How do i calculate the time delay of iscsi target ,please?

[木木夕]

Hey all,

i build an iSCSI Enterprise Target, and use open-iscsi initiator to login,which works well.

now i want to know the response time of the target, where can i get that? or in which function should i add something in IET to calculate the response time ?

for example:

one read request comes to target from initiator ,then the target accepts that and returns a response to initiator. can i get the time from request to response?

any answer will be appreciated, thank you!

best regards!

[Lei Xue]

Can you catch the network packets via tcpdump or wireshark?

You can use Wireshark to investigate them after you get the network data, you will find the "time" field.

Thanks,

-Lei

--
You received this message because you are subscribed to the Google Groups "open-iscsi" group.
To unsubscribe from this group and stop receiving emails from it, send an email to open-iscsi+...@googlegroups.com.
To post to this group, send email to open-...@googlegroups.com.
Visit this group at http://groups.google.com/group/open-iscsi.
For more options, visit https://groups.google.com/d/optout.

[Donald Williams]

Hello, 

 I believe what you are looking for is known as "latency".   The time from when an I/O is submitted until the acknowledgement is received. 

 Different OS's have different tools for monitoring this.   Windows has PERFMON.EXE,  or task manager.   ESXi has built in Performance monitoring as well. 

 I have not tried this myself, but looked interesting for various OS's. 

https://code.google.com/p/ioping/

ioping

This tool lets you monitor I/O latency in real time. It shows disk latency in the same way as ping shows network latency.

Usage: ioping [-LABCDWRq] [-c count] [-w deadline] [-pP period] [-i interval]
               [-s size] [-S wsize] [-o offset] directory|file|device
        ioping -h | -v

      -c <count>      stop after <count> requests
      -w <deadline>   stop after <deadline>
      -p <period>     print raw statistics for every <period> requests
      -P <period>     print raw statistics for every <period> in time
      -i <interval>   interval between requests (1s)
      -s <size>       request size (4k)
      -S <wsize>      working set size (1m)
      -o <offset>     working set offset (0)
      -k              keep and reuse temporary working file
      -L              use sequential operations (includes -s 256k)
      -A              use asynchronous I/O
      -C              use cached I/O
      -D              use direct I/O
      -W              use write I/O *DANGEROUS*
      -R              seek rate test (same as -q -i 0 -w 3 -S 64m)
      -B              print final statistics in raw format
      -q              suppress human-readable output
      -h              display this message and exit
      -v              display version and exit