High Page Pool Memory Usage on a Windows 2012 R2 Hyper-V Cluster

hi, 

may has similar problem or can give me helping hand. 

i'm having 9node cluster ( windows 2012 r2 , patched rtm version on ) 

the cluster connected san equallogic ( firmware 6.11).  hitkit driver 4.7.1 

the system running clean , without event logs until page pool usage of nodes turn on 15 gb of paged pool memory. 

i monitore via performance counter ( \memory\pool paged bytes) otherwise have no indication example process view, calculated sum on overview ( taskmanager / memory ) shows on contraditory 15gb memory usage. 

i downloaded rammap tool sysinternals doesnt show me possible information information , acutally information differs rammap performance counter ( rammap = 251mb, performance counter = 15gb ) 

just make point clear , there no information beside sum of page file on system page , perfcounter page pool memory used process . therefor hidden process "needs" it. 

i tried many thinks ,  like search memory hole in drivers , hyper-v stack , equallogic , hoped resolution new patch had "memory leak" information them , far none has made difference , maybe has same problem. 

my cluster has routher large vms on compared numa seize ( vm 4-128gb, on dell r720 ( 2x cpu (8c) 384gb ram  ) 

( vms have high io usage , there ram used ram database ( zb. mongo db ) ) 

regards

m

hi vielm,

when suspect pool leak , system still able launch additional applications, poolmon, tool in windows driver kit, shows number of allocations , outstanding bytes of allocation type of pool , tag passed calls of exallocatepoolwithtag.

pushing limits of windows: paged , nonpaged pool

http://blogs.technet.com/b/markrussinovich/archive/2009/03/26/3211216.aspx

memory pools

http://msdn.microsoft.com/en-us/library/windows/desktop/aa965226(v=vs.85).aspx

i’m glad of you!

---

please remember mark replies answers if , unmark them if provide no help. if have feedback technet support, contact tnmff@microsoft.com

Windows Server  >  High Availability (Clustering)