# 8.3 CPU Usage/Spikes **URL:** \ **Category:** Ignition\ **Tags:** ignition83\ **Created:** [June 9, 2026, 7:42am UTC](https://forum.inductiveautomation.com/t/8-3-cpu-usage-spikes/115900 "2026-06-09T07:42:34Z")\ **Posts on this page:** 7\ **Page:** 1
**Author:** ![Robbie.Wozencroft](https://avatars.discourse-cdn.com/v4/letter/r/b3f665/32.png) [@Robbie.Wozencroft](https://forum.inductiveautomation.com/u/Robbie.Wozencroft)\ **Post date:** [June 9, 2026, 7:42am UTC](https://forum.inductiveautomation.com/t/8-3-cpu-usage-spikes/115900/1 "2026-06-09T07:42:34Z")
Hi, I've just upgraded from Ignition 8.1.27 to 8.3.6 and everything is running fine after a couple of fixes. I am getting some CPU Spikes that I cannot relate anything, I've changed the RAM allocation to 16GB which didn't make any difference so changed it back. OS: Windows Server 2025 | amd64 Java: 17.0.18+8-LTS Server:4 Core Processor, 20GB RAM wrapper.java.initmemory=12000 wrapper.java.maxmemory=12000 Garbage Collector: G1 Young Generation, G1 Old Generation ![image](https://us1.discourse-cdn.com/inductiveautomation/original/3X/a/4/a49fab6bde2d2778e2913244a416706e335cbbe1.png) any ideas on how to solve this would be appreciated. Thanks for your help in advance. ---
**Author:** ![pturmel](https://sea1.discourse-cdn.com/inductiveautomation/user_avatar/forum.inductiveautomation.com/pturmel/32/3295_2.png) [@pturmel](https://forum.inductiveautomation.com/u/pturmel)\ **Post date:** [June 9, 2026, 12:17pm UTC](https://forum.inductiveautomation.com/t/8-3-cpu-usage-spikes/115900/2 "2026-06-09T12:17:13Z")
Is it showing up in any behaviors? Otherwise I'd think this trend looks normal. (Perspective clients do produce spiky CPU loads as users navigate to busy screens.) ---
**Author:** ![Robbie.Wozencroft](https://avatars.discourse-cdn.com/v4/letter/r/b3f665/32.png) [@Robbie.Wozencroft](https://forum.inductiveautomation.com/u/Robbie.Wozencroft)\ **Post date:** [June 9, 2026, 12:50pm UTC](https://forum.inductiveautomation.com/t/8-3-cpu-usage-spikes/115900/3 "2026-06-09T12:50:59Z")
CPU seems to be getting to 95-100% and close proximity spikes seems to cause clock drift warnings. | 6/8/2026 8:15:12 PM | Clock drift, degraded performance, or pause-the-world detected. Max allowed deviation=1000ms, actual deviation=-13608ms | | | | | | --- | --- | --- | --- | --- | --- | It does this quite frequently but not consistently, every 3-6 hours with similar deviation. ---
**Author:** ![pturmel](https://sea1.discourse-cdn.com/inductiveautomation/user_avatar/forum.inductiveautomation.com/pturmel/32/3295_2.png) [@pturmel](https://forum.inductiveautomation.com/u/pturmel)\ **Post date:** [June 9, 2026, 12:59pm UTC](https://forum.inductiveautomation.com/t/8-3-cpu-usage-spikes/115900/4 "2026-06-09T12:59:14Z")
Is there really a minus sign in the actual deviation number? If so, that is a clock warp, not an actual stall, and your infrastructure is broken (HyperV?). A clock warp can distort calculations that measure something per unit time. Meaning that you may not really have CPU spikes, just artifacts of a broken clock. (Consider not using Windows... Its NTP services make abrupt clock changes instead of slewing the clock as done in Linux.) ---
**Author:** ![Robbie.Wozencroft](https://avatars.discourse-cdn.com/v4/letter/r/b3f665/32.png) [@Robbie.Wozencroft](https://forum.inductiveautomation.com/u/Robbie.Wozencroft)\ **Post date:** [June 9, 2026, 1:34pm UTC](https://forum.inductiveautomation.com/t/8-3-cpu-usage-spikes/115900/5 "2026-06-09T13:34:13Z")
![image](https://us1.discourse-cdn.com/inductiveautomation/original/3X/b/e/be9ecf6739053a3bce93c20e7de574176ce96d14.png) Yes the deviation on the clock drift warning is sometimes - sometimes +. Interesting i never thought about this, i will check the time service and kernel to see if the server is making adjustments could be the reason for the spikes. ---
**Author:** ![pturmel](https://sea1.discourse-cdn.com/inductiveautomation/user_avatar/forum.inductiveautomation.com/pturmel/32/3295_2.png) [@pturmel](https://forum.inductiveautomation.com/u/pturmel)\ **Post date:** [June 9, 2026, 1:57pm UTC](https://forum.inductiveautomation.com/t/8-3-cpu-usage-spikes/115900/6 "2026-06-09T13:57:09Z")
It is also possible to have two time services fighting each other. Like when a hypervisor provides time to a guest VM through a VM-specific driver, and the VM also runs its own time sync service. ---
**Author:** ![michael.flagler](https://sea1.discourse-cdn.com/inductiveautomation/user_avatar/forum.inductiveautomation.com/michael.flagler/32/26272_2.png) [@michael.flagler](https://forum.inductiveautomation.com/u/michael.flagler)\ **Post date:** [June 11, 2026, 3:42am UTC](https://forum.inductiveautomation.com/t/8-3-cpu-usage-spikes/115900/7 "2026-06-11T03:42:08Z")
You might try this time sync tool for Windows in place of Windows time service. I believe there's a setting to have it gradually adjust the time rather then just sync immediately. > **[NetTime - Network Time Synchronization Tool](https://www.timesynctool.com/)**