Free benchmark software for linux

bandwidth project is an artificial benchmark for measuring memory bandwidth, useful for identifying a computers weak areas.
It tests several types of memory:
- Main memory read accesses
- Main memory write accesses
- Level 2 cache read accesses
- Level 2 cache write accesses
- Framebuffer read accesses
- Framebuffer write accesses
- String library routines
Enhancements:
- 64-bit support has been added.

John The Ripper MPI is an updated version of Ryan Lims patch for john the ripper to support MPI, in addition to a large number of third party patches to support additional ciphers and such.
MPI allows you to use multiple processors on a single system, or a cluster of systems for cracking passwords using john the ripper. Incredibly usefull in these days of multi core processors.
A compatible MPI implementation is required.
Benchmarks
Dual Core Intel Core2 Duo, 2.16GHz, MacBook Pro MacOSX
John MPI - Version 1.7.2-bp17-mpi4, 2 threads
Benchmarking: Traditional DES [128/128 BS SSE2]... DONE
Many salts: 3433087.00 c/s real, 3636739.00 c/s virtual
Only one salt: 2852658.00 c/s real, 3021898.00 c/s virtual
Dual AMD Opteron 250 (2.2ghz), Gentoo Linux 64bit
John MPI - Version 1.7.2-bp17-mpi, 2 threads
Benchmarking: Traditional DES [128/128 BS SSE2-16]... DONE
Many salts: 2132632.00 c/s real, 2139034.00 c/s virtual
Only one salt: 1951692.00 c/s real, 1957552.00 c/s virtual
Version restrictions:
- If you use Gentoo, make sure you compile mpich2 _WITHOUT_ the "threads" use flag, otherwise john-mpi will fail to compile with undefined references to MPIU_Free and MPIU_Malloc
- If you use OpenMPI instead of mpich2, the SIGHUP signal doesnt get passed to john. It is necessary to send a SIGUSR1 instead.
Enhancements:
- Support for SIGUSR1 as well as SIGHUP, required for use with OpenMPI

fio is an I/O tool meant to be used both for benchmark and stress/hardware verification.
fio project has support for 4 different types of I/O engines (sync, mmap, libaio, posixaio), I/O priorities (for newer Linux kernels), rate I/O, forked or threaded jobs, and much more. It can work on block devices as well as files.
fio accepts job descriptions in a simple-to-understand text format. Several example job files are included. fio displays all sorts of I/O performance information, such as completion and submission latencies (avg/mean/deviation), bandwidth stats, cpu and disk utilization, and more.
Options:
$ fio
-s IO is sequential
-b block size in KiB for each io
-t < sec > Runtime in seconds
-r For random io, sequence must be repeatable
-R < on > If one thread fails to meet rate, quit all
-o < on > Use direct IO is 1, buffered if 0
-l Generate per-job latency logs
-w Generate per-job bandwidth logs
-f < file > Read < file > for job descriptions
Examples using a job file:
A sample job file doing the same as above would look like this:
[read_file]
rw=0
bs=4096
[write_file]
rw=1
bs=16384
And fio would be invoked as:
$ fio -o1 -s -f file_with_above
The second example would look like this:
[rf1]
rw=0
prio=6
[rf2]
rw=0
prio=3
[rf3]
rw=0
prio=0
direct=1
And fio would be invoked as:
$ fio -o0 -s -b4096 -f file_with_above
global is a reserved keyword. When used as the filename, it sets the default options for the threads following that section. It is possible to have more than one global section in the file, as it only affects subsequent jobs.
Enhancements:
- Support for replaying blktrace logs, full splice support (network and normal), adaptive latency histograms, many data verification improvements and additions, and many bugfixes and improvements.

svk is a decentralized version control system written in Perl. svk project uses the Subversion filesystem but provides additional, powerful features. svk was originally developed by ChiaLiangKao, with the help from SVKPeople and generous SVKSponsors.
Main features:
- svk uses the subversion fs library. the SVN::XD module is a svn_wc (working copy) replacement (XD is character increment of WC).
- SVN::XD does not use the .svn for checkout meta-data. it uses Data::Hierarchy for storing per-directory meta-data in a compact way. Read about WhyXD.
- svk uses svm (svn mirror) for retrieving remote repositories (via ra) to a certain point of the local repository. There are plans to integrate svm and vcp so we have instant ability to branch remote cvs or perforce.
- the depot spec is like : . ~/.svk/config holds a depotmap hash for mapping depotname (which could be empty) to a path containing a subversion repository.
- when committing (or merging) into the mirrored directory, svm provides a commit editor for svk to use. the merge brings changes from local to remote repository if theres no conflict.
- using a local repository holding the subversion tree for benchmark, svk checkout time is 3 times faster than svn.
Enhancements:
- Perl 5.9.5 compatibility, various fixes in the merge subsystem, and some Win32 fixes.

GliBench is a Gui based benchmarking tool to check your computers CPU and hard disk performance. The software is based on the benchmarks I developed for CliBench Mk III SMP a SMP enabled benchmark program for Win32.
There were several tries to port it to other architectures than Win32, but this was not that easy. So I decided to to a Linux port, based on the GTK toolkit with Gnome support, as Linux runs on almost every hardware around and GTK is quite easy to port to other OSs.
The CPU tests are all ported to ANSI C. They run fully multithreaded using posix threads. You can already use the program for benchmarking your hardware.
Enhancements:
- The application was completely rewritten for GTK+ 2.x and Glib 2.x.
- New stress tests and threaded I/O tests for benchmarking hard drives are available.
- Both console and GUI interfaces are available.

Array::Unique is a tie-able array that allows only unique values.

SYNOPSIS

use Array::Unique;
tie @a, Array::Unique;

Now use @a as a regular array.

This package lets you create an array which will allow only one occurrence of any value.

In other words no matter how many times you put in 42 it will keep only the first occurrence and the rest will be dropped.

You use the module via tie and once you tied your array to this module it will behave correctly.

Uniqueness is checked with the eq operator so among other things it is case sensitive.

As a side effect the module does not allow undef as a value in the array.

EXAMPLES

use Array::Unique;
tie @a, Array::Unique;

@a = qw(a b c a d e f);
push @a, qw(x b z);
print "@an"; # a b c d e f x z

When you are collecting a list of items and you want to make sure there is only one occurrence of each item, you have several option:

1) using an array and extracting the unique elements later. You might use a regular array to hold this unique set of values and either remove duplicates on each update by that keeping the array always unique or remove duplicates just before you want to use the uniqueness feature of the array. In either case you might run a function you call @a = unique_value(@a);

The problem with this approach is that you have to implement the unique_value function (see later) AND you have to make sure you dont forget to call it. I would say dont rely on remembering this.

There is good discussion about it in the 1st edition of the Perl Cookbook of OReilly. I have copied the solutions here, you can see further discussion in the book.

----------------------------------------
Extracting Unique Elements from a List (Section 4.6 in the Perl Cookbook 1st ed.)

# Straightforward

%seen = ();
@uniq = ();
foreach $item (@list) [
unless ($seen{$item}) {
# if we get here we have not seen it before
$seen{$item} = 1;
push (@uniq, $item);
}
}

# Faster
%seen = ();
foreach $item (@list) {
push(@uniq, $item) unless $seen{$item}++;
}

# Faster but different
%seen;
foreach $item (@list) {
$seen{$item}++;
}
@uniq = keys %seen;

# Faster and even more different
%seen;
@uniq = grep {! $seen{$_}++} @list;

----------------------------------------
2) using a hash
Some people use the keys of a hash to keep the items and
put an arbitrary value as the values of the hash:

To build such a list:
%unique = map { $_ => 1 } qw( one two one two three four! );

To print it:
print join ", ", sort keys %unique;

To add values to it:
$unique{$_}=1 foreach qw( one after the nine oh nine );

To remove values:
delete @unique{ qw(oh nine) };

To check if a value is there:
$unique{ $value }; # which is why I like to use "1" as my value

(thanks to Gaal Yahas for the above examples)

There are three drawbacks I see:

1) You type more.
2) Your reader might not understand at first why did you use hash and what will be the values.
3) You lose the order.

Usually non of them is critical but when I saw this the 10th time in a code I had to understand with 0 documentation I got frustrated.

3) using Array::Unique

So I decided to write this module because I got frustrated by my lack of understanding whats going on in that code I mentioned. In addition I thought it might be interesting to write this and then benchmark it. Additionally it is nice to have your name displayed in bright lights all over CPAN ... or at least in a module.

Array::Unique lets you tie an aray to hmmm, itself (?) and makes sure the values of the array are always unique.

Since writing this I am not sure if I really recommend its usage. I would say stick with the hash version and document that the variable is aggregating a unique list of values.

4) Using real SET

There are modules on CPAN that let you create and maintain SETs. I have not checked any of those but I guess they just as much of an overkill for this functionality as Unique::Array.