Sponsored Content
Top Forums UNIX for Advanced & Expert Users Gurus needed to diagnose severe performance degradation Post 302350045 by achenle on Wednesday 2nd of September 2009 04:13:05 PM
Old 09-02-2009
Quote:
Originally Posted by Neo
Well, it has been a long time since I had to do these calculations. The limits are published by IEEE (assuming point-to-point in this discussion) - I would have to Google for the numbers.

---------- Post updated at 18:14 ---------- Previous update was at 18:03 ----------

Also, I forgot to mention that the theoretical maximum for point-to-point Ethernet (assuming no other network devices talking on the channel), is different, of course, than the practical maximum based on things like "length of cable run", "crimps in the cable" , "connector losses", etc.

I once was on a site where the entire performance of the network management system was terrible and the problem was a crimped cable (I think someone rolled their chair across it in the data center, LOL)

That is why I advise to focus on the network channel(s) when you are debugging performance issues on distributed applications.
If you really want to get into the weeds, the theoretical max bandwidth utilization of an unswitched Ethernet network is 32% of the nominal rate. The key word is "unswitched".

Once you throw switching into the equation, it's a lot easier to get higher rates. I can sustain 90+ megabytes/sec on a gigE point-to-point link, as long as it's dedicated traffic. Now, it takes newer hardware to do that as even a not-too-old IBM x305 that I've used as a WAN emulator (WANEM : The Wide Area Network Emulator) starts falling behind at about 30 or 40 megabytes/sec. And that's a bit newer albeit smaller box than the Fujitsu PrimePower that's the subject of this discussion.

Also, what are the NFS settings? The NFS version? What are the TCP send and receive hiwat settings? Jumbo frames?

Is direct IO enabled?

What is the exact version of Solaris 9? I'd suspect it needs to be as recent as possible.

Also, was the IO utilization of the older fiber channel configuration ever measured? That'd be nice to know in order to solve this problem.

Knowing the IO utilization now would be good, too. If it's moving 3.8 gbps NOW over NFS, it'd be hard to go much faster than that.
 

9 More Discussions You Might Find Interesting

1. Shell Programming and Scripting

SED GURUS - Help!

I wish to substituite a string on each line but ONLY if it appears within double-quotes: this_string="abc#def#geh" # Comment here I wish to change the "#" characters within the double quoted string to "_": this_string="abc_def_geh" # Comment here ... but as you see, the "comment" hash... (2 Replies)
Discussion started by: Simerian
2 Replies

2. Solaris

error notification and diagnose

Hi All, How does Solaris 9/10 alert the server? Where do you get the error on the server? Is there some kind of verifying of errors (like in AIX, CERTIFY resources or diagnose)? Please let me know. Thanks, itik (4 Replies)
Discussion started by: itik
4 Replies

3. Red Hat

Severe Error while starting the System

Dear All, I am facing a unknown error, I start the Linux (RHEL 4 update 6) as usual. After starting the various services(like network,sendmail,portmap etc) a error appears suddenly. The error looks like : Post_create: setxattr failed, rc=28 (dev=hda2 ino=772685) Post_create: setxattr... (2 Replies)
Discussion started by: akhtar.bhat
2 Replies

4. Shell Programming and Scripting

Performance degradation with KSH93

Hi, I have a script that calls an external program to perform some calculations and then I read with "grep" and "sed" values from the output files. I've noticed that performance of KSH93 degrades with every iteration. The output files are all the same size, so I don't understand why after the... (2 Replies)
Discussion started by: i.f.schulz
2 Replies

5. Shell Programming and Scripting

Performance monitoring help needed.

How would i check for following? 1)open ports in my linux machine. 2)Hard disk read speed. 3)Hard disk write speed. (2 Replies)
Discussion started by: pinga123
2 Replies

6. AIX

Diagnose high disk write IO

Hi, say for example if there is high disk write IO in one disk (detected from NMON), how to we identify what processes is writing on that particular disk? (3 Replies)
Discussion started by: ngaisteve1
3 Replies

7. Shell Programming and Scripting

Severe performance issue while 'grep'ing on large volume of data

Background ------------- The Unix flavor can be any amongst Solaris, AIX, HP-UX and Linux. I have below 2 flat files. File-1 ------ Contains 50,000 rows with 2 fields in each row, separated by pipe. Row structure is like Object_Id|Object_Name, as following: 111|XXX 222|YYY 333|ZZZ ... (6 Replies)
Discussion started by: Souvik
6 Replies

8. Shell Programming and Scripting

Shell script to diagnose the network

i have learnt a little bit of shell scripting but not alot. i want to write a script to diagnose the network using ping and another script to traceroute. how would i do this? (6 Replies)
Discussion started by: stefanere2k9
6 Replies

9. AIX

Ld: 0711-851 SEVERE ERROR:

I need to install python 3.3.0 to AIX 6.1 I created folder where I want to install I downloaded files archive from python official website I extracted it into new folder and ran; 1)./configure --with-gcc="xlc_r" --with-cxx="xlC_r" --disable-ipv6 --prefix=my_folder CXX=xlC_r... (2 Replies)
Discussion started by: AIX_30
2 Replies
NFSIOSTAT(1)                                                    Linux User's Manual                                                   NFSIOSTAT(1)

NAME
nfsiostat - Report input/output statistics for network filesystems (NFS). SYNOPSIS
nfsiostat [ -h ] [ -k | -m ] [ -t ] [ -V ] [ interval [ count ] ] DESCRIPTION
The nfsiostat command displays statistics about read and write operations on NFS filesystems. The interval parameter specifies the amount of time in seconds between each report. The first report contains statistics for the time since system startup (boot). Each subsequent report contains statistics collected during the interval since the previous report. A report con- sists of an NFS header row followed by a line of statistics for each network filesystem that is mounted. The count parameter can be speci- fied in conjunction with the interval parameter. If the count parameter is specified, the value of count determines the number of reports generated at interval seconds apart. If the interval parameter is specified without the count parameter, the nfsiostat command generates reports continuously. REPORT
The Network Filesystem (NFS) report provides statistics for each mounted network filesystem. Transfer rates are shown in 1K blocks by default, unless the environment variable POSIXLY_CORRECT is set, in which case 512-byte blocks are used. The report shows the following fields: Filesystem: This columns shows the hostname of the NFS server followed by a colon and by the directory name where the network filesystem is mounted. rBlk_nor/s (rkB_nor/s, rMB_nor) Indicate the number of blocks (kilobytes, megabytes) read by applications via the read(2) system call interface. A block has a size of 512 bytes. wBlk_nor/s (wkB_nor/s, wMB_nor/s) Indicate the number of blocks (kilobytes, megabytes) written by applications via the write(2) system call interface. rBlk_dir/s (rkB_dir/s, rMB_dir/s) Indicate the number of blocks (kilobytes, megabytes) read from files opened with the O_DIRECT flag. wBlk_dir/s (wkB_dir/s, wMB_dir/s) Indicate the number of blocks (kilobytes, megabytes) written to files opened with the O_DIRECT flag. rBlk_svr/s (rkB_svr/s, rMB_svr/s) Indicate the number of blocks (kilobytes, megabytes) read from the server by the NFS client via an NFS READ request. wBlk_svr/s (wkB_svr/s, wMB_svr/s) Indicate the number of blocks (kilobytes, megabytes) written to the server by the NFS client via an NFS WRITE request. ops/s Indicate the number of operations that were issued to the filesystem per second. rops/s Indicate the number of 'read' operations that were issued to the filesystem per second. wops/s Indicate the number of 'write' operations that were issued to the filesystem per second. OPTIONS
-h Make the NFS report easier to read by a human. -k Display statistics in kilobytes per second. -m Display statistics in megabytes per second. -t Print the time for each report displayed. The timestamp format may depend on the value of the S_TIME_FORMAT environment variable (see below). -V Print version number then exit. ENVIRONMENT
The nfsiostat command takes into account the following environment variables: S_TIME_FORMAT If this variable exists and its value is ISO then the current locale will be ignored when printing the date in the report header. The nfsiostat command will use the ISO 8601 format (YYYY-MM-DD) instead. The timestamp displayed with option -t will also be com- pliant with ISO 8601 format. POSIXLY_CORRECT When this variable is set, transfer rates are shown in 512-byte blocks instead of the default 1K blocks. BUG
/proc filesystem must be mounted for nfsiostat to work. FILE
/proc/self/mountstats contains statistics for network filesystems. AUTHORS
Written by Ivana Varekova (varekova <at> redhat.com) Maintained by Sebastien Godard (sysstat <at> orange.fr) SEE ALSO
sar(1), pidstat(1), mpstat(1), vmstat(8), iostat(1), cifsiostat(1) http://pagesperso-orange.fr/sebastien.godard/ Linux JANUARY 2011 NFSIOSTAT(1)
All times are GMT -4. The time now is 09:37 PM.
Unix & Linux Forums Content Copyright 1993-2022. All Rights Reserved.
Privacy Policy