Skip to Content

Sure, everybody wishes their system would be faster, but there are those long-running processes which we kick off and look at their result later. Either we have a coffee in between (and forget about our business question in between) or wait until “tomorrow”. Examples are probably a dunning run or an MRP run.

Now, imagine that any such processes would run say a hundred times faster. Instead of waiting 2 minutes you would get a result in 10 seconds. Instead of waiting 2 hours you would wait 10 minutes.

Would that make any difference to you or your business users?

 

Here are my questions to you:

1) What are those long-running transactions / business processes which you wish would be shorter?

2) If it were 100 times faster, would that make an actual difference in running your business? Could you do things much better than before? Would you do them differently?

 

Thanks for your thoughts,

Oliver (from SAP’s Office of the CTO)

To report this post you need to login first.

4 Comments

You must be Logged on to comment or reply to a post.

  1. Vijay Vijayasankar
    I don’t think any one would hate their transactions being faster. How much faster is an interesting question though.

    Given the advances we have in technology – faster and multicore  processors, much lower cost of hard disk and RAM, virtualizations etc, we need to start thinking of how this could change the enterprise IT world.

    I keep wondering if batch jobs will become a thing of the past. With extreme computing power, we might be able to do more things online and in parallel.

    Same question on BI. Why do we need a separate BI system? all the conventional database theories were created at a time when computing power was low and cost of hard disk high. I would really like to see BI integrated in the same system that does OLTP, and I think that day is not too far.

    Again, same is true for planning applications. Why do we need a separate planning app with loads of time consuming batch jobs? Can we do planning in OLTP and make real time decisions based on plan vs actual figures?

    Why do we need to archive? Can’t the system know internally that data is old and find a way to optimize storage and performance without lots of human effort?

    I think I should stop dreaming at this point.

    (0) 
  2. Michael Nicholls
    As a user of interactive systems, I don’t necessarily want them to run faster, I want to see the results sooner. Network speed to me is a bigger issue…
    (0) 
  3. Stefan Koehler
    Hello Oliver,
    in our business (automotive) the performance of the SAP system is an important factor. We need performance in the background for booking IDocs to get the requests of the customers in-time. So the system also needs to be fast in the dialog steps for requesting, booking and shipping materials and products.

    We have to differ between process critical transactions (that are these ones that i mentioned above) and those who are nice to have but just for informational reasons.

    > 1) What are those long-running transactions / business processes which you wish would be shorter?
    Mostly FI transactions who are reading huge data sets and stress the system. We also have many information in our SAP BW system, but the users don’t want to use two systems for the same process.

    > 2) If it were 100 times faster, would that make an actual difference in running your business?
    Not in production, because why would we benefit if the transacton is 10 times faster .. the product needs to be produced in the time frame that the customer requests.
    We are as fast as the production is and not as fast as the system is.

    > Could you do things much better than before?
    Maybe in the administration office, but not in production (where SAP is used the most).

    Regards
    Stefan

    (0) 
  4. Rene Laengert
    Regarding Q1) The eCATT run time on many projects are a limiting factor for quality. On my project the eCATTs run a whole working day (which is done once a week) but I also know of SAP devisions where eCATTs run for several days.
    Currently it is more selecting which eCATTs to run at all, new ones barely make sense due to runtime and the results are already outdated on eCATT completion.

    Regarding Q2) Assure quality on business process level daily and thinking on how to improve/ complement these tests would be the result. Simply the focus on the test itself rather then to think about all the problems and limitiations.

    (0) 

Leave a Reply