Illustration podcast section

IBM i DevOps TechTalk
AI-Driven IBM i Application Analysis with ARCAD Discover #20

by the experts at ARCAD

In this Podcast, listen in as Arcad’s Senior Solutions Architects introduce ARCAD Discover, the ultimate visualization tool for IBM i applications:

  • Graphical insights: Instantly visualize applications, data flows & dependencies.
  • AI-powered analysis: Ask questions in natural language & get clear explanations.
  • Cross-team accessibility: RPG, Java developers & analysts gain instant understanding.
  • Fast & easy: Web-based, no install, real-time updates from ARCAD repository.

Listen to the Podcast

Spotify Podcast Badge
Apple Podcast Badge

The Story Behind the Mic: Podcast Transcription

Ray Bernardi – Welcome to IBM i DevOps TechTalk, where we discuss key topics and questions with ARCAD experts. I’m Ray Bernardi and I’ll be your host today. I’m also one of the experts today. I’m joined by Alan Ashley. He’s a senior solutions architect here at ARCAD software. We’ll be discussing ARCAD’s new product, ARCAD Discover. I’ll let Alan start the conversation.

Alan Ashley – Thanks, Ray. With DevOps and modernization really at the forefront of application development these days, particularly around legacy applications, one of the key focuses is the analysis. Which fields are impacted? Which tables are impacted? Understanding the application, particularly now that we’re starting to lose some of the historical knowledge from our developers on the i. What can we do about this?

R.B. – Our catch trying to answer that with a product called ARCAD discover. Now, this is a new product that’s coming out right now. It’s all about visualizing IBM i applications and doing that in a graphical form, which of course, a picture is worth a thousand words. Everybody has heard that before. So what our discoveries are all about is giving you a really unique way of taking a look at all of the objects, everything within an application, how they cross-reference to each other, and doing that all visually.
You can take a program and you can look at that program literally in graphical form using ARCAD discover. So it’s basically a tool, that’s a graphical display and a really efficient way to understand how your IBM i environment is built, the relationships between the different components and so on. With ARCAD discover, you’re capable of viewing an input output chart in seconds.
You can do a flow diagram on a program in seconds. There’s even high level views of the application. We call those macroscopic views where you can see how certain modules of an application relate to each other and so on. So there’s a lot going on there. And all of this works off of the ARCAD repository. So it takes that repository, it builds itself a data warehouse.
And then it can give you almost instant answers to how your application is built and how your application behaves on the IBM i. Now that you might be able to do that now with other tools. ARCAD Observer does that, but it’s pretty data oriented, you get data grids and so on. Again, this is graphical. And that’s the major difference here.

A.A. – Now one of the things we discover is it’s not necessarily always for the RPG developer. This could be used by a Java developer trying to understand how an application is built, or from a business analyst trying to figure out how to maybe document a project that’s coming in where they may not have access to the IBM i itself, but they can access this web page and go through it.

R.B. – Absolutely true. Alan, you and I deal with a lot of shops that are modernizing right now, and what they’re doing is they’re taking old 5052 interfaces, and they’re bringing those up to web type interfaces, but they’re still using the IBM i as the system of record.
So it’s still got the back-end database and that type of thing. That’s where somebody who’s coding that new front-end can almost instantly and graphically take a look at a set of programs, how they work, what they’re doing and understand. Even if they can’t read RPG code, they can understand what’s going on within that application and they can start to build the new modern front end that mimics that application.

A.A. – Now, we talked about and you’ve mentioned this, that you can get this information almost instantly. Anybody that’s familiar with ARCAD observer knows you can go through. And you can right click and look at utilization and it’ll look at right click from there and go in and look at some different other information. What makes ARCAD discover a little bit more special when it comes to finding that information? Does it use any kind of AI?

R.B. – It does. It has AI engines built into it, and there’s several different ways you can use AI. One of the things that you can do is you can ask AI a question. And this is in the area of architecture of your applications. So for instance, if you’ve seen any of our webinars and discover you’ve seen the architecture, tab where you click on that and you start to examine the architecture of your application.
It used to be that. You would have to click on a program and say, start here and show me what’s related and then go from there, etc. but with the new AI components built into discovery, you can ask a question. You can say, show me all the features related to or that will update the country file, for instance, and you press the enter button or the send button.
It’s like you’re chatting with the application and what it does. Like I said this, it’s got that data warehouse built from the repository.

A.A. -It’s the metadata. It’s got all that metadata out there. That’s how it all ties together, which is what makes ARCAD so strong is it’s got that metadata that tells you where everything is and how it all relates.

R.B. – That’s right. That’s exactly what it’s all about. But again, I mean, it does it almost instantly. It shows you what you’re looking at. And it will understand your application and show it to you graphically from you asking a question in human form. You’re talking to it conversationally. The other thing that it can do in the AI area is when you’re examining a program.
You gave the example just a few seconds ago. You asked about Java developers, for instance. We talked about modernizing a front end. So picture me a Java developer here trying to understand what this program is doing. I get some graphical pictures of it and how it flows and so on. But I can also get paragraphs written out from the AI that describe it in plain English.
What the program is doing this program opens the item master file. It then uses this key to search the item master file. It will allow these functions. It’s all spelled out for you. That is AI at its best.

A.A. – That right there can help a Java developer that is maybe trying to tie into a table and replace an old legacy RPG program, and he needs to know what is the input. Where’s that input hook. And by looking at this and it’s literally telling him this module does the right to this table. And that’s how you know that’s the access end to that table.
And that’s something that can be very valuable. Particularly like I said, we’re starting to lose some of our historical programmers. They’re starting to retire. And this is a great way to understand your application, not only for your RPG developers, but for your other developers as well.

R.B. – Yes, it comes into play in a lot of ways to be honest. Natural language queries like we just mentioned. We just talk to it in English. It’s going to automate creation of complete functional maps of your application. You ask it a question and it finds the components. It builds the diagram. It displays the diagram.
You’ve got no work to do. It’s suddenly there. It’s going to help with retro documentation. It’s going to help you with database analysis. There’s tons of things that I can do. It’s all about that impact analysis to a change. What’s going to happen. And being able to gain that information extremely quickly simply by asking a question.

A.A. – I guess the the bigger question here is, is do you have to be an ARCAD user today to make use of ARCAD discover? Is it a standalone or does it have to be included with the larger ARCAD DevOps suite?

R.B. – It doesn’t have to be included with the larger DevOps suite. We need the ARCAD repository, the metadata. So in order to use discover, you are going to go through the process of building an ARCAD repository, which is basically a batch job that you run. We do offer some analysis tools so that you can look at your environment before you run it, so you get good metadata.
That’s called ARCAD audit, if I recall correctly. So you can run an audit, you can clean things up a little bit, you can load that repository and then that repository is used by discover as its data warehouse. And you can sync that as many times as you want. And keep it up to date and so on.
But that’s the instantaneous aspect of discover, through. And that’s why it’s so quick and so fast. When you ask it questions, it’s already got all the information it needs and off it goes.

A.A. – One of the other things that I’ve seen and discover is the ability to tie in to some of the other products that ARCAD has, such as using the healthcheck to tie in and look at information provided by CodeChecker, or the larger ARCAD DevOps suite where you can, or the drops product as well. So you can start tying those all together and discover does have a link to that.
So that can be set up. So you can look at a Grafana dashboard to see how it all ties together. Maybe you want to see it in a slightly different light. So you have that option as well. If you have some of those other products available to you.

R.B. – This is true. We do have offer application health heck and I can be used to help you find things like inconsistencies in your code, redundancies in your code, obsolete programs, etc.. And these are all critical metrics that you need to understand what’s going on out there in your organization, things like the complexity of a program, how often where’s my database being access the most things like that the volume and so on. So we can monitor all of that at an extremely high level with our dashboards, and we can show you that information and you can make informed decisions. As to where to focus your time to make sure that the health of your application is maintained.

A.A. – And particularly when it comes to the modernization efforts, you can start to see those almost like a pinwheel of how many times something is used. And let’s get rid of some of this and start dialing that wheel back. And you can start to develop projects to clean that up. Now, one of the things that I think, I don’t know if we mentioned that: this is a web interface.
There’s no studio to install. It’s just a server component that runs on the i or windows. And then it’s just a web interface. So I think we left out that kind of important part there.

R.B. – This is true. It’s very easy to maintain. There’s nothing to maintain. As long as people have an active browser they can they can access it. You give them the URL and they’re off and running.

A.A. – So we’ve talked about AI and how useful it is and how impactful it can be in understanding what you’re looking at. Are there any other products that are coming down through the ARCAD pipeline that are going to start gaining some AI help?

R.B. – I know that we’re exploring AI in many of our products. I don’t have the inside knowledge on exactly what R&D is doing on some of these. You might know a little bit more of it than I do, believe it or not, Alan, you’ve worked a little bit closer with R&D than I have, I believe.

A.A. – Yes, I know that there’s and I don’t want to give away too many secrets here, but I know the CodeChecker here, and, our new product transformer Microservices is starting to look at that to see how it can help. But one thing that, ARCAD is aware of like many companies, AI is just a baby.

R.B. – Sure it is.

A.A. – It is still learning and and we’re still learning how to make the most of it. So when we’re going through and we’re seeing how I can work with discover or work with other products, we’re keeping in mind that, is that really the best way to gather that information? Is it mature enough to give it a proper answer?
So that doesn’t lead a consumer down the wrong path? And so that’s what we’re very cognizant of. And so you may hear during the years of oh we’re looking at doing it in CodeChecker or we’re looking at doing it a microservices. We may find out through some of our testing that this isn’t giving us good results because the language model isn’t mature enough yet.
So that’s something to keep in mind. And that’s just a broader kind of AI, discovery and you can have a whole discussion on the ethics of AI, but that’s where a different form there. But as you’ve heard discover, it’s a pretty simple install. You run your audit, you build your application in this repository of what’s going to be there. The cleaner it is, the better the information, which is always the case. And you’re off and running and you can start to see how your application is really built.
Because I remember when I was a system admin way back in the late 1900s, they printed out on one of the wide format printers how the application that we used was all hooked together. And it was thousands of pages of how it was all hooked together. And this you can see it in one nice graphical picture. This ties to this, this ties to this. Oh here’s three inputs here. Here’s two inputs there. That would have been fabulous to have. When you’re trying to understand the impact in your application. What does this change in this one program do. It impacts a lot of stuff.

R.B. – Like you said discovery shows you the repository. You can see relationships between all the components within an application. Discover graphically shows them to you. You’ve got flow diagrams. You’ve got input output diagrams. You’ve got database relationship diagrams built into the thing. And again the AI assistant allows you to make those very quick and very fast. So you’re very efficient when you’re using this tool.
We need to end the conversation there today. We’re running out of time. But as you can see, ARCAD discover is ready to revolutionize application analysis for IBM i applications.

Our Hosts

Cécile Masson

Ray Bernardi

Senior Consultant, ARCAD Software

Ray is a 30-year IT veteran and currently a Pre/Post Sales technical Support Specialist for ARCAD Software, international ISV and IBM Business Partner. He has been involved with the development and sales of many cutting edge software products throughout his career, with specialist knowledge in Application Lifecycle Management (ALM) products from ARCAD Software covering a broad range of functional areas including enterprise IBM i modernization and DevOps.

 

Alan Ashley

Alan Ashley

Solution Architect, ARCAD Software

Alan has been in support and promotion of the IBM i platform for over 30 years and is the Presales Consultant for DevOps on IBM i role with ARCAD Software. Prior to joining ARCAD Software, he spent many years in multiple roles within IBM from supporting customers through HA to DR to Application promotion to migrations of the IBM i to the cloud. In those roles, he saw first hand the pains many have with Application Lifecycle Management, modernization, and data protection. His passion in those areas fits right in with the ARCAD suite of products.