CA and Cybermation and workload automation

Let's see what the two can do

Comment While I may focus on data and things to do with data, my tastes tend to be more eclectic than that so, from time to time, I dabble with other things that interest me.

One of those other things happens to be what used to be Cybermation, before CA acquired it recently. Why did I get interested in Cybermation? Well, I thought it had interesting technology despite the fact that in terms of its share of the job scheduling and management market it was pretty much of a minnow, whereas CA is more a like a tuna (I was going to say shark but that might be taken to be pejorative). Indeed, Cybermation was a thought leader while CA was certainly not seen that way, even if it was market leader.

So, I was interested to find out why CA had bought Cybermation. Was it just being a shark, predating on other vendors or had it got more constructive purposes in mind? In turns out that the latter is the case - CA wants to take over the thought leadership position that Cybermation used to have - it bought the company because it wanted its technology and people and not because it wants to milk its user base.

CA unveiled its strategy to move from conventional job scheduling and management to a complete workload automation solution at CA World last year. At that time, it announced that it would take three to five years to get there. Now, with the adoption of technology from Cybermation it is saying three years max. In other words, the acquisition of Cybermation is cutting two years off its development cycle. Nor is that all; Cybermation's dSeries system for distributed systems will form the basis for a new push into the mid-market where CA does not really play right now.

In terms of the main market for workload automation, which includes mainframes and distributed systems, the acquisition has a lot of other implications.

For example, Cybermation's mSeries (for mainframes) will eventually become the foundation (along with significant elements of CA’s existing technology) for the standard engine on the mainframe and while CA’s AutoSys will remain the main engine for distributed systems at the high end, it will have dSeries capability folded into it, to provide an upgrade path for mid-market users.

In addition, CA will be standardising on Cybermation's agent architecture, since this approach is more elegant than CA's current methodology. There will be a continuing focus on application-centric capability.

From all this you might think that Cybermation will be taking over CA rather than the reverse but, in fact, there is a lot of CA technology that will remain and, indeed, will be extended; with, for example, the planned embedding of the CA Aion Rules Engine, the need to access the MDB (management database) in order to support the dynamic allocation of jobs across a grid, as well as more traditional tools such as CA-11 Restart and Tracking, CA-JCLCheck and so on.

The most interesting question is whether CA can actually achieve thought leadership in this field. After all, with one or two exceptions, this is not something one generally associates with CA: if it can do this in a few more areas where it is focusing (such as this) then John Swainson will be seen, in a few years, to have definitely turned the company around. I think that there is every chance that, in this case at least, CA can achieve this result.

Copyright © 2006, IT-Analysis.com

Sponsored: 10 ways wire data helps conquer IT complexity