Globus toolkit

Globus toolkit

Un ejemplo popular de middleware de grid es Globus Toolkit. Éste es un conjunto de herramientas de software requeridas para la construcción de una grid, cubriendo mediciones de seguridad, localización de recursos, manejo de recursos, comunicaciones y más.

Muchos de los grandes proyectos grid emplean Globus Toolkit, software desarrollado por Globus Alliance, un equipo inicialmente compuesto por el grupo de Ian Foster, en el Laboratorio Nacional Argonne, y por el de Carl Kesselman, en la Universidad de California del Sur en Los Ángeles (Estados Unidos).

Many of the protocols and functions defined by the Globus Toolkit are similar to those in networking and storage today, but have been optimized for grid-specific deployments.

Muchos de los protocolos y funciones definidas por Globus Toolkit son similares a aquellas que hoy operan para las redes y los almacenajes, pero han sido optimizadas para desarrollos específicos de grid.

Globus incluye programas como:

• GRAM (Globus Resource Allocation Manager – Recurso Globus de Manejo de Asignación): se encarga de convertir las solicitudes de recursos en comandos que lo computadores locales puedan comprender.

• GSI (Grid Security Infrastructure – Infraestructura de Seguridad Grid): autentifica a los usuarios y determina sus derechos de acceso.

• MDS (Monitoring and Discovery Service - Servicio de Monitoreo y Descubrimiento): reúne información acerca de recursos tales como capacidad de procesamiento, capacidad de ancho de banda, tipo de almacenamiento y más.

• GRIS (Grid Resource Information Centre – Centro de Recursos de Información Grid): recursos de consultas para sus actuales configuraciones, capacidades y status.

• GIIS (Grid Index Information Service – Servicio de Índice de Información Grid): coordina arbitrariamente los servicios GRIS.

• GridFTP (Grid File Transfer Protocol - Protocolo de Transferencia de Archivos Grid): provee un mecanismo de transferencia de datos de alto rendimiento, seguro y robusto.

• Replica Catalog (Catálogo de Réplicas): provee la ubicación en la grid de las distintas réplicas de un grupo de datos determinado.

• Replica Management system (Sistema de Manejo de Réplicas): maneja el Catálogo de Réplicas y el GridFTP, permitiendo a las aplicaciones crear y manejar réplicas de grandes grupos de datos.  

Hay dos razones para el fortalecimiento y la popularidad de Globus Toolkit:

1.      Las grid necesitan soportar una amplia variedad de aplicaciones creadas de acuerdo a diferentes paradigmas de programación. En vez de proveer un modelo uniforme de programación para las aplicaciones grid, Globus Toolkit tiene un “acercamiento orientado al objeto”, proveyendo una bolsa de servicios para que los desarrolladores puedan escoger el servicio que mejor se adecua a sus necesidades. Además, las herramientas pueden introducirse de a una a la vez. Por ejemplo, una aplicación puede usar GRAM o GRIS sin tener que usar necesariamente Globus Security o los sistemas de manejo de réplicas.

2.      Globus Toolkit está disponible bajo acuerdo de licencia de “fuente abierta”, lo que significa que cualquiera el libre de utilizar o mejorar el software. Esto es similar a la World Wide Web y a los sistemas de operación Linux.

http://www.globus.org/toolkit/

sinfo-Advanced Network Monitoring

sinfo-Advanced Network Monitoring

Are you looking to set up some kind of network cluster, but dealing with many different computers, all of which are nearly impossible to keep track of? What if you're in charge of a room full of computers and also of those who are using them (some of whom may be looking to slack off or run something I'll politely dub "objectionable")? sinfo may be what you're looking for. According to its Freshmeat entry:

sinfo is a monitoring tool that uses a broadcast scheme to distribute information on the status of each computer on your local network. It supports CPU, memory usage, network load and information about the top five processes on each computer. sinfo uses ncurses to display the information in an attractive manner.

sinfo displays invaluable system information over multiple PCs for easy administration.
Further extending sinfo's use with the -s switch provides even meatier information.

Installation

A binary is available for those with Debian-based systems, such as Debian, Ubuntu and the like, and chances are that it's already sitting in your repository. Given that this software also includes a startup dæmon, sinfod, I thoroughly recommend taking the binary option if you can, because a great deal of the process is automated (it's also the version I cover here). Nevertheless, in the interests of distro neutrality, I also cover the source version during the installation, as usual.
As far as library requirements are concerned, you need the following, according to the documentation:

• ncurses: libraries for terminal handling (tested with version 5.7).
• boost: peer-reviewed portable C++ source libraries using Boost.Bind and Boost.Signals (tested with version 1.42).
• asio (>=1.1.0): asio is a cross-platform C++ library for network programming (tested with version 1.4.1).

If you're compiling from source, you need the pesky developer packages as well (-dev). The number of packages under libboost- in particular are quite extensive, so if you run into any problems during make, the libboost libraries may be the culprit, requiring more of its packages before you can install properly.
For those running with source, once you have the library requirements out of the way, grab the latest tarball and extract it. Open a terminal in the new folder and enter the following commands:

$ ./configure
$ make

If your distro uses sudo:

$ sudo make install

If your distro uses root:

$ su
# make install

Alternative

$ sudo apt-get install sinfo
Before I continue, I should explain that sinfo is broken into two parts: the everyday application and the background dæmon. As for installing the dæmon, I'm going to leave this part to you, because it seems that just about every distro has a different method of dealing with startup processes (and the Debian package takes care of all that). Check the readme files in the source tarball and the Web site for more information if you're still using the source.

Usage

sinfo is a "semi-GUI" command-line program that's actually pretty easy to use, although advanced users can make it do some pretty cool stuff via command-line switches. To run the program in its basic mode, simply enter:

$ sinfo

Assuming that you have sinfo installed only on your machine for now, the information being displayed will be just that of your machine. From this screen, you see all kinds of useful information, such as available memory, CPU utilization, hostnames and so on. The sinfo Keyboard Commands sidebar shows a list of the keyboard commands that drive sinfo, toggling parts of the program with a single keystroke.
However, in this state, it's really just a glorified version of top. The whole point is that you can display information from several machines at once to keep tabs on a LAN. If the other PCs on your network have sinfo in their distro's repository, it should be as easy as installing sinfo via apt-get and the like, then running the program on those machines. As soon as I did that on a second machine, voilà, both PCs were displaying under both installations. Keep installing it on other networked PCs, and the list will grow larger and larger.
Those are the basics out of the way, but what about the extended features? Obviously, I don't have the space to cover everything here (and you really should check the man page for more details), but let's look at some of my favorite features.
At the command line, if you add the -W switch (or alternatively --wwwmode) like so:

$ sinfo -W

the output changes from the usual top-like screen to HTML output instead—very handy for those into things like remote administration with automated Web pages and whatnot.
This is slightly redundant while in the sinfo user interface (simply press the s key), but it's awesome when doing some kind of command-line scripting: add the switch -s (or alternatively --systeminfo), and a big chunk of serious system information is output as well. As an example, my two machines had the following extra information:

192.168.1.2   knightro-bigdesktop i686 
 ↪Linux 2.6.32-27-generic #49-Ubuntu SMP Wed De
cpus: 4  MHz:  800.0
RAM: 3276.5 MByte   swap: 7629.4 Mbyte
load 1min:  0.0   load 5min:  0.1   load 15min:  0.1

192.168.1.1   nhoj-desktop x86_64 
 ↪Linux 2.6.38-8-generic #42-Ubuntu SMP Mon Apr 11 0
cpus: 2  MHz: 1000.0
RAM: 2007.6 MByte   swap: 2047.3 Mbyte
load 1min:  0.1   load 5min:  0.2   load 15min:  0.1
uptime     0 days, 19:13:03

This kind of information suggests many potential uses, and keeping watch and troubleshooting at LAN parties springs instantly to mind. If any one node is having trouble, there's a good chance that the host will be able to hit the ground running when trying to isolate the problem.
In the end, sinfo is not only well designed, but if it's installed as binary, it's also convenient. Ultimately, I think this application will carve out an instant niche for itself, and hopefully, it will turn into one of those standard applications that are so commonplace, they just become part of the landscape. Maybe porting it would do exactly that.

sinfo Keyboard Commands

• q — quit sinfo.
• Page up, Page down — scroll the screen by one page.
• Up arrow/u, down arrow/d — scroll the screen by one line.
• Home — jump to the top line.
• s — toggle display of system information.
• o — toggle display of your own processes.
• n — toggle display of network information.
• D — toggle display of disk load.
• t — toggle display of the top X processes.
• c — toggle the scaling of the CPU load bars from "log", "lin" to full.

 http://www.ant.uni-bremen.de/whomes/rinas/sinfo

Reporte de Actividades

Esta semana lo que realize fue instalar la VPN en el nuevo servidor tambien instale 64 bits en los nodos que tengo y funciona excelente esta mas rápido el procesamiento se instalo ubuntu 10.04 LTS 64 bits  ya se empezo a  generar los certificados a cada quien para que ingresen a la vpn y puedan trabajar

Se hizo una lista donde se le asignara un hotsname a cada computadora para que no se repitan

Se realizo una reunion organizacional con los integrantes del grupo se hizo un calendario donde se discutio diversos puntos sobre el proyecto tambien se definio el objetivo ya que varios tenian diferentes ideas

Una captura de pantalla donde se muestra el Servidor y el Nodo en la vpn en una arquitectura de 64 bits