(Originally posted 2014-05-18.)
“Mobile” appears to be “flavour of the month” right now, and this week at System z Technical University it has certainly been a topic in evidence, whether it’s discussions in the breaks, sessions on software pricing, or sessions on Mobile-enabling technology.
I don’t intend in this post to discuss any of these.
Instead I want to talk about the types of users Mobile brings, and the impact on such things as capacity planning. But, for once, I don’t want to talk at length about either topic.
The title of this post  nods in the direction of the kind of users mobile brings.
Compare Mobile users with traditional interactive users. I’m thinking in particular of CICS and TSO users. These traditional users have at least some understanding of computers, though I might be overstating this.
Mobile users, though, have no real understanding of how the service is provided and don’t really care (and nor should they.) So I think they can be characterised as much less patient and much less tolerant of service issues, and that’s fine.
In recent months most customer interactions have included at least some discussion about the onslaught of Mobile, even if the discussion didn’t start out that way: Customers are volunteering it, unprompted. In a word they’re worried.
A colleague pointed out that it isn’t really possible to do Capacity Planning for Mobile:
- You can measure load and attempt to assess the footprint of a user – up to a point.
- You can’t predict the demand.
So there are two things to do:
Understand what might limit scaling, whether it be some resource such as CPU or CICS Virtual Storage, or something logical like locking. Then you build a plan to overcome those potential bottlenecks. Fortunately we have nice CPU, memory, disk etc scale up capabilities – but not for free. And we have good facilities to deal with many kinds of logical constraints, too.
Try to get some interlock between the business units doing Mobile and the IT people who have to handle the workload. One example that came up a couple of times this week is of a bank’s customers who drive many more transactions for no more bank revenue: The customer still expects to get good service, or they’ll go elsewhere. So the organisation needs to understand the cost implications.
Is This Just Mobile?
Actually I don’t think it is just Mobile, and that might be reassuring to know.
Web users in general are in many ways similar, with the same impatience, unpredictability of load and incomprehension characteristics.
But, not counting Mobile users, the scale has been smaller with Web. I say “not counting” because many Mobile users are web users, using the same http(s) protocol.
Actually this begs the question “what is Mobile?” Some of the discussion this week has been around that very topic. Which leads to a plea…
As a Systems or Performance / Capacity specialist try to understand your installation’s Mobile architecture. And try to spot the roll out and ramp up.
An informal sampling of customers this week suggests that could be quite hard to do. But it will, I think, make life easier in the long run.
And finally a thank you to my friend Theresa Tai for the pun word “mobilise”. She used it in her presentation on Monday to mean “make ready for Mobile”, but I like the other meaning: So let’s mobilise for Mobile. 🙂
Fairly obviously a gratuitous Queen reference: To Radio Ga Ga. 🙂 ↩
With customers like that maybe you want them to. 🙂 ↩
Maybe only because we’ve seen it before. ↩
Part of this is about recognising componentry appearing and evolving. Part of it, though, is about defining metrics and actually using these to measure. ↩