to do that on their own. We set up a deployment package and everyone runs it when they're told to execute the upgrade.

Which deployment model are you using for this solution?

**Disclosure: IT Central Station contacted the reviewer to collect the review and to validate authenticity. The reviewer was referred by the vendor, but the review is not subject to editing or approval by the vendor.
More erwin Data Modeler (DM) reviews from users
...who work at a Insurance Company
...who compared it with SAP PowerDesigner
Add a Comment