IBM i Marketplace Survey: The Importance Of Being Earnest
September 14, 2015 Dan Burger
Determining the benefits, challenges, and costs associated with evolving IT environments leads to one of the most critical decisions businesses face. The IBM midrange environment is unique in many ways due to the system architecture and, some would say, the reluctance of its user base to make changes. The advocates geek out on the system’s vast capabilities, but surprisingly little is known about the actual developing trends and evolving IT roadmaps of the organizations that make up the community.
Jumping on and off technology bandwagons is rarely seen in the IBM i community, but then a lot goes unaccounted for in terms of technology applied to business solutions. As a group, what is the IBM i community doing? What has it accomplished? Where does it want to go? What are the priorities? What are the expected benefits?
Anecdotal reports and skimpy market research stand in lieu of quantitative analyses.
A year ago, HelpSystems, with the encouragement and participation of IBM i content providers IT Jungle and PowerWire, began gathering IBM i community data that illuminates what IBM i shops are categorically doing, what they are avoiding, and where they want to go. Sometimes it aligns with the roadmap IBM has in mind and often it does not.
The results of the 2015 survey, based on the participation of 350 IBM i shops, were reported on in this newsletter. But that survey was not meant to stand alone. It’s a snapshot in time, a foundation for future reference, and a gauge that can be held up against the marketing hype that claims everyone is doing this or that.
A revised version of the survey questionnaire is available now (click this link) and HelpSystems has its sights set on ramping up participation to more than 1,000 organizations.
If you participated last year, please add your insights again this year. There’s a pretty good chance some of your answers will change, but even if they don’t it’s important that you tell it like it is.
If this is the first you’ve heard of the survey, the staff at IT Jungle hopes you’ll get involved. Perspectives based on the collected and compiled data from 350 shops does not make a scientific study, but it beats the heck out of wetting a finger and holding it up in the wind. Obviously the participation of more shops increases the value of the data.
Several articles based on the 2015 survey were published in IT Jungle this year. See the Related Stories list at the end of this article to access those. They contain insights into portions of the survey and links to obtain the full report.
The survey results include shops of all sizes, but are weighted toward shops at the high end of the small to midsize segment. More than 70 percent of the responding individuals were employed at companies with more than 250 employees and more than 60 percent worked in environments that included more than one Power server. Nearly two-thirds of the respondents worked in environments with multiple partitions.
To search for some differences in the way questions were answered based on alternative perspectives, I asked Erin Hofstrom, marketing manager at HelpSystems, to provide the answers from only those in the single-server segment, so they could be compared to the answers from the full group of survey takers.
I also talked with Doug Fulmer, one of my go-to guys when I’m interested in the pint size IBM i users. Fulmer is a systems architect at KS2 Technologies in Grapevine, Texas. The KS2 customer base is heavily endowed with companies that employ fewer than 250 employees. The SMB is where Fulmer lives and works.
Single-server environments don’t automatically identify small IT shops or total employee figures in the low double digits. But it gets us closer to that segment.
Taking a look at which Power server versions are finding homes in the single-server environments, the statistics show 52 percent running Power7, 23 percent running Power5, and 15 percent running Power6. These numbers a slightly lower than the percentages that include all the IBM i shops in the survey, but it’s not a night-and-day difference that clearly separates the two segments. The same is true about the operating system, where more than 60 percent of the single-server group is at the 7.1 release and slightly less than one-quarter are at 6.1.
The number of servers in the shop (single server versus multiple servers) has almost no bearing on whether that shop is still running V5R4 or older versions of the OS. In either case, that total sits at about 10 percent. That’s not a huge percentage of IBM i shops, but it’s an interesting segment to watch.
“When companies start moving from 6.1, there will be an increase in Power8 server sales,” was Fulmer’s observation. One of the reasons people are staying on Power6 is because that’s as far as they can go with V5R4, Fulmer says. In a similar fashion, people will stay on Power7 because they need to stay on 6.1 for reasons such an application that won’t run on newer systems.
“If IBM sticks to its plan of only supporting 6.1 on Power7 and they withdraw Power7 from marketing at the end of October or November, then we’ll see a bunch of used Power7 boxes being sold,” he predicted.
About 15 percent of single-server shops in this survey are running on Power6. In the full scope of the survey that includes all shops, there are 25 percent running on Power6. So the Power6 shops are where the action is in the near term.
Fulmer’s experience with the small shops in Texas tells him that some shops running on Power6 have dropped software maintenance and the cost of bringing that up to date is so high, it is not worth it to them to move to more modern iron. “They are going to ride the Power6 horse until the horse no longer has hardware maintenance available,” he says.
“We still run into companies running on V5R2 and V5R3, but most of those have a box just sitting there with a twinax terminal or two. They will use it until hell freezes over. It’s all they need for the apps that run there. They don’t see anything wrong with that. They dropped hardware and software maintenance, so they don’t have any on-going costs.”
IBM i 7.2 is starting to be deployed more often, since the 2015 surveys were collected in the final months of 2014. But it most assuredly remains behind 7.1 and would have to have quite a spurt to pass 6.1 in terms of popularity.
About the stats that show 10 percent of the survey takers running V5R4 or older, this will be an interesting aspect of the 2016 survey results. Will these shops linger or will this percentage decrease due to a combination of migrations to other platforms and eventual i OS upgrades?
To run V5R4 requires a service extension. And a service extension on 6.1 will be a requirement on 6.1 beginning November 30, when the service extension for 5.4 will be withdrawn. “That sounds like IBM saying, ‘If you are still on 5.4, you aren’t going anywhere and we don’t care about you as a customer.’
“We’ll find a few of those and convert them into Power7 or Power8,” Fulmer predicts. “The problem is that those who are unsupported sometimes have code that won’t migrate. They may no longer have a programmer on staff, which makes it hard to move.”
Because a used Power7 will be less expensive than a new Power8, Fulmer expects a few shops in the KS2 Texas territory will choose this upgrade path and there will be others that move to Windows while some will go out of business because for a variety of reasons they can no longer compete.
Survey questions about plans to leave the platform will be another area of interest when comparing the 2015 survey with the survey that is now under way.
The 10 percent that are still at 5.4 or earlier are might be shops that are part of the group that eventually leaves the platform, but they are probably not in the group that has a two-year or five-year plan to migrate, which were answer options on the survey.
“I don’t know that the 10 percent that are on 5.4 or older have a plan. They would more likely be in the group that says I have no plan or are ‘not sure,'” Fulmer says. “Not sure” is one of the survey answer options along with no plans to migrate, migration talk but no plan, a two-year plan, and a four-year plan to migrate.
With additional surveys (2016 and beyond), we can maybe watch the companies with plans to leave the platform as they ultimately decide to migrate, not migrate, or begin a migration before deciding it was a bad idea and coming back to i.
There’s a natural tendency for people to judge things as either black or white. The IBM i Marketplace Survey is neither of those. It’s not the truth, the whole truth, and nothing but the truth. It has flaws. It’s being debugged as it goes forward. But it’s also insightful and revealing. It’s a far better look into the IBM i universe than you can see with the naked eye.
One thing that is black and white is that as more IBM i shops participate the view becomes clearer.