Sponsored Content
Full Discussion: performance
Top Forums UNIX for Advanced & Expert Users performance Post 82829 by blowtorch on Wednesday 7th of September 2005 02:23:23 AM
Old 09-07-2005
Check the procs columns, r is for process that are currently runnable and waiting for cpu time, b is for processes that are blocked for some IO. The paging column is also important. You should not have very high values of po - pageouts and sr - pages scanned by the scan algo. These mean that your system is spending more time doing paging than carrying out user tasks.

But the exact threshold values of these depend on your system config, the applications running on the system and a host of other parameters, which makes it quite difficult to come up with exact values.
 

7 More Discussions You Might Find Interesting

1. UNIX for Advanced & Expert Users

Performance

Hello, i have changed a slow server with Solaris 7 to a bigger one with Solaris 8 (Sun Ultra 2). Now i have a real bad performance problem (only CPU). Solaris 7 ran with standard FTP and Samba 2.0.7. The new machine is running ProFTP and Samba 2.0.9. There are a lot of NFS Shares and... (5 Replies)
Discussion started by: olso
5 Replies

2. UNIX for Advanced & Expert Users

performance

Hi, I have this on a AIX UNIX machine : ps aux| head -20 USER PID %CPU %MEM SZ RSS TTY STAT STIME TIME COMMAND root 516 23.7 0.0 12 15808 - A 19:38:15 903:13 wait root 774 23.7 0.0 12 15808 - A 19:38:15 902:13 wait root 1290 23.6 0.0 ... (2 Replies)
Discussion started by: big123456
2 Replies

3. UNIX for Advanced & Expert Users

I/O performance

i want to determine I/O performance of an executable, but iostat dont give correct results because the disk that i am writing to and reading from, are not physical disk of the host machine, instead of these local disks we are using a network storage. is there any standard way in unix to get... (2 Replies)
Discussion started by: gfhgfnhhn
2 Replies

4. UNIX for Advanced & Expert Users

performance issue

Hi, on a linux server I have the following : vmstat 2 10 procs memory swap io system cpu r b w swpd free buff cache si so bi bo in cs us sy id 0 4 0 675236 39836 206060 1617660 3 3 3 6 8 7 1 1 ... (1 Reply)
Discussion started by: big123456
1 Replies

5. News, Links, Events and Announcements

Announcing collectl - new performance linux performance monitor

About 4 years ago I wrote this tool inspired by Rob Urban's collect tool for DEC's Tru64 Unix. What makes this tool as different as collect was in its day is its ability to run at a low overhead and collect tons of stuff. I've expanded the general concept and even include data not available in... (0 Replies)
Discussion started by: MarkSeger
0 Replies

6. Solaris

best way and best performance

Hi all, I have two storadge 3510 Fc .. with 12 disks 146Gb ..total 1752Gb each storadge. I need to use about 1.4 Tb of it. and want RAID1 .. I need 13 mount points .. So question: for best performance and redundjancy how I must do it. create 13 logical drives on each stordge with same size... (1 Reply)
Discussion started by: samar
1 Replies

7. UNIX for Dummies Questions & Answers

Performance issue

hi I am having a performance issue with the following requirement i have to create a permutation and combination on a set of three files such that each record in each file is picked and the output is redirected in a specific format but it is taking around 70 odd hours to prepare a combination... (7 Replies)
Discussion started by: mad_man12
7 Replies
lotsfree_pct(5) 						File Formats Manual						   lotsfree_pct(5)

NAME
lotsfree_pct - sets the upper bound before paging daemon starts stealing pages, specified as a percentage of available physical memory VALUES
Default Allowed values Must not be less than (see desfree_pct(5)). DESCRIPTION
is a tunable parameter to set the upper bound before the paging daemon starts stealing pages. It is specified as a percentage of physical memory available after boot. must be a positive integer between 0 and 30 and must be equal to or greater than is the lower bound before paging daemon starts stealing pages (see desfree_pct(5)). The paging daemon in HP-UX acts on a "two hand" model. The daemon runs at least once a second, with one part marking pages of virtual address space as "unused". If the page is referenced before this "aging" hand returns, it will be marked as "used" again. Another section of the daemon, the "steal" hand follows the age hand (the distance between them varies in a well bounded range) and processes pages which are still marked unused, since needed or frequently accessed pages would have a high likelihood of being referenced in the gap between the hands. Exactly how the steal hand treats the pages still marked as unused depends on the comparison of free system physical memory and three pag- ing parameters: and Between and is a periodically re-calculated and, hence, floating threshold known as Thus and are the upper and lower bounds between which moves. is an expression how much the system maintains, in percentage terms (percentage of physical memory available after boot). o If is greater than the system's memory availability is in good shape. The steal hand does nothing as a lot of memory is still available on the system, so there is no need to "steal" a page that might be needed soon. o In the more common case, when is lower than (but higher than the steal hand will begin to steal pages that have remained unreferenced from when the age hand last marked them. "Stealing" a page refers to the process of freeing the page from being allocated for a partic- ular virtual page and making it available for general allocation again. If falls below but still is more than memory availability is still in reasonable shape, but the paging daemon begins stealing pages more aggressively, including putting lower-priority processes to sleep to free up their memory. o When is less than memory availability is getting more critical; and the paging daemon begins deactivating low-priority processes, in other words, swapping out pages to the swap device. Since the exact memory topography varies widely across supported platforms, these three boundary values are tunable to allow for cases where the default values are insufficient, or where the system administrator requires more control over the actions of the paging daemon. In general, however, the automatic calculation performed by default should suffice for most systems. is a system calculated value. Who Is Expected to Change This Tunable? Anyone with super-user privileges. Restrictions on Changing Changes to this tunable take effect immediately. When Should the Value of This Tunable Be Raised? The tunable should only be raised if an administrator feels that the current value provides insufficient free physical memory required by a short lived but critical process. Increasing (and hence to an amout equal to or greater than the amount of physical memory required by this process allows the swapper to move out less frequently used pages from longer running programs, increasing the probability of having the desired amount of memory on hand without waiting for to drop below or during the execution of this critical process. Unless the current is set very low, raising for a process that is not short lived is fairly pointless as will almost certainly drop below and more likely even below thus waking up the swapper and allowing for the physical memory to be freed up over the longer running time as the process requires it. What are the Side Effects of Raising the Value? The swap daemon will begin to steal less-frequently accessed pages from running processes. If large amounts of physical memory is already available without the above situation, this will only slow the already running processes if they need to swap back in their stolen pages. Large amounts of free physical memory without short-lived processes that need it (as mentioned above) is simply a waste. When Should the Value of This Tunable Be Lowered? The tunable should be lowered on systems where swapping is occurring that still have more physical memory available than is needed on aver- age, and that have no sudden sharp peaks in memory usage. What are the Side Effects of Lowering the Value? System swap will occur at a lower memory level. If is set too low (keeping in mind that can never be greater than the system might stall due to memory starvation upon new allocation even though swap space is available. The stall would not be fatal, as the swap daemon run eventually, but will certainly hamper performance. What Other Tunable Values Should Be Changed at the Same Time? should be considered when modifying since Attempting to set above is meaningless and will result in an error. Notes Starting with HP-UX 11iv2, HP recommends the use of to specify the upper bound for the paging daemon. ERROR VALUES
Memory resource management infrastructure is busy. Please try later. Attempting to set above An invalid value is entered that is not an integer between 0 and 30, such as an negative number. WARNINGS
All HP-UX kernel tunable parameters are release specific. This parameter may be removed or have its meaning changed in future releases of HP-UX. Installation of optional kernel software, from HP or other vendors, may cause changes to tunable parameter values. After installation, some tunable parameters may no longer be at the default or recommended values. For information about the effects of installation on tun- able values, consult the documentation for the kernel software being installed. For information about optional kernel software that was factory installed on your system, see at AUTHOR
was developed by HP. SEE ALSO
desfree_pct(5). Tunable Kernel Parameters lotsfree_pct(5)
All times are GMT -4. The time now is 11:43 AM.
Unix & Linux Forums Content Copyright 1993-2022. All Rights Reserved.
Privacy Policy