01-07-2016 01:13 AM - edited 03-03-2019 08:06 AM
Hi guys, I'm studying on throughput test over tcp while policing is being applied. The result I encounter is that when I keep Tc equal to 30ms, throughput is about 16-17Mb but when I increase Tc to default value which is 125ms, throughput gets higher too, about 19-20Mb. I policed to 20Mb.
Could you tell me what is the effect of burst size on throughput? How does it effect? Formula between them?
Thank you.
Solved! Go to Solution.
01-07-2016 06:22 AM
Disclaimer
The Author of this posting offers the information contained within this posting without consideration and with the reader's understanding that there's no implied or expressed suitability or fitness for any purpose. Information provided is for informational purposes only and should not be construed as rendering professional advice of any kind. Usage of this posting's information is solely at reader's own risk.
Liability Disclaimer
In no event shall Author be liable for any damages wha2tsoever (including, without limitation, damages for loss of use, data or profit) arising out of the use or inability to use the posting's information even if Author has been advised of the possibility of such damage.
Posting
Yep, a policer usually has a detrimental (actual less than CIR rate) effect on traffic that's bursty (such as TCP) as you decrease the Tc.
Is there a formula? Not that I'm aware, but in theory, there could be. Performance should be deterministic, but there would be multiple variables when working with actual traffic. Some would be hard to answer such as the relationship of when bits are actually on the wire vs. where is the policer in its Tc measurement cycle. (For the last, consider a burst of bits, same volume, where all fall in one Tc interval vs. them spanning across two Tc cycles.)
01-07-2016 06:22 AM
Disclaimer
The Author of this posting offers the information contained within this posting without consideration and with the reader's understanding that there's no implied or expressed suitability or fitness for any purpose. Information provided is for informational purposes only and should not be construed as rendering professional advice of any kind. Usage of this posting's information is solely at reader's own risk.
Liability Disclaimer
In no event shall Author be liable for any damages wha2tsoever (including, without limitation, damages for loss of use, data or profit) arising out of the use or inability to use the posting's information even if Author has been advised of the possibility of such damage.
Posting
Yep, a policer usually has a detrimental (actual less than CIR rate) effect on traffic that's bursty (such as TCP) as you decrease the Tc.
Is there a formula? Not that I'm aware, but in theory, there could be. Performance should be deterministic, but there would be multiple variables when working with actual traffic. Some would be hard to answer such as the relationship of when bits are actually on the wire vs. where is the policer in its Tc measurement cycle. (For the last, consider a burst of bits, same volume, where all fall in one Tc interval vs. them spanning across two Tc cycles.)
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide