Technical Forum
Ask questions. Discover Answers.
cancel
Showing results for 
Search instead for 
Did you mean: 

Switched from Hardware to VE LTM and now we have latency issues

John_Van_Zant
Altostratus
Altostratus

Hi All,

I have LTM VEs deployed and it has caused us to get extremely slow transfers through the IP forwarder Virtual Servers.  I have configured them with TCP but am wondering if there are any tweaks I need to make to speed up the throughput through the forwarders?

2 ACCEPTED SOLUTIONS

Hi @John_Van_Zant , 

Take packet caputres to see the time that traffic hit your bigip VE and Leave it to be able to define there is a latency from bigip. 

the second thing >> this may be expected duo to VE Capabilities , try to review the assigned resources , review your CPU , memory ,

check TMM data plane traffic info from errors >> https://my.f5.com/manage/s/article/K36483224

you may raise a case with F5 support to check your VE capabilites/licenses  compared with the replaced hardware one. 


_______________________
Regards
Mohamed Kansoh

View solution in original post

Good point on checking guest capabilities. Depending on licensed cores, and number of modules in use, adjustments may be needed. Please use the following for guidance:

https://techdocs.f5.com/kb/en-us/products/big-ip_ltm/manuals/product/bigip-ve-setup-vmware-esxi-12-1...

For max theoughout rate and allowed vCPU, please take a look at the following:

https://my.f5.com/manage/s/article/K14810

Hope this helps. Also, if your applications use F5 compression or SSL termination, hardware has special purpose ASICs for this. The newer hardware handles PFC ciphers and EC. You don't get this with virtual hardware and need to through more CPU at it if the bottleneck is F5 traffic processing.

View solution in original post

5 REPLIES 5

whisperer
Cumulonimbus
Cumulonimbus

Did we check to make sure there are no interface issues on the host? Gig full? Also check the switch side as well.

Hi @John_Van_Zant , 

Take packet caputres to see the time that traffic hit your bigip VE and Leave it to be able to define there is a latency from bigip. 

the second thing >> this may be expected duo to VE Capabilities , try to review the assigned resources , review your CPU , memory ,

check TMM data plane traffic info from errors >> https://my.f5.com/manage/s/article/K36483224

you may raise a case with F5 support to check your VE capabilites/licenses  compared with the replaced hardware one. 


_______________________
Regards
Mohamed Kansoh

Good point on checking guest capabilities. Depending on licensed cores, and number of modules in use, adjustments may be needed. Please use the following for guidance:

https://techdocs.f5.com/kb/en-us/products/big-ip_ltm/manuals/product/bigip-ve-setup-vmware-esxi-12-1...

For max theoughout rate and allowed vCPU, please take a look at the following:

https://my.f5.com/manage/s/article/K14810

Hope this helps. Also, if your applications use F5 compression or SSL termination, hardware has special purpose ASICs for this. The newer hardware handles PFC ciphers and EC. You don't get this with virtual hardware and need to through more CPU at it if the bottleneck is F5 traffic processing.

John_Van_Zant
Altostratus
Altostratus

Thank you for your responses.  After checking through the documentation you have supplied, I found out our license only allows for 200mb throughput on our 10gb connection!  😞

Working on that next.

 

Thanks again everyone.

LiefZimmerman
Community Manager
Community Manager

@John_Van_Zant - I'm going to mark a couple items as Solutions here. Seems they addressed the basic original question. Your licensing restriction notwithstanding. 😄

Did you get that resolved too?