06-16-2003
Unix has a buffer cache which is a very large chunk of main memory.
When you issue a disk read, unix sees if the data is already in the buffer cache, if so no i/o occurs, the data is just moved into your program's buffer. When you issue a write, the data is just moved into the buffer cache.
The system periodicly scans the buffer cache and flushes new data to disk. This is called "write-behind". The system also predicts data that you are likely to need and issues reads to get the data into the buffer cache before your program wants it. This is called "read-ahead".
Read-ahead and write-behind work so well that most disk i/o is reduced to just moving data in memory.
10 More Discussions You Might Find Interesting
1. Shell Programming and Scripting
i have done a coding in shell scripting which reads a file line by line and does something....it takes more than 30 seconds to execute for a single file. when i do the same with perl scripting it takes less than a second.
is that shell scripting is not efficient while working with large number of... (1 Reply)
Discussion started by: brkavi_in
1 Replies
2. Solaris
getting delayed to connect to my database in solaris
Does anyone help me to get out of this issue
help will be more appreciated. (2 Replies)
Discussion started by: Gods_help
2 Replies
3. Programming
My application is coded in C++ and uses pro C to get data from oracle database. My executable file is taking more time to execute for some specified input. Is there any way to check where it spends the maximum time? (1 Reply)
Discussion started by: SamRoj
1 Replies
4. UNIX for Dummies Questions & Answers
Hello,
like the title says, how can i measure the time it takes to load a module in Linux, and how how can i measure the time it takes to load a statically compiled module.
/Best Regards Olle
---------- Post updated at 01:13 PM ---------- Previous update was at 11:54 AM ----------
For... (0 Replies)
Discussion started by: ollebanan
0 Replies
5. UNIX for Dummies Questions & Answers
When you open a file on a fileserver from your client, what type of operation takes place on the server? I imagine the server's hard drive reads the file that the client has requested, and then makes it available somehow through the network share, is that accurate? Also what service on Unix/Linux... (0 Replies)
Discussion started by: glev2005
0 Replies
6. Shell Programming and Scripting
Dear experts
I have a 200MG text file in this format:
text \tab number
I try to sort using options -fd and it takes very long! is that normal or I can speed it up in some ways?
I dont want to split the file since this one is already splitted.
I use this command: sort -fd file >... (12 Replies)
Discussion started by: voolek
12 Replies
7. Red Hat
When I am trying to login to a server using Putty ssh from my windows xp machine it is taking too much time for login.
My server name is http2 and when I enter this address into putty I immediately get
login as:
screen, but after I put my username root and enter the cursor moves downwards... (14 Replies)
Discussion started by: abhilashkrishn
14 Replies
8. AIX
telnet or rsh or ftp to an AIX machine takes too much time, like have to wait 3 minutes before the prompt appears.
What can I look to troubleshoot this problem
thanks (4 Replies)
Discussion started by: filosophizer
4 Replies
9. Shell Programming and Scripting
Hi,
I wish to check the return value for wget $url.
However, some urls are designed to take 45 minutes or more to return.
All i need to check if the URL can be reached or not using wget.
How can i get wget to return the value in a few seconds ? (8 Replies)
Discussion started by: mohtashims
8 Replies
10. Shell Programming and Scripting
It takes 6 hrs for a 90 GB zip file that i am copying / transferring from serverA onto serverB.
scp user1@serverA:/opt/setup/cash.zip .
Output:
cash.zip 21% 19GB 4.7MB/s 4:11:46 ETA
uname -a
SunOS serverB 5.11 11.2 sun4v sparc sun4vCan you please suggest if i could do... (11 Replies)
Discussion started by: mohtashims
11 Replies
LEARN ABOUT HPUX
dbc_min_pct
dbc_max_pct(5) OBSOLETED dbc_max_pct(5)
NAME
dbc_max_pct, dbc_min_pct, bufcache_max_pct, bufpages, nbuf - OBSOLETED kernel tunable parameter
DESCRIPTION
These tunables have been obsoleted and removed. Do not make any changes to these tunables, as they have no effect on the kernel.
Use the file cache tunables and (see filecache_max(5)).
In previous releases, the tunables and were used to set limits to the dynamic buffer cache, and the tunables and were used for tuning the
buffer cache when a static cache was desired.
This release of HP-UX offers improved file caching technology and improved physical memory control associated to caching file I/O data.
The number of tunable parameters used to control file caching memory usage is reduced from five to two, and previous usability issues are
addressed.
Restrictions on Changing
These tunables should not be modified. Attempting to tune any of the obsolete buffer cache tunables, or results in an error.
Use the tunables and to set limits on the file cache. Note that, on a any given system, the optimum values of these two new tunables are
not necessarily equivalent to the optimum values of the obsolete buffer cache tunable values in the older systems. You should first deter-
mine if the new default values yield acceptable performance on your system, before attempting to change the values of the new file cache
tunables.
AUTHOR
and were developed by HP.
SEE ALSO
filecache_max(5).
Kernel Tunable Parameter dbc_max_pct(5)