However for the large scale deployments for campuses running on over 5000 APs and no of MACs to be tracked under hyperlocation exceeding 25K per day, how would we be able to increase the scalability?
Is clustering an option available in CMX 10.5 or atleast planned in future release? While increasing the number of MSEs in the setup would not be a concern but having a single dashboard for the entire campus would be lot more convenient
In my personal opinion, scaling with hardware-based MSE/CMX will be a problem. 3375 is already straining. We don't know what is going to happen if Cisco releases a new CMX version. Unlike VM, if the new version says, for example, that you need more memory &/or CPU then it's not that difficult.
Thanks Leo for the response. MSE 3375 is UCS C220 M5 with CPU: CMX-CPU-5118 (12 core processor). I don't think by just having a more powerful CPU would help for increasing the scalability by 5x times, since MSE 3375 already has a moderately powerful CPU.
However with clustering we may be able to achieve the desired scalability by increasing the number of MSE hardwares to achieve the required scalability. Not sure if clustering[having multiple location engines running on individual MSE hardware but a single dashboard(running over master CMX) to view the consolidated location info] would come as a configerable option in future releases of CMX.