Tool to simulate non-sequential disk I/O (simulate db file sequential read) in C POSIX


 
Thread Tools Search this Thread
Top Forums Programming Tool to simulate non-sequential disk I/O (simulate db file sequential read) in C POSIX
# 8  
Old 06-18-2011
Happy with it not being completely random, as long as it jumps around enough to not catch itself on the read advance routines when running on the SAN, and as simulation goes, the database would more than likely read specific files or parts of files more than others anyway.

Here comes my next exciting problem, a database will only read a certain amount of time, other bits are sorting, writing and whatever else the database gets up to. So I am thinking of adding how manny cycles there are between reads, because just doing random reads and nothing else is giving rather silly results. In its purest form I seem to be able to get 100% WaitI/O and read about 6-12 Kb/s, which is pretty impressive in its ow right.

And todays exiting question, anyone feel like guessing how many cycles in a million that a rather streamlined database spends reading disk dataSmilie
Anyone who guesses the right answer wins an icecream, collectable in Camden, London (U.K.)

/Ben
Login or Register to Ask a Question

Previous Thread | Next Thread

10 More Discussions You Might Find Interesting

1. Shell Programming and Scripting

Sequential Reading from two file in a loop

Hello All, I have two files with me file1.txt and file2.txt file1.txt has: 333 222 111 file2.txt has ccc bbb aaa ccc is related to 333 only, bbb is related to 222 only and aaa is related to 111 only. I have to get the values from each of the file and pass them in the URL... (3 Replies)
Discussion started by: ankur328
3 Replies

2. Shell Programming and Scripting

Read directories sequential based on timestamp

Hi, I have a directory structure like below Directoryname create time d1 12:00 d2 12:05 d3 12:08 I want to read the directories based on timestamp.That is oldest directory must be read first and kick off certain process. ... (7 Replies)
Discussion started by: chetan.c
7 Replies

3. UNIX for Dummies Questions & Answers

Inserting a sequential number into a field on a flat file

I have a csv flatfile with a few million rows. I need to replace a field (field number is 85) in the file with a sequential number. As an example, let's assume there are only 4 fields in the file: A,A,,32 A,A,,27 A,B,,43 C,C,,354 If I wanted to amend the 3rd field in this way my... (2 Replies)
Discussion started by: BristolSmithy
2 Replies

4. Shell Programming and Scripting

sequential to line sequential

Hi I have a file sequential way i.e. written in contineous mode and the Record Seperator is AM from which the record is seperated .Now to process I have to make line sequential,and more over record length is not same it varies as per the input address, AM1234563 John Murray 24 Old streeet old... (5 Replies)
Discussion started by: vakharia Mahesh
5 Replies

5. Shell Programming and Scripting

Sequential comparison (one row with file and so on)

Dear linux experts, I'd like to ask for your support, I've read some posts in this forum about files comparison but haven't found what I'm looking for. I need to create a sequential script to compare row-by-row one file with 34 similar files but without success so far. This is what I get: ... (2 Replies)
Discussion started by: Gery
2 Replies

6. Shell Programming and Scripting

How to sca a sequential file and fetch some substring data from it

Hi, I have a task where i need to scan second column of seuential file and fetch first 3 digits of that column For e.g. FOLLOWING IS THE SAMPLE FOR MY SEQUENTIAL FILE AU_ID ACCT_NUM CRNCY_CDE THHSBC001 30045678 THB THHSBC001 10154267 THB THHSBC001 ... (2 Replies)
Discussion started by: manmeet
2 Replies

7. Shell Programming and Scripting

how to scan a sequential file to fetch some of the records?

Hi I am working on a script which needs to scan a sequential file and fetch the row where 2nd column = 'HUB' Can any one help me with this... Thanks (1 Reply)
Discussion started by: manmeet
1 Replies

8. Shell Programming and Scripting

Finding missing sequential file names

So, I've got a ton of files that I want to go through (ie something like 300,000), and they're all labeled sequentially. However I'm not 100% positive that they are all there. Is there any way of running through a sequence of numbers, checking if the file is in the folder, if not appending it... (2 Replies)
Discussion started by: Julolidine
2 Replies

9. Programming

Reading special characters while converting sequential file to line sequential

We have to convert a sequential file to a 80 char line sequential file (HP UX platform).The sequential file contains special characters. which after conversion of the file to line sequential are getting coverted into "new line" or "tab" and file is getting distorted. Is there any way to read these... (2 Replies)
Discussion started by: Rajeshsu
2 Replies

10. UNIX for Dummies Questions & Answers

inserting uniq sequential numbers at the start of the file

Hi Unix gurus, I have a file. I need to insert sequential number at the starting of the file. Fields are delimited by "|". I know the starting number. Example: File is as follows |123|4test|test |121|2test|test |x12|1test|test |vd123|5test|test starting number is : 120 ... (7 Replies)
Discussion started by: jingi1234
7 Replies
Login or Register to Ask a Question
prealloc(2)							System Calls Manual						       prealloc(2)

NAME
prealloc - preallocate fast disk storage SYNOPSIS
DESCRIPTION
is used to preallocate space on a disk for faster storage operations. fildes is a file descriptor obtained from a or system call for an ordinary file of zero length. It must be opened writable, because it will be written to by size is the size in bytes to be preallocated for the file specified by fildes. At least size bytes will be allo- cated. Space is allocated in an implementation-dependent fashion for fast sequential reads and writes. The EOF in an extended file is left at the end of the preallocated area. The current file pointer is left at zero. The file is zero-filled. Using on a file does not give the file an attribute that is inherited when copying or restoring the file using a program such as or (see cp(1) and tar(1)). It simply ensures that disk space has been preallocated for size bytes in a manner suited for sequential access. The file can be extended beyond these limits by operations past the original end of file. However, this space will not necessarily be allo- cated using any special strategy. RETURN VALUE
Upon successful completion, returns 0; otherwise, it returns -1 and sets to indicate the error. ERRORS
fails and no disk space is allocated if any of the following conditions are encountered: [EBADF] fildes is not a valid open file descriptor opened for writing. [EDQUOT] User's disk quota block limit has been reached for this file system. [EFBIG] size exceeds the maximum file size or the process's file size limit. See ulimit(2). [ENOSPC] Not enough space is left on the device to allocate the requested amount; no space was allocated. [ENOTEMPTY] fildes not associated with an ordinary file of zero length. EXAMPLES
Assuming a process has opened a file for writing, the following call to preallocates at least 50000 bytes on disk for the file represented by file descriptor outfd: WARNINGS
Allocation of the file space is highly dependent on current disk usage. A successful return does not tell you how fragmented the file actually might be if the disk is nearing its capacity. AUTHOR
was developed by HP. SEE ALSO
prealloc(1), creat(2), dup(2), fcntl(2), open(2), prealloc64(2), read(2), ulimit(2), write(2). prealloc(2)