Free rhes cluster software for linux

Rocks Cluster is a complete "cluster on a CD" solution for x86 and IA64 Red Hat Linux COTS clusters.
Building a Rocks cluster does not require any experience in clustering, yet a cluster architect will find a flexible and programmatic way to redesign the entire software stack just below the surface (appropriately hidden from the majority of users).
Although Rocks includes the tools expected from any clustering software stack (PBS, Maui, GM support, Ganglia, etc), it is unique in its simplicity of installation.
From a hardware component and raw processing power perspective, commodity clusters are phenomenal price/performance compute engines. However, if a scalable ``cluster management strategy is not adopted, the favorable economics of clusters are offset by the additional on-going personnel costs involved to ``care and feed for the machine. The complexity of cluster management (e.g., determining if all nodes have a consistent set of software) often overwhelms part-time cluster administrators, who are usually domain application scientists. When this occurs, machine state is forced to either of two extremes: the cluster is not stable due to configuration problems, or software becomes stale, security holes abound, and known software bugs remain unpatched.
While earlier clustering toolkits expend a great deal of effort (i.e., software) to compare configurations of nodes, Rocks makes complete Operating System (OS) installation on a node the basic management tool. With attention to complete automation of this process, it becomes faster to reinstall all nodes to a known configuration than it is to determine if nodes were out of synchronization in the first place. Unlike a users desktop, the OS on a cluster node is considered to be soft state that can be changed and/or updated rapidly.
This is clearly more heavywieght than the philosophy of configuration management tools [Cfengine] that perform exhaustive examination and parity checking of an installed OS. At first glance, it seems wrong to reinstall the OS when a configuration parameter needs to be changed. Indeed, for a single node this might seem too severe. However, this approach scales exceptionally well, making it a preferred mode for even a modest-sized cluster. Because the OS can be installed from scratch in a short period of time, different (and perhaps incompatible) application-specific configurations can easily be installed on nodes. In addition, this structure insures any upgrade will not interfere with actively running jobs.
One of the key ingredients of Rocks is a robust mechanism to produce customized distributions (with security patches pre-applied) that define the complete set of software for a particular node. A cluster may require several node types including compute nodes, frontend nodes file servers, and monitoring nodes. Each of these roles requires a specialized software set. Within a distribution, different node types are defined with a machine specific Red Hat Kickstart file, made from a Rocks Kickstart Graph.
A Kickstart file is a text-based description of all the software packages and software configuration to be deployed on a node. The Rocks Kickstart Graph is an XML-based tree structure used to define RedHat Kickstart files. By using a graph, Rocks can efficiently define node types without duplicating shared components. Similiar to mammalian species sharing 80% of their genes, Rocks node types share much of their software set. The Rocks Kickstart Graph easily defines the differences between node types without duplicating the description of their similarities. See the Bibliography section for papers that describe the design of this structure in more depth.
By leveraging this installation technology, we can abstract out many of the hardware differences and allow the Kickstart process to autodetect the correct hardware modules to load (e.g., disk subsystem type: SCSI, IDE, integrated RAID adapter; Ethernet interfaces; and high-speed network interfaces). Further, we benefit from the robust and rich support that commercial Linux distributions must have to be viable in todays rapidly advancing marketplace.
Wherever possible, Rocks uses automatic methods to determine configuration differences. Yet, because clusters are unified machines, there are a few services that require ``global knowledge of the machine -- e.g., a listing of all compute nodes for the hosts database and queuing system. Rocks uses an SQL database to store the definitions of these global configurations and then generates database reports to create service-specific configuration files (e.g., DHCP configuration file, /etc/hosts, and PBS nodes file).
Enhancements:
- Rocks v4.3 is released for i386 and x86_64 CPU architectures. New features: Rocks command line - initial release of the Rocks command line which facilitates non-SQL administrative access to the database; PXE First - hosts can now be configured in BIOS with a boot order of CD, PXE, hard disk. Enhancements: based on CentOS 4.5 and all updates as of July 4, 2007; Anaconda installer updated to 10.1.1.63; performance improvement when building torrent files for the Avalanche Installer; database indirects, more flexibility with Rocks variables; Globus updated to gt4.0.4 with web services....

Image Cluster copies and renames images based on Exif data and file number names. Image Cluster also clusters those images into directories based on a variable sliding window (with a default of 36 hours), which makes it easy to group images based on events without manual intervention.

The inspiration for this program came from recently getting a new Canon SD500 camera to replace my Canon S30 that Id had for years. The upside, the Canon SD500 rocks! The downside, I now have 2 cameras that are burning through the same sequence numbers, so my previous solution of just putting all the files in to directories by the first 2 digits of the sequence numbers was no longer going to work.

Imagecluster solves this problem, plus another grouping problem that Id been thinking about, by extracting the CreateDate and FileNumber exif tags from the images, and using that as the basis of a new filename (typically YYYY:mm:dd_HH:MM:SS_FileNumber.jpg). This ensures that 2 images taken at the same second have an even smaller chance of colliding, as their camera sequence numbers would have to also be the same at that second.

But that is just the first step. I have noticed that I am an occational photographer, so take pictures in bursts, often for a weekend of hanging out with folks, though sometimes for a vacation as well. This got me thinking. What I really needed is a tool that also creates directories that allows for some minimum tollerance between CreateDate, that is used to cluster images. For me, the optimum value seems to be 36 hours, though this is configurable via the command line.

This took me an afternoon to pull together, Im sure it could be smarter, but it is useful enough to post for others to use.

Options:

-d directory
Set the target directory for images. Defaults to /tmp/photos, which is probably not what you want.

-D
Dryrun. Tells you what the program would have done.

-h
Print out help message

-s
Seperator character. It defaults to : (i.e. 2005:10:09...), but is user configurable because my friend Clemens wants to use - (i.e. 2005-10-09) instead.

-t
Set the tollerance for image clustering. This is the maximum time between any 2 pictures in a cluster, which will cause a new cluster to be created. The name of the cluster will be YYYY:MM:DD of the first image in the cluster, even if it spans multiple days. Because this tollerance is the maximum time between any two images in the cluster, it is possible that all images you have ever taken could be in 1 cluster, if you took a picture every day of your life. Hence, this feature isnt useful to everyone. If you are that kind of person, set tollerance to 16 hours or something, and youll tend to get 1 day sized buckets.

-v
Prints verbose output

SenseClusters is a natural language processing package that allows you to cluster similar contexts or to identify clusters of related words.
SenseClusters supports its own native methods based on first and second order representations of context, and also supports Latent Semantic Analysis. It is fully unsupervised, and can automatically discover the optimal number of clusters in your text.
SenseClusters is a complete system that takes users from preprocessing of raw text to providing clustered output.
Enhancements:
- Full support for Latent Semantic Analysis was introduced.
- Both contexts and words may be clustered using either native SenseClusters methods (first or second order) or Latent Semantic Analysis.

Algorithm::Cluster is a Perl interface to the C Clustering Library.

This module is an interface to the C Clustering Library, a general purpose library implementing functions for hierarchical clustering (pairwise simple, complete, average, and centroid linkage), along with k-means and k-medians clustering, and 2D self-organizing maps.

This library was developed at the Human Genome Center of the University of Tokyo. The C Clustering Library is distributed along with Cluster 3.0, an enhanced version of the famous Cluster program originally written by Michael Eisen while at Stanford University.

Cluster Live can do the following:
- Boots a cluster of diskless Thin Clients.
- Automatically loads a Web browser.
- Share workload between Thin Clients.
Why?
- No individual installations required.
- Fast deployment of centralised software over multiple computers.
- No need to access the hard disk of your existing system.
- Portable yet scalable infrastructure.
- Less spending on high end computer hardware such as hard drives, processors and memory.
Who can benefit from it?
- Public kiosk
- Internet cafe
- Home, Office and other intranet premises
- Training centre
- Government sectors, and other organisations requiring large deployment of dedicated applications.
- Ideal for academics in developing countries to make use of computers cost effectively.
- Open Source developers can freely create their own Cluster Live CDs.
How does it work?
- A server gets booted by a Cluster Live CD.
- The clients BIOS is configured to boot through the LAN by default.
- When the server has finished the bootup sequence, the client can be booted through the LAN.
What was it tested with?
- IBM Thinkpad T30 (CPU: Pentium 4 - 1GHz; RAM: 512MB)
- Virtual Machines (Software to allow running of multiple O.S. simultaneously)
- IDE CDRW (To rewrite the ISO image on to the CD for testing)
- A hub with several LAN cables connecting the infrastructure.
- Source built with Red Hat Fedora Core 3 Linux

Bio::ClusterI module is a cluster Perl interface.

SYNOPSIS

# see the implementations of this interface for details but # basically
my $cluster= $cluster->new(-description=>"POLYUBIQUITIN",
-members =>[$seq1,$seq2]);
my @members = $cluster->get_members();
my @sub_members = $cluster->get_members(-species=>"homo sapiens");

This interface is the basic structure for a cluster of bioperl objects. In this case it is up to the implementer to check arguments and initialize whatever new object the implementing class is designed for.

ClusterControl project is a web-interface to simplify distributing and monitoring bioinformatics applications on Linux cluster systems.

This software is based on a modular concept that enables integration of command-line oriented programs into the application framework of ClusterControl. The system enables integration of different applications accessed through one interface and executed on a distributed cluster system.

The use of wide spread and freely available technologies like Apache as web-server, PHP as server-side scripting-language and OpenPBS or Sun Grid Engine (SGE) as queuing system enables straightforward installation on common operating systems.

Installation:

For local installation, ClusterControl requires a running installation of the web server Apache (http://www.apache.org) with PHP(Version 4) and LDAP support, that comes with nearly every Linux distribution, and a local installation of the queuing system Open Portable Batch System (PBS) which can be downloaded from http://www.openpbs.org or Sun Grid Engine (SGE) which is available at http://gridengine.sunsource.net.

If an LDAP-Server is already available for user-management of the cluster it can be easily integrated by modifying the configuration file. To activate the monitoring mechanism of ClusterControl a status collecting server-process must be started on every cluster node. With the adaptation of one well documented configuration file the installation process is completed.

Klusters is a powerful and easy-to-use graphical application for spike sorting of extracellular neuronal recordings.
While the accuracy and speed of spike sorting is dramatically improved by automatic classification algorithms, the results of automatic algorithms must be examined manually to ensure that the assigned clusters correspond to single neurons. Klusters has been carefully designed to make this process as fast and error-free as possible.
Main features:
- Multiple type of views including Cluster View displaying spikes in the PCA feature space, Waveform View and Correlation View displaying auto-and cross-correlograms.
- Cluster colors making it easy to quickly identify a given cluster across different views.
- Time trajectory display to identify possible electrode drift.
- "Grouping Assistant" to suggest which pairs of clusters may result from overclustering of a single neuron.
- "Recluster" feature to automatically split clusters that may be several distinct neurons (underclustering), using an external classification program.
- Multiple displays open at once, which can be useful to examine different subsets of the data in parallel.
- Multiple undo and redo on the last actions.
- Print and export to postcript and pdf.
- Ergonomic user interface for maximum speed and reliability.
Enhancements:
- The source archive klusters-1.4.tar.gz released on July 08 did not compiled on certain distributions, this is now corrected.
- If you have encounter this problem, please download a copy of the corrected archive and try again. [This does not affect the debian package].
- In addition a layout bug was corrected in the documentation. [This affects both the source archive and the debian package].