In this episode of BMC AMI Z Talk, Anne Hoelscher, Director of R&D for BMC’s Db2 solutions, and Craig Mullins, President of Mullins Consulting and widely known Db2 thought leader discuss the need for intelligent automation in modernizing Db2 data management to help companies achieve greater availability, resiliency, and agility. Below is a condensed transcript of our conversation.
Anne: Craig, thank you for joining us today. Can we get started by having you share a little bit about yourself and your Db2 background as well as your data management background?
Craig: Sure Ann, and thanks for having me on the podcast today. I look forward to chatting about what’s going on these days in the world of Db2 and data management. So, about me. I’m currently an independent consultant, have my own company, Mullins Consulting and there I specialize in data and database management research and strategy with a big focus on mainframes and Db2. Now, I’ve been working with data for over three decades and during that time I’ve taught database classes. I worked as a systems analyst and programmer. I spent a lot of time as a DBA, first working with IMS and then with Db2, but I’ve also worked with some other database systems like Oracle and Microsoft SQL Server.
I spent a little bit of time at Gartner as an analyst covering database systems and DBA, but really, I’m proud to say I’ve been working with Db2 on the mainframe since Version 1, so seen it grow up before my eyes. I’ve also written a couple of books on Db2 as well as a book on database administration practices and procedures and I’ve also been recognized by IBM as a Champion for Data & AI as well as an IBM Db2 Gold Consultant. And, Analytic Suite magazine recognized me as one of the top 200 thought leaders on big data and analytics. So, that’s me in a nutshell.
Anne: Thanks, Craig. That’s great. So you said you’ve been on Db2 since Version 1. That was a long time ago and I remember when segmented table spaces came out and some people thought the world was going to end. But what do you think has been the biggest impact to Db2 management you’ve seen over all that time?
Craig: Wow. That’s a long period of time, 30 plus years. I think there are several things. I would say things are more heterogeneous today than they’ve ever been. When I started out most organizations ran most of their mission-critical work if not all of it on the mainframe. These are large organizations. And then as we move forward into the ‘90s things like client servers and into the 2000s things like node SQL and then the wide adoption of the internet as a platform have really changed things. So when I started the ‘80s heads down CRTs, green screens, everything running in an insular mainframe environment and that’s not the case anymore.
Today organizations, large ones anyway, still store the bulk of their data on the mainframe and that means mainframe data management skills are at a premium, but they’re accessing it not just with CICS and IMS and not just with PSO and Batch, but also distributed from multiple platforms. A lot of things coming in over the web, WebSphere being booked into this as well and moving from an environment where almost everything is static SQL prebound and we know the access path before they’re run.
To today where almost everything that’s new is coming in through DDF and distributed applications and that means dynamic SQL and we don’t know those access paths to runtime and that’s a little bit more of a challenging environment. So lots of things have changed, but I think that heterogeneity as well as the static to dynamic shift are two of the biggest.
Anne: So that definitely makes our modern data management much different than what it’s been in the past. How do you see DevOps impacting the jobs of Db2 professionals?
Craig: Well, I think with DevOps it means that the DBAs need to be more closely tied to the applications. Again, go back to when I first came into this business in the ‘80s and the DBA was the curmudgeon in the corner back there smoking a pipe and there’s pipe smoke coming out and you’re afraid to go into his cubicle. Now, it’s the DBA has to have a presence, has to be known, has to be a communicator. No longer a curmudgeon, but someone who is adept at working with people, has to actually have a personality. And that is a significant change, but that’s really not the only change with DevOps because DevOps is orchestration and implementation of the entire software development lifecycle.
So as we are adopting DevOps, we are automating. And one of the things I’ve seen as a data bigot, a guy who loves data, is that we had organizations spend a lot of time automating for the application developer and that’s good. I’m not saying there’s anything wrong with that. We’re seeing things like Jenkins for coordinating and orchestrating and Ansible for configuration resource management and GitHub for source control and Docker to containerize. And what we need really is the same type of tools to automate and integrate DBA into the DevOps pipeline.
One of the things I’m really fond of saying, DevOps has become more DEV Ops. The focus is on Dev. We need Dev and Ops to be at the same pitch and volume.
Anne: That’s a great distinction because the DBAs do fall more into that Ops section than the development section. So as this is all changing, what do you think are the most pressing issues that IT leaders are facing with probably both managing their professionals as well as their data and the volumes of data in their environments?