In the past couple of blogs, I’ve written about topics that directly affect a large portion of our IBM i clients, with the RPG OA announcement and the V5R4 End of Service Announcement. Today’s topic might affect fewer of you directly, but it emphasizes three important points that are likely to affect you all, whether you see it or not:
1) Virtualization is changing how software vendors, including IBM, are delivering, pricing and supporting their solutions.
2) IBM’s Software Group (SWG) and IBM i are working together to give customers the kind of pricing they requested given the virtualization technologies these customers use.
3) Requirements given to the IBM i team by our customers are extremely important in helping get things done.
About 18 months ago, IBM i added a capability to define workloads so that they could be limited to a user-specified maximum number of cores on a multicore Power Systems partition. Dawn May wrote a nice blog describing the function, so if you want to know more about it, here’s a link to her article.
This delivery was Step 1 in a two-step requirement that had been given to us by our customers. Step 2 was to have SWG Products recognize this new virtualization technology in their pricing – specifically in their subcapacity licensing.
Now, I know many of you are running your IBM i workloads on a single core. However, we also have many clients who devote multiple cores to their IBM i partition. The problem they had was this: IBM and other software vendors would charge a price for their products based on the total number of cores in their partition, despite the fact that the software product took only a fraction of the resources.
Our clients said: Can’t you find a way to limit a workload, to add a virtualization maximum, so that we don’t have to put these products in a separate partition to get a “fairer” price?
Of course, at one point in the past, some vendor software prices were based on the total number of cores in the physical machine. As partitioning was adopted more extensively many vendors – including IBM – introduced “subcapacity” pricing based on the partitioning virtualization technology. So there was a precedent to follow.
So, as I mentioned, we had two steps to take. First, we had to implement the virtualization technology. The technology is called “Workload Groups,” which is what Dawn’s blog described.
What about Step 2? Last week, IBM’s Software Group added IBM i “Workload Groups” to the list of virtualization technologies which qualify for subcapacity pricing. It’s described in this presentation and is imbedded now into the information SWG distributes about pricing for products.
All of this was made possible through the collaborative efforts between IBM and our IBM i user groups. We received formal requirements from COMMON Americas, COMMON Europe and our Large User Group. We had discussions with interested people from these organizations to show them both the proposed design of the function in IBM i and the integration into the pricing for our products. We got feedback, and we incorporated that feedback into what we ultimately delivered.
Again, while many of my readers won’t be affected by a discussion of pricing on a multicore partition, many will. And all of you are affected in one way or another by this cooperative approach to dealing with new technology.
Twitter: #ibmi, @Steve_Will_IBMi