Sponsored Content
Full Discussion: Making things run faster
Top Forums UNIX for Advanced & Expert Users Making things run faster Post 302248460 by shamrock on Friday 17th of October 2008 05:33:21 PM
Old 10-17-2008
Quote:
Originally Posted by Legend986
Added to that, I have a small question (not so sure if its silly though but can't seem to understand it completely)...

If I have four datasets like in the problem above and all I have to do is grep some text out of it, does it really make a difference doing the jobs parallely on all the datasets or doing them in a sequential order? In fact, to be more precise, the argument goes something like this:

Four datasets are stored on the disk. The CPU has to fetch some data everytime for the four processes to process them and write back to the disk. Now, if it has to provide data to all the four processes, then shouldn't the head keep moving around to provide the data as opposed to just one process where it just keeps reading the data (provided there is no fragmentation). As I said, I'm sorry if my question seems silly but just want to clear some basic concepts.
And that is the reason for caching data and striping it over multiple disks in order to reduce disk arm contention. This way reads/writes are done in parallel and with caching in play most reads/write are logical instead of physical. As you have terabytes of data I am assuming that all of it isn't on a single drive like a JBOD of some sort and that it is on a high end storage array with significant intelligence and caching built into it while being striped for performace and mirrored for availability.
 

8 More Discussions You Might Find Interesting

1. UNIX for Dummies Questions & Answers

making ssh run without password

Hello Everybody, Could anyone please tell me how to get ssh to work without asking for passwords? (i want to do a ssh <hostname> without getting a request for a password but getting connected straight away) I have attempted the following but to no avail :( ... I tried to generate a SSH... (5 Replies)
Discussion started by: rkap
5 Replies

2. Programming

Complicating things?

So basically what im trying to do is ... Open file, read that file, than try to find .. We or we and replace them with I, but not replace the cases where words contain We or we, such as Went, went, etc a and replace them with the, but not replace the cases where words contain a, such as... (1 Reply)
Discussion started by: bconnor
1 Replies

3. Shell Programming and Scripting

When things doesn't run into crontab???

Could someone explain my problem? I've the following script... #! /bin/ksh ... vmquery -m $MediaID | awk ' BEGIN {FS=": " getline expdate <"ExpDate.txt" } $1 ~ /media ID/ {MediaNumber = $NF} ... $1 ~ /number of mounts/ { "date +%Y"|getline YearToday Year4 = YearToday - 4 if... (4 Replies)
Discussion started by: nymus7
4 Replies

4. Shell Programming and Scripting

Can anyone make this script run faster?

One of our servers runs Solaris 8 and does not have "ls -lh" as a valid command. I wrote the following script to make the ls output easier to read and emulate "ls -lh" functionality. The script works, but it is slow when executed on a directory that contains a large number of files. Can anyone make... (10 Replies)
Discussion started by: shew01
10 Replies

5. UNIX for Dummies Questions & Answers

things root can't do

Hey all my co-workers and I are trying to put together a list of things root "Can't" do on any *NIX OS, so I wanted to come here and see what all we could come up with. Here are two to start this off: write to a read only mount FS kill a tape rewind Please add what you know. Thanks,... (5 Replies)
Discussion started by: sunadmn
5 Replies

6. Shell Programming and Scripting

Making script run faster

Can someone help me edit the below script to make it run faster? Shell: bash OS: Linux Red Hat The point of the script is to grab entire chunks of information that concerns the service "MEMORY_CHECK". For each chunk, the beginning starts with "service {", and ends with "}". I should... (15 Replies)
Discussion started by: SkySmart
15 Replies

7. Shell Programming and Scripting

Making a faster alternative to a slow awk command

Hi, I have a large number of input files with two columns of numbers. For example: 83 1453 99 3255 99 8482 99 7372 83 175 I only wish to retain lines where the numbers fullfil two requirements. E.g: =83 1000<=<=2000 To do this I use the following... (10 Replies)
Discussion started by: s052866
10 Replies

8. Shell Programming and Scripting

Optimize shell script to run faster

data.file: contact { contact_name=royce-rolls modified_attributes=0 modified_host_attributes=0 modified_service_attributes=0 host_notification_period=24x7 service_notification_period=24x7 last_host_notification=0 last_service_notification=0 host_notifications_enabled=1... (8 Replies)
Discussion started by: SkySmart
8 Replies
MPB-DATA(1)						    MIT Photonic-Bands Package						       MPB-DATA(1)

NAME
mpb-data - transformations of HDF5 files output by MPB SYNOPSIS
mpb-data [OPTION]... [HDF5FILE]... DESCRIPTION
mpb-data is a utility to perform additional processing and transformations of HDF5 files output by MPB, the MIT Photonic-Bands program. In particular, it is designed to make the output more amenable to visualization by reformatting it into a rectangular grid, extending it to multiple periods, and rescaling the data. MIT Photonic Bands (MPB) is a free program to compute the band structures (dispersion relations) and electromagnetic modes of periodic dielectric structures. HDF5 is a free, portable binary format and supporting library developed by the National Center for Supercomputing Applications at the Uni- versity of Illinois in Urbana-Champaign. A single h5 file can contain multiple data sets; by default, mpb-data operates on all of the MPB- produced datasets in the file, but this can be changed via the -d option, or by using the syntax HDF5FILE:DATASET. mpb-data writes its output datasets as additional datasets in the input file(s), with "-new" appended to the dataset names. Alternatively, it can write its output to a separate file, specified by the -o option. Note also that, by default, the output datasets are identical to the input datasets; you must use one or more of the options below to spec- ify a transformation (e.g. the -r/-e and -n options are very useful). OPTIONS
-h Display help on the command-line options and usage. -V Print the version number and copyright info for mpb-data. -v Verbose output. -o file Write output datasets to file (for the first input file only) rather than as additional datasets in the input file(s) (the default). -r Output a rectangular cell with the same volume as the cell of the input data. This option is particularly useful for visualizing data from non-orthogonal unit cells (e.g. a triangular lattice), as otherwise the data will appear skewed or warped in most graphics programs. This option should almost always be accompanied by the -n option to ensure a uniform resolution. -e x,y,z As the -r option, but also make the first axis of the output along the x,y,z direction (in Cartesian coordinates) instead of along the first lattice vector as for -r. -n n Output n grid points per lattice unit ("a"). This is useful not only for interpolating to finer (or coarser) resolutions, but also to ensure that the resolution is uniform in each direction (to prevent the data from looking distorted when you visualize it). -x mx, -y my, -z mz This tells mpb-data to output multiple periods in the corresponding lattice directions. to use a particular slice of a two- or three-dimensional dataset. e.g. -x 3.2 causes the output of 3.2 periods in the first lattice direction. The default is to output only a single period. -m s Output s periods in each lattice direction; equivalent to: -x s -y s -z s. -T The output has the first two dimensions (x and y) transposed. This is useful in conjunction with the parallel (MPI) version of MPB, which for performance reasons outputs all arrays with the first two dimensions transposed. -T can undo this transposition. -p Pixellized output. Normally, the input data is linearly interpolated to the output grid, but the -p option causes it to instead use the nearest grid point in the input data. This is useful, for example, if you want to study the discretization of the dielectric- function representation. -d name Use dataset name from the input files; otherwise, the first dataset from each file is used. Alternatively, use the syntax HDF5FILE:DATASET, which allows you to specify a different dataset for each file. You can use the h5ls command (included with hdf5) to find the names of datasets within a file. BUGS
Send bug reports to S. G. Johnson, stevenj@alum.mit.edu. AUTHORS
Written by Steven G. Johnson. Copyright (c) 1999, 2000, 2001, 2002 by the Massachusetts Institute of Technology. SEE ALSO
mpb(1) MPB
January 27, 2000 MPB-DATA(1)
All times are GMT -4. The time now is 02:06 AM.
Unix & Linux Forums Content Copyright 1993-2022. All Rights Reserved.
Privacy Policy