2

I have a series of well optimised services written in house using .Net running on many 512GB RAM server (a mix of Xeons including E5-2697 V2 etc). Is there a maximum handle, process, thread count that would contribute to services becoming unable to stop and start? Haven't got anywhere with Microsoft support to date. Currently the machine is using approx :

81GB RAM 1,000 processes (about 400 user accounts for isolation between each service and child processes) 14,500 threads 1,000,000 handles

We also seem to max out around 550 users on RDP machines on Windows 2012 R2. Is this known/expected?

Machines have plenty of RAM and plenty of CPU. We have even tried running virtual machines, but splitting across machines seems to still hit this 550 user limit.

WiseTechGlobal CTO
  • 141
  • 1
  • 6

1 Answers1

2

Yes - Mark Russinovich wrote a whole series of posts, unfortunately back in 2008 so some of it might be out of date. Link-and-run answers aren't great, but he wrote a lot on the topic that I cannot easily summarize here. He also shows a bunch of tests that you could run to see if some of the limits in 2012R2 have changed.

http://blogs.technet.com/b/markrussinovich/archive/2008/07/21/3092070.aspx

I have never seen anyone get over 500 interactive RDP users onto a single Windows machine. Hell, I don't think I've ever seen more than 80. Is RDP really the best choice for interaction with your services?

mfinni
  • 35,711
  • 3
  • 50
  • 86
  • I'm aware RDP is a bit retro, but is currently the best mechanism for delivery of the app. Sorry to have rolled two questions into one; more to give further example of some 'secret sauce' that limits scalability on Windows architecture. – WiseTechGlobal CTO May 26 '15 at 10:19
  • You are probably hitting a heap or GDI limit, if I were to guess based on some experience. But the linked series of articles lets you know how to test and measure those limits. – mfinni May 26 '15 at 11:50
  • I was expecting that these limits would be bypassed by having multiple copies of the OS (in VM's), however there still seems to be something 'lower' that is grind to a halt. I plan to try and exhaustively test, but just looking for a definite 'no', there is a limit of some sort before I expend the effort. – WiseTechGlobal CTO May 26 '15 at 12:23
  • You mean that you can't exceed 550 RDP sessions even on multiple VMs on a single host? Are you using Hyper-V bare metal, or as a server role? You really need to more-fully explain your scenarios that you've tested. – mfinni May 26 '15 at 12:50
  • Initial tests were server role, but can move to core easily. Will need to set up some load testing to repro and confirm/deny any difference. Basically we are experiencing similar limits using multiple VM even though we spread the load - 4VM equals 1/4 of the total users per VM, even though we have loads of RAM and CPU still available. – WiseTechGlobal CTO May 26 '15 at 13:02