File cache portion of memory on Solaris Post: 302324270

Sponsored Content

Operating Systems Solaris File cache portion of memory on Solaris Post 302324270 by Devyn on Wednesday 10th of June 2009 11:03:55 AM

06-10-2009

Registered User

Thanks jlliagre. That moves me forward.

> Only pages are cached, i.e. portions of files that are actually used, not whole files.
Ok. That's fine too. Would you know which parts of kstat output shows the total # of pages cached or at least which vars contain this so I can calculate this?

> Assuming you are only using UFS and no FZS, the file cache is displayed in this line on your report:
Yes I know. I was interested in getting this number using kstat values as mdb is slow and because of this impractical.

> Also,
> Is the portion of physical memory used allocated to the cache?
> Is it part of the physical memory allocated to swap?
-> The file cache is obviously RAM based. There would be no interest to cache disks
-> blocks into different other disk blocks.

Yes of course. Let me rephrase. On Solaris, the memory reported by vmstat per my understanding already does not include cache. Solaris considers the cache portion as memory it can grab as needed so it sees it as free. This makes sense to me and so any usage reported through Solaris commands is the portion used that is not cache. Please correct me if I'm wrong.

In light of above, of the physical ram Solaris sees as free, what portion of that is Cache?

Again, thank you for the help so far.

Devyn

View Public Profile for Devyn

Find all posts by Devyn

10 More Discussions You Might Find Interesting

1. Windows & DOS: Issues & Discussions

Clearing of cache memory

Pls! will someone help me on how to clear my cache memory of my Internet browser Thanks Kayode

2. Solaris

Solaris File Cache

I started a previous thread : https://www.unix.com/sun-solaris/76721-coredumps-swap-part-solaris-mem-consumption.html But my fault as there is no xml feed for this forum i lost track and thread got closed. So in the mean time i went through some docs and here are my queries : 1. mdb -k...

3. UNIX for Dummies Questions & Answers

High use of cache memory

Hi, I'm running a debian lenny 1GB ram, but with a high I/O. This server has 400IOPS and 3MB/s sustain. So, I noted cached memory use 800MB, buffered memory use 50MB, and no free memory is available. Questions: What does mean such a high cached memory? Who's using this cached memory? Is...

4. Linux

File cache /Page cache Linux

Hi All, could any one point out any open source test-suites for "File cache" testing and as well as performance test suites for the same. Currently my system is up with Linux/ext4. Regards Manish

5. Red Hat

Need to release Cache memory

Right now i am using Red Hat Enterprise Linux AS release 4 and cache memory occupying around 1.5GB mentioned below, total used free shared buffers cached Mem: 2026 2021 5 0 161 1477 -/+ buffers/cache: 382 1644 ...

6. Programming

How access a specific memory portion through printf() function????

Hi friends, Hope everyone is doing well. Please have a look at this simple program, you will figure out what I want. #include <stdio.h> int main() { printf("Enter an integer!\n"); scanf( "%d", 134511890 ); // Valid address on my computer printf( "%d\n", ???? ); return 0; } ...

7. Programming

Ways to preserve a memory cache

The environment is Java/Windows. The program keeps near real-time state in memory cache, which is updated by multiple sources, size of the cache is roughly 500 MB, frequency of updates is ~ 20 per second. I am looking into different ways to keep current snapshot of the memory on the disk for a)...

8. Solaris

clear cache memory

hi all, i have noticed that my server has 64 GB RAM and i have application in this server but the server has free memory only 15% and utilized 85% however it didn't eat from swap . does any parameter can be configured in kernel to make the system clear memory from cache like linux i found...

9. UNIX for Dummies Questions & Answers

Clearing memory cache on Linux server

i wish to clear memory cache on a production box and i was wondering what is the worst that can happen if i do? i already tested this on a backup server and everything seemed fine. but i need to know from you experts what are the worst things that can happen when i run it on a real server: ...

10. Programming

Memory and cache access time discrepancy

#include<stdio.h> #include<stdlib.h> #include<sys/time.h> #include<time.h> #include "rdtsc.h" #define SIZE 4*64*1024 int main() { unsigned long long a,b; int arr={0}; int i; register int r; a=rdtsc(); r=arr; b=rdtsc(); printf("1st element Access Cycles = %llu\n",b-a);

LEARN ABOUT DEBIAN

fs_getcacheparms

FS_GETCACHEPARMS(1)					       AFS Command Reference					       FS_GETCACHEPARMS(1)

NAME

       fs_getcacheparms - Displays the current size and usage of the cache

SYNOPSIS

       fs getcacheparms [-help] [-files] [-excessive]

       fs getca [-h] [-files] [-excessive]

DESCRIPTION

       The fs getcacheparms command displays the current size of the cache (which can be in memory or on disk), and the amount currently in use.

       The reported statistics are from kernel memory, so the reported size can differ from the setting specified in the /etc/openafs/cacheinfo
       file on a machine using a disk cache, if the fs setcachesize command has been used to alter cache size.

OPTIONS

       -help
	   Prints the online help for this command. All other valid options are ignored.

       -files
	   Displays the current number of cache files in use, and the maximum available, as well as displaying the current cache size.

       -excessive
	   Displays detailed cache statistics, including the flags set on each cache element, and the distribution of cache entries by size.

OUTPUT

       The output reports

	  AFS using <amount> of the cache's available <size> 1K byte blocks.

       where <amount> is the number of kilobyte blocks currently used to cache data and status information, and <size> is the total current cache
       size.

EXAMPLES

       The following example shows the output on a machine with a 25000 kilobyte cache.

	  % fs getcacheparms
	  AFS using 22876 of the cache's available 25000 1K byte blocks.

       The following shows the output when asked to show file information.

	  % fs getcacheparm -files
	  AFS using	0% of cache blocks (897 of 200000 1k blocks)
			1% of the cache files (79 of 6250 files)

       And finally, the following detailed information can be obtained by using the excessive flag

	  % fs getcacheparm -excessive
	  AFS using	0% of cache blocks (897 of 200000 1k blocks)
			1% of the cache files (79 of 6250 files)
		  afs_cacheFiles:	6250
		  IFFree:		6171
		  IFEverUsed:		  79
		  IFDataMod:		   0
		  IFDirtyPages: 	   0
		  IFAnyPages:		   0
		  IFDiscarded:		   1
		  DCentries:	    3122
		   0k-	 4K:	     74
		   4k-	16k:	   3045
		  16k-	64k:	      2
		  64k- 256k:	      0
		 256k-	 1M:	      1
		       >=1M:	      0

PRIVILEGE REQUIRED

       None

SEE ALSO

       fs_setcachesize(1)

COPYRIGHT

       IBM Corporation 2000. <http://www.ibm.com/> All Rights Reserved.

       This documentation is covered by the IBM Public License Version 1.0.  It was converted from HTML to POD by software written by Chas
       Williams and Russ Allbery, based on work by Alf Wachsmann and Elizabeth Cassell.

OpenAFS 							    2012-03-26						       FS_GETCACHEPARMS(1)

10 More Discussions You Might Find Interesting

1. Windows & DOS: Issues & Discussions

Clearing of cache memory

Discussion started by: kayode

2. Solaris

Solaris File Cache

Discussion started by: rajwinder

3. UNIX for Dummies Questions & Answers

High use of cache memory

Discussion started by: iga3725

4. Linux

File cache /Page cache Linux

Discussion started by: hmanish