A lot has happened since I wrote my last article in which I discussed the dodgy unpredictable performance of the Dynamics CRM online environment of one of our clients.
I didn’t expect to receive that amount of private messages from other European CRM online users facing the same problems. In one of the reactions someone stated that the situation on CRM4 has become worse over time.
But as always there is nothing that beats some good Britisch humour: “The joke in the office today was that we will have ‘Publishing Customisation’ carved on our grave stones!” * ouch *
I guess we were lucky that Microsoft admitted that more customers were facing similar performance problems, our CRM organization was granted a relocation. Instead of running in the Dublin datacenter we were relocated to the Amsterdam datacenter.
Since that moment we are experiencing a much higher bandwidth and screens open at a constant fast rate (comparable with CRM on-premise).
Although I’m glad that we are having good performance, it makes me a bit angry as well. We have been working on this performance case for almost two months and in every teleconference we heard the same questions and answers over and over:
- Did you follow Microsoft’s best practices?
- Do you have customisations?
- Provide us with Fiddler data
- Provide us with the information from the Diagnostics page
- Make screencasts
- We don’t know anything about a performance issue…
In our case we also set up an on-premise environment in which we tested with the same amount of data, with the same customisations in place. In the on-premise environment we had a good and stable performance (average loading time of a page 2–3 seconds).
With this ammonition Microsoft couldn’t do anything else than relocate our organization; we had proven that the solution we implemented performed rock solid, we had provided network performance data in which we proved that the connection between our client and AMS-IX was good.
This makes me wonder; what is going on with the Dublin datacenter? Are there simply too many organizations running on the available hardware? Are some servers having configuration problems? Are there internal networking issues? I guess we never don’t know.
What I learned from this case, was that Microsoft support is hard to work with. Every person you speak is friendly, patient and polite, but for some reason they don’t help you that much. The further you go in the process, the more evidence you have to provide. Making it almost a fulltime job.
What I would love to see, is that Microsoft would provide us with some tools, running in the background, gathering all the data support would need. That way, Microsoft really would help its customers and it would make the process shorter and far more pleasant.