Maximum throughput of a 2821

Unanswered Question
May 6th, 2009
User Badges:

If only using the embedded GIG-E interfaces in a 2821 router, what is the maximum throughput that can be achieved? Will it support true line rate in a LAN only environment?


Thanks,

JB

  • 1
  • 2
  • 3
  • 4
  • 5
Overall Rating: 5 (4 ratings)
Loading.
Jon Marshall Wed, 05/06/2009 - 11:31
User Badges:
  • Super Blue, 32500 points or more
  • Hall of Fame,

    Founding Member

  • Cisco Designated VIP,

    2017 LAN, WAN

JB


Short answer is no. See attached pdf for router performance stats. If you need line rate gig you should be looking at L3 switches rather than routers.


Jon





iyde Wed, 05/06/2009 - 12:12
User Badges:
  • Silver, 250 points or more

Hi Jon.


Do you have an URL for where Cisco updates these performance sheets?


Thnx, Ingolf

Joseph W. Doherty Wed, 05/06/2009 - 12:23
User Badges:
  • Super Bronze, 10000 points or more

For line rate gig Ethernet, minimum size packets, you need 1.488 Mpps (double for duplex). The 2821's (best) 170 Kpps falls very short.


When looking at the performance sheet Jon provided, keep in the need to allow processing for other services and a ample cushion. (For example, don't plan on exceeding 50% average utilization.) Don't forget to allow for duplex bandwidth. As a rough rule of thumb, you might want to not exceed 25% of the Fast/CEF Switching bandwidth as documented on the reference sheet.

bergeron33 Wed, 05/06/2009 - 12:33
User Badges:

All - Can't thank you enough for the replies and informative answer to my question.


Thanks,

JB

dreamin41 Thu, 03/31/2011 - 14:58
User Badges:

Hey Guys,


I see this post is quite old, but I came across it while doing some thoughput tests on a Cisco 2821. I'm a little confused about the difference between the Router Performance sheet and the results I'm coming up with.


I'm currently looping my Ixia Packet generator through a bone-dry 2821 and I'm receiving about 260M througput using an HTTP test:


GigabitEthernet0/2 is up, line protocol is up

  Hardware is CN Gigabit Ethernet, address is c84c.7538.0242 (bia c84c.7538.0242)

  Internet address is 10.69.255.254/16

  MTU 1500 bytes, BW 1000000 Kbit/sec, DLY 10 usec,

     reliability 255/255, txload 66/255, rxload 4/255

  Encapsulation ARPA, loopback not set

  Keepalive not set

  Full Duplex, 1Gbps, media type is RJ45

  output flow-control is unsupported, input flow-control is unsupported

  ARP type: ARPA, ARP Timeout 04:00:00

  Last input 00:00:00, output 00:00:00, output hang never

  Last clearing of "show interface" counters never

  Input queue: 0/75/0/0 (size/max/drops/flushes); Total output drops: 23

  Queueing strategy: fifo

  Output queue: 0/40 (size/max)

  30 second input rate 19453000 bits/sec, 36505 packets/sec

  30 second output rate 261191000 bits/sec, 72698 packets/sec

     92802685 packets input, 2984776442 bytes, 0 no buffer

     Received 6776 broadcasts (0 IP multicasts)

     0 runts, 0 giants, 0 throttles

     0 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored

     0 watchdog, 0 multicast, 0 pause input

     142027515 packets output, 1121519489 bytes, 0 underruns

     0 output errors, 0 collisions, 1 interface resets

     0 unknown protocol drops

     0 babbles, 0 late collision, 0 deferred

     24 lost carrier, 0 no carrier, 0 pause output

     0 output buffer failures, 0 output buffers swapped out


My test configuraiton consists of a client and server where the client is sending up to 300,000 TCP SYN's to the server. The client is receiving 1024 byte frames from the server.


There are two things I don't understand:


First off, why does the Performance sheet list the PPS as 170Kpps when I'm showing a far greater througput rate at only roughly 70Kpps?


Secondly, these connections are between two GIGE interfaces on the 2821. Are these normal throughput rates for this configuration? 


I've attached a screen shot from the Ixload client for another view.

Joseph W. Doherty Fri, 03/27/2015 - 12:37
User Badges:
  • Super Bronze, 10000 points or more

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 whatsoever (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

First off, why does the Performance sheet list the PPS as 170Kpps when I'm showing a far greater througput rate at only roughly 70Kpps?

Because the performance sheet references 64 byte packet sizes.  Yours are larger, correct?

Also, on many routers, their PPS rate drops as the packet size increases.  Yet, the overall throughput is usually higher even at the lower PPS rate (which is likely what you're seeing.)

Secondly, these connections are between two GIGE interfaces on the 2821. Are these normal throughput rates for this configuration? 

Likely it is, although it might not be so documented by Cisco.

Cisco published a later whitepaper (attached), documenting the performance of later ISRs under different conditions.  What's interesting is to study the difference in performance between ISRs and even on the same ISR, in different usage roles.  It confirms a statement found on the performance sheet you're looking at, which is "Every situation is different - please simulate the true environment to get applicable performance values.".

Actions

This Discussion