I’m moderating a panel on BI, Data Governance and Data Modeling using Embarcadero ER/Studio on 16 November 2011, at noon EST.

Panelists include

• Kevin Phelps – Data Architect, American Century Investments, Inc., 
• Chris Bradley, Business Consulting Director, Information Processing Limited
• Jason Tiret – Director of Modeling and Architecture Solutions, Embarcadero Technologies

In this webinar, you’ll learn how to:

• Increase visibility and buy-in with business users
• Ensure smooth compliance with changing data privacy regulations
• Enable data stewards to meet requirements strategically

What makes this even more interesting is that audience members, like you, will have a chance to submit questions beforehand and during the event.

Details of the event:
Wednesday, November 16, 2011
9 a.m. Pacific, noon Eastern and 5 p.m. in the UK

When you register you’ll be asked to submit your questions.  Please do so.  We love having questions from YOU!

I’m deeming the hashtag for this event to be #DMExperts  I will try to take Tweeted questions, too. 

In working with Embarcadero ER/Studio (currently branded as ER/Studio XE), I often find that accidental system admins are confused by what appear to be duplicate or overlapping set up steps to get someone up and running with ER/Studio and the ER/Studio Repository.  This is especially confusing if the admin does this only a few times a year or does not have a documented process for setting up accounts.

Before getting into the details, let’s first define some of the terms for this how-to:

Embarcadero License Server: An application that runs on a server for managing the allocation of licenses for users of the ER/Studio suite of tools, including ER/Studio Data Architect, ER/Studio Business Architect, ER/Studio Software Architect, and Schema Validator. If you have concurrent licensing for ER/Studio, you must use the License Server to hand out licenses and keep track of how many are being used.  Other licensing schemes may or may not make use of the License Server.

ER/Studio Data Architect: The ER/Studio data modeling client application.  This application runs on your desktop/laptop.  Also known as the ER/Studio client.  This is the data modeling tool.

ER/Studio Repository: An application that runs on a server for managing versioning and releases of ER/Studio Data and Process models.  This is the service that allows you to check out a model and check it back in.

While the ER/Studio client can be deployed without these other services, enterprises typically deploy both the License Server and the Repository to help manage the complexity of enterprise projects.  However, to most modelers, it’s all magic to them: they run their local copy of ER/Studio and all these services work together seamlessly.

So when a user needs to be set up, they may need something to be set up in two locations:

1. The License Server
2. The Repository

And that’s where the confusion sets in, as it can appear they are having to set up users in duplicate.  It is even more confusing because organizations can customize how they manage licenses (letting anyone who has network to use a license, limiting licenses to specific machines or login IDs, etc.)  So I’ll leave the exact set up steps for another post. In my diagram above, that license server uses a concurrent user list to manage which machine logins (not ER/Studio logins, but Windows User Accounts)  are allowed to use a license.  Your licensing scheme may be different. 

The reason why users, accounts, or machines must be set up in the License server is that this services provides licenses for each of the ER/Studio suite of tools, not just ER/Studio Data Architect.   It’s also because an organization could deploy ER/Studio clients without the Repository or without the License Server.  They are two different services that can be used independently from each other.

Once an ER/Studio client application has obtained a license, a user can start working with models.  If that user is going to access the Repository, they need to have a login and a password in the Repository application.  Note that this is different than any data that was provided to the License Server.

To set up a Repository Account in ER/Studio Data Architect:

1. Run ER/Studio client.
2. Choose Repository from the menu bar, then Security and Security Center

   

3. You will be presented with the Security Center window.  Click on the Manage Users Tab.

   

4. Add the new User here.  You will want to use your organization’s standards for login IDs and passwords.
5. Once a User has been added, he can be assigned access to certain projects and diagrams via roles.

   

Once you have completed BOTH allowing the user to get a license (via the License Server) and set them up with a login (User Account) in ER/Studio Data Architect, they should be ready to work.

Special note: This post is part of TSQL Tuesday , a special blog posting monthly event based on a SQL Server/data topic chosen by one blogger.  I’d love to see more data architects be part of the blog conversations.

I frequently hear from project team members that they’ve never used the automation features of their data modeling tools.  Some of the reasons they give:

• I’m not a programmer.  It is not fair to expect me to to know how to program in order to use these tools.
• I have no idea how to use it and don’t have the time to find out
• Nothing I do can be automated; it’s all one-off tasks
• You can automate some of my work???

I think if you are giving these reasons without even trying, you a missing out on one of the niftiest features of your tools. 

I’m not a Programmer

Great! Neither am I.  The good news is that many tools don’t require you to have full blown programmer skills in order to automate data modeling tasks.  They have macro-like languages that require a bit of logic skills, but not much more.  For instance, Embarcadero ER/Studio XE uses a macro language called Sax Basic that is very similar to VBscript.  I’m lucky in that I used BASIC early in my career and am generally familiar with the language.  The toughest part is learning the functions, objects, and properties that are specific to ER/Studio, but thankfully there is a built in Help system that does a half decent job of helping you use them.

I Have No Idea How to Use It

That’s okay, because I don’t start with a blank macro when I go to automate a task; I just start with an existing one that is close to the same thing and tailor it to what I need to do.  I needed to export some meta data from my model to Excel, so I opened an existing macro that exported a bunch of data and tailored it to include the data I wanted, in the format I wanted it.

We host a mailing list/forum just for macros and automation of data modeling tools in our InfoAdvisors User Discussion Groups.  There are sample models that others have tailored and community members who are willing to help you through a tough part.  Also, most vendors have similar resources on their websites.

InfoAdvisors ER/Studio Macro

Nothing I Do Can Be Automated

I think you must be spending a huge amount of time clicking and waiting when you could be pushing a button and doing something else like grabbing a cup of coffee or answering a question on a forum…or even helping out a team member.  For instance, some of the macros that have been posted to our communities are:

ER/Studio Exporting All a Model’s Submodels as Images

ER/Studio Macro to Convert 1 Datatype to another Datatype

ER/Studio Macro to Add all Entities from a Model to It’s Submodel

ER/Studio Macro to Print PDFs of all Submodels Completely Unattended

ERwin Macro to Generate DDL for a DBMS not Currently Supported

Most of the macros I write tend to be to make some boring aspects of my job less boring by allowing me to do something else.  This means printing out my entire model, exporting images, making mass updates, etc.  If it has an algorithm that I can automate, I’m going to invest 15-20 minutes so that I don’t have to spend hours or even hundreds of hours over the course of a project doing those non-architecture tasks.

Another major use of automation I do is for setting properties for my DBAs.  Their standards and preferences should be automatable.  How FKs are named, how indexes should be names, which datatypes should be used, etc.  The fact that I can run a quick macro to do these and keep my DBAs happy, well, that’s priceless.  They love me for it <crickets, crickets>…well they should love me for it.

You can Automate Some of my Work???

Yes.  Not really the analytical parts, but some of the more mundane, boring, "change all these but not those" tasks.  Sure, finding and tailoring a macro takes time, but it is so worth it the next time you push a button, wander off to the coffee room to refill your 31 oz. Trenta cup with high-octane coffee.  Managers really don’t want you spending times on tasks that nearly anyone with mouse skills could do.  By using macros and APIs, you can add significant hours of productivity to your day.  Let’s also admit that computers are generally better at doing mass changes more accurately than we are. 

So let’s summarize:

1. Automating boring tasks makes you happier.
2. Happier Data Architects are better Data Architects
3. Automated recurring, boring tasks make bosses happier
4. Automating tasks makes for more accurate work
5. Saving time for you and your team members makes everyone happier.

A Lazy Data Modeler is a Better Data Modeler.

First, I’d like to thank all of you who registered for my Enterprise Data Modeling: 7 Mistakes you Can’t Afford to Make. This was an incredibly successful event, with:

• 723 Registrants
• 313 Attendees
• 90+% attendees who had the webinar "active"

That last stat, a guess at the average I saw as I was presenting, is a feature of GoToWebinar that measures at any point in time how many attendees have the meeting "in focus" on their screen.  I am also guilty of multi-tasking while attending webinars — checking my e-mail, chatting, conversing on Twitter, etc. — so I do understand that unlike an in-person presentation I can’t expect everyone to be completely focused on the slides during the whole hour.  I do think that this average means that people were actively engaged in what was happening for nearly all of the presentation.  I’m very happy with the registration and attendee numbers.  I think these are the highest I’ve ever attained for an online event.  I’m honoured that so many of you were able to make time in your busy schedules to spend with me.  Thank you.

One of the questions I got was about data modeling books I recommend.  These are the ones I mentioned:

• The Data Model Resource Book, Volumes I, II, and III – Wiley – Len Silverston and others – ISBN 0471380237, 0471353485, 0470178450
• Data Modeling Essentials, 3rd Edition – Morgan Kaufmann – Graeme Simsion and Graham Witt – ISBN 0126445516

Some of the questions I received via GoToWebinar:

Q: Any thoughts on the common pit falls in reverse engineering databases

Q: Is it possible for Karen to identify the pros/cons between Erwin and ER/Studio?

Q: I’ve heard other presentations which point out business users are intimidated or put off by "modeling". Do you have suggestions for making the topic accessible and valuable?

Q: What are the 7 mistakes?

Q: Can you advise if you have any tips on Industry Models to Launch your enterprise modelling with ER Studio? Does ER Studio Partner with any industry model suppliers?

Q: just a confirmation: when a non-Data architect/modeler (like a Business User) is given access to the model, the access is purely a read only, right?

Q: The reality with which I’m faced is that legacy databases use different terminology for the same entities. For example, in one table the entity may be called "deal" and in another table the same entity may be called "product" Does ER Studio provide a way to link both of these to a single entity in the logical model?

Q: What level of detail do you recommend that we display to the end users?

Q: If an organization is starting out or looking to start out with enterprise data architecture, where would you recommend that they start? Any books?

Q: Currently I create data models for specific application databases but I have noticed E/R Studio having sample models comprising of source and target systems. What is the recommend practice?

Q: In ER Studio, is the diagram considered to be a ‘report’ of what is in the model, or is the diagram the model itself?

Those were all great questions.  You can view the recording to see how I tackled them.  Also, I added 18 MORE mistakes to the end of the presentation as a bonus. 

We used the hashtag #7Mistakes on Twitter.  It’s harder for me to respond to questions on Twitter while I’m presenting, but I did respond during the webinar to a couple of them.  Again, listen to the recording to see what I said.

In the next few days you should be receiving the companion whitepaper that Embarcadero sponsored.  I hope you find it thought provoking.  I’d love to continue the conversation about these topics, even the "Wait, there’s More!" bullet points that I teased you with at the end of the presentation.

Perhaps we could discuss this during one of my upcoming Office Hours.