Grid5000:Software: Difference between revisions
No edit summary |
No edit summary |
||
| Line 3: | Line 3: | ||
__TOC__ | __TOC__ | ||
{{Template:Software| | {{Template:Software| | ||
|title=[[OAR]] 2 | |title=[[OAR]] 2 | ||
|purpose=OAR is a resource manager (or batch scheduler) for large clusters. It allows cluster users to submit or reserve nodes either in an interactive or a batch mode. | |purpose=OAR is a resource manager (or batch scheduler) for large clusters. It allows cluster users to submit or reserve nodes either in an interactive or a batch mode. | ||
|logo=[[Image:Logo_oar.png|150px|right]] | |logo=[[Image:Logo_oar.png|150px|right]] | ||
| Line 9: | Line 9: | ||
|status=Production/Stable | |status=Production/Stable | ||
|homepage=http://oar.imag.fr | |homepage=http://oar.imag.fr | ||
|availability=OAR 2 | |availability=OAR 2 is installed on Idpot (Grenoble) | ||
}} | }} | ||
Revision as of 09:40, 2 August 2007
Softwares mainly developed in Grid'5000 and available for its users.
OAR 2
OAR is a resource manager (or batch scheduler) for large clusters. It allows cluster users to submit or reserve nodes either in an interactive or a batch mode.
- Status: Production/Stable
- Contacts: Olivier Richard
- Homepage: http://oar.imag.fr
- Availability: OAR 2 is installed on Idpot (Grenoble)
Kadeploy 2.1.5
Kadeploy is a fast and scalable deployment system towards cluster and grid computing. It provides a set of tools, for cloning, configuring (post installation) and managing a set of nodes. Currently it deploys successfully linux, *BSD, Windows, Solaris on x86 and 64 bits computers.
- Status: Production/Stable
- Contacts: Julien Leduc
- Homepage: http://kadeploy.imag.fr
- Availability: Kadeploy is installed on all Grid'5000 clusters
TakTuk 3
TakTuk is a tool for deploying parallel remote executions of commands to a potentially large set of remote nodes. It spreads itself using an adaptive algorithm and sets up an interconnection network to transport commands and perform I/Os multiplexing/demultiplexing. The TakTuk mechanics dynamically adapts to environment (machine performance and current load, network contention) by using a reactive work-stealing algorithm that mix local parallelization and work distribution.
- Status: Production/Stable
- Contacts: Guillaume Huard
- Homepage: http://taktuk.gforge.inria.fr/
- Availability: -
Kaapi
KAAPI means Kernel for Adaptative, Asynchronous Parallel and Interactive programming. It is a C++ library that allows to execute multithreaded computation with data flow synchronization between threads. The library is able to schedule fine/medium size grain program on distributed machine. The data flow graph is dynamic (unfold at runtime).Target architectures are clusters of SMP machines.
- Status: Release versions are stable
- Contacts: kaapi-dev-info@lists.gforge.inria.fr, Thierry Gautier
- Homepage: http://kaapi.gforge.inria.fr
- Availability: Should work on all clusters
KaVlan
VLAN manipulation tool for network isolation of experiment
- Status: Alpha
- Contacts: Pierre Neyron
- Homepage: http://kadeploy.gforge.inria.fr/
- Availability: tested in Idpot (Grenoble), Grillon (Nancy), Capricorne (Lyon)
Katapult
Katapult is a small, well-tested script to automatically start experiments using deployments. Most experiments start by deploying the nodes, re-deploying the nodes if too many of them failed, copying the user's SSH key to the node, etc. Katapult automates all those tasks.
- Status: Production/Stable
- Contacts: Lucas Nussbaum
- Homepage: http://www-id.imag.fr/~nussbaum/katapult.php
- Availability: tested on most clusters. Doesn't work on Sophia/Azur (Totally obsolete ruby version). Should work fine everywhere else.
GRUDU
GRUDU for Grid5000 Reservation Utility for Deployment Usage is a tool for managing Grid5000 resources, reservations and deployments. It provides :
- the access to the Grid5000 grid environment through a GUI interface allowing the user to get detailed informations on the grid, the clusters, and the present jobs.
- the reservation of resources on default or deploy queue thanks to an user-friendly interface
- the deployment of images on the nodes reserved on the deploy queue (through Kadeploy)
- the acces to a terminal on each level of the grid (the connection frontale, a cluster frontale or the master node of a reservation)
As GRUDU is a JAVA-based and client-side software it is cross-platform and you only need a JVM 1.5.0 or later on your computer to run it.
- Status: Production/Stable
- Contacts: Abdelkader Amar
- Homepage: http://graal.ens-lyon.fr/DIET/grudu.html
- Availability: Portable (written in Java)
Marcel
Marcel is a POSIX-compliant thread library featuring a programmable scheduler designed for hierarchical multiprocessor architectures.
- Status: Production
- Contacts: Samuel Thibault
- Homepage: http://runtime.futurs.inria.fr/marcel/
- Availability: Bordeaux
Mad-MPI
Mad-MPI is an efficient implementation of MPI for fast networks.
- Status: Production
- Contacts: Nathalie Furmento
- Homepage: http://runtime.futurs.inria.fr/MadMPI/
- Availability: Bordeaux
MPICH-Madeleine
MPICH-Madeleine is an MPI implementation for clusters and clusters of clusters with heterogenous networks.
- Status: Production
- Contacts: Nathalie Furmento, Guillaume Mercier
- Homepage: http://runtime.futurs.inria.fr/mpi/
- Availability: Bordeaux
NewMadeleine
the NewMadeleine communication library provides extended capabilities for dynamic communication optimization on top of high performance networks.
- Status: Production
- Contacts: Olivier Aumage, Elisabeth Brunet
- Homepage: http://runtime.futurs.inria.fr/newmadeleine/
- Availability: Bordeaux
|
Grid'5000 users, please send an email to web-staff if you would like to see a new software appear on this page. |