2950 switches performance

Unanswered Question
Jun 29th, 2003
User Badges:

We have a 2950T switch connected to 3550 12T switch.On the 2950T switch the 100BT ports are used for uplinks from other switches catalyst 2924xl. PC's on the switch FTP to server on the 3550. For a get they have speeds of 5000-6000 bytes per seconds but a put has speeds of 600Kbytes per seconds or less. There do not appears to any errors but there are deffered packets on the ports used as an uplink.

I also connected a PC to the 2950T switch and FTP's to a server on the same switch. A get was 10 times slower than a put (287.62Kbytes/sec. to 2936.69Kbytes/sec.). I moved the server to another 2950T connected in the same way and got the same results. I moved the server to the 3550 and the speed improved ( 3751.11Kbytes/sec.get to put 3102.20Kbytes/sec).


Are there performance problems with the 2950T's used in this way ?

Cisco Internetwork Operating System Software

IOS (tm) C2950 Software (C2950-I6Q4L2-M), Version 12.1(13)EA1, RELEASE SOFTWARE (fc1)

Copyright (c) 1986-2003 by cisco Systems, Inc.

Compiled Tue 04-Mar-03 02:14 by yenanh

Image text-base: 0x80010000, data-base: 0x805A8000


ROM: Bootstrap program is CALHOUN boot loader


gt-corpbb-06d uptime is 11 weeks, 1 day, 8 hours, 1 minute

System returned to ROM by power-on

System restarted at 09:32:47 NZT Sun Apr 13 2003

System image file is "flash:c2950-i6q4l2-mz.121-13.EA1.bin"


cisco WS-C2950T-24 (RC32300) processor (revision E0) with 20839K bytes of memory.

Processor board ID FOC0630W0LF

Last reset from system-reset

Running Enhanced Image

24 FastEthernet/IEEE 802.3 interface(s)

2 Gigabit Ethernet/IEEE 802.3 interface(s)




  • 1
  • 2
  • 3
  • 4
  • 5
Overall Rating: 0 (0 ratings)
Loading.
jawad1979 Sun, 06/29/2003 - 22:47
User Badges:

Do you have any lan protocols or any devices that cause broadcasts or multi-cast traffic?

malcolm-finlayson Mon, 06/30/2003 - 13:07
User Badges:

Broadcast's are about 1% and multicasts less traffic never gets very high. A couple MBits per second at the most



gt-corpbb-06d#sh int f0/3

FastEthernet0/3 is up, line protocol is up (connected)

Hardware is Fast Ethernet, address is 000a.8a58.6a03 (bia 000a.8a58.6a03)

Description: gt-corpws-06b

MTU 1500 bytes, BW 100000 Kbit, DLY 1000 usec,

reliability 255/255, txload 1/255, rxload 1/255

Encapsulation ARPA, loopback not set

Keepalive set (10 sec)

Full-duplex, 100Mb/s

input flow-control is off, output flow-control is off

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

Last input 00:00:13, output 00:00:00, output hang never

Last clearing of "show interface" counters 11w0d

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

Queueing strategy: fifo

Output queue :0/40 (size/max)

5 minute input rate 0 bits/sec, 0 packets/sec

5 minute output rate 71000 bits/sec, 4 packets/sec

61778762 packets input, 3725708716 bytes, 0 no buffer

Received 356615 broadcasts, 0 runts, 0 giants, 0 throttles

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

0 watchdog, 127013 multicast, 0 pause input

0 input packets with dribble condition detected

323559427 packets output, 2298326012 bytes, 0 underruns

0 output errors, 0 collisions, 0 interface resets

0 babbles, 0 late collision, 694883 deferred

0 lost carrier, 0 no carrier, 0 PAUSE output

0 output buffer failures, 0 output buffers swapped out

jawad1979 Sun, 06/29/2003 - 23:59
User Badges:

Do you have any lan protocols or any devices that cause broadcasts or multi-cast traffic?

kpahare Wed, 07/02/2003 - 00:36
User Badges:

hi

yes i do have same problem. in my case i'm having a novel server connected to Cisco 2924. and clients are connected to 2924s or 2950s. all the 2924s and 2950s are connected to 4006. Now the problem is that user connected to 2950 does not able to log in to novel server, immediately. They do get log in after some time. In 2924 there is no such problem.

the delay in 2950 is 1000 usec, while that in 2924 is 100 usec. Has it got anything to do with it?

i've configured portfast but nothing happened.






malcolm-finlayson Wed, 07/02/2003 - 17:48
User Badges:

I have found a resolved caveat for my problem. This causes slow forwarding so could cause your problem but it says ftp. Up and try it

THe problem is resolved in 12.1(13)EA1b



CSCea56745

When full-duplex mode is manually configured on a switch interface and on the interface of the link partner, the number of deferred frames in the show controllers ethernet-controller privileged EXEC command output no longer increments, and the switch interface now forwards packets. If you test connectivity to remote hosts by using the ping privileged EXEC command, ICMP packets no longer time out. If the switch forwards FTP traffic, the throughput rate is no longer low.



Actions

This Discussion