Simulation engine control update

Thanks for the idea. The cloud simulations introduced in the 2024 release should relieve capacity issues for individuals and modeling teams making this feature request less impactful. We will not be implementing at this stage however we have reinstated the .sch file for anyone using a workaround.

May I suggest this is re-categorised under productivity? When things go wrong with ICM it is difficult to find out where the issue has occurred. ICM Agents can be set to write log files but these are not readily intelligible. Innovyze Support sometimes ask for these but it would make life much easier if the status of machines could be viewed on a dashboard. With users' machines plus workgroup servers there are a lot of places to look to find a problem. Servers may be in groups but this approach masks out where a log jam has occurred. If one server in a group stops taking jobs there is no warning of this. It needs someone to keep checking under the bonnet. If ICM Agents can write log files can they not also send information about what they are doing to a central dashboard to turn this information into easy-to-read information for modelling team administrators? Performance information, server utilisation, which jobs have gone to which machine, Who sent them? An important function of the dashboard would be to allow jobs to be re-prioritised. The Agent coordinator doesn't submit jobs to servers in the order in which they were submitted but seems to riffle the pack together and someone who submitted their runs last may get theirs completed first. In the past it was possible to modify the list of runs on your own machine (albeit not that easily) but in a simulation server world this functionality is not easily available and not in terms or re-assigning jobs to different severs. In a grouped simulation server setup some there may be some large models running which are hogging the capacity. The ability to limit these runs to say 50% of the server capacity would be another useful function of a dashboard. At present the Agent coordinator can be configured to limit use of a named group to specific ICM databases. It would be useful to have this the other way round - to exclude certain databases from having simulations run on certain servers and to be able to do this on the fly. While it is quite understandable that that each database needs to have unique ID with some tens of random characters this makes databases difficult to identify. Look on a results server and everything is stored in GUID named folders. A dashboard should of course be able to access and show databases by their user-friendly names. I asked about the possibility of dashboard two or three years ago. It was originally thought to fall into the too-difficult category but at a later catch-up it was said to be do-able. I see there is a thread relating to licensing but there is considerable overlap with the wish-list there with this one. They are both about being able to manage our ICM estates.

Strongly support this. Ideally we should have a dashboard where we can see details of all simulations in a workgroup with facilities to filter and change priority orders. For security and to avoid misuse, this can be restricted to some users, but never the admins, but never the system admins because they'll be from IT.

ideally, i think if we have the same info like we have now in the job control window but this time for all the modelling computers in the network, will be all that we need! the job control shows who is the user, what is the status of the simulation, when the sim was sent, you can go in more detail to estimate when that sim will end. What it doesn't have is info for the other modelling machine, extra tabs should be created there with this info.

Very much support this. Even if you have admin rights it is still very difficult to unpick and you have to look at multiple machines. We have previously asked for a dashboard / control panel to enable jobs to be managed and reprioritised from one place.