Showing posts with label Grid Computing. Show all posts
Showing posts with label Grid Computing. Show all posts

Tuesday, February 24, 2009

SheevaPlug, Make Every Wall Plug A Computer Node


Perhaps this is an easy way to build a multinode grid computer. If you do not have enough plug points you can get those monstrous power extenders and plug in 10 or 20 of Sheevaplugs.
Marvell®’s SheevaPlug™ is a plug computer — an embedded computer that plugs into the wall socket and can run networkbased services that normally require a dedicated personal computer. Featuring a 1.2GHz Marvell Sheeva™ CPU with 512MB of flash memory and 512MB of DDR2 memory, the SheevaPlug provides ample processing power and resources to run any embedded computing application. Network connectivity is via Gigabit Ethernet; peripheral devices can be connected using USB2.0. Software for the SheevaPlug includes multiple Linux distributions and follows the open-source model, making the SheevaPlug an ideal platform on which to develop or port any application. The SheevaPlug development kit contains the SheevaPlug as well as all of the software tools needed to develop applications for the platform.
Marvell

Thursday, April 03, 2008

The Ninf-G5 is now available from APGRID.

I recieved the following information from Ninf-G group. I have been working with Ninf-G for a long time now and the new features suggested about the Version 5.0, Ninf-G5 is making me wanting to upgrade. But because I am working with a bunch of other people, I need to plan out the upgrade. Thank you, Yoshio Tanaka.
The Ninf-G version 5.0.0 is now available for download at the Ninf project home page at: http://ninf.apgrid.org/ .

Ninf-G Version 5.0.0 (Ninf-G5) is a new version of Ninf-G which is a reference implementation of the GridRPC API.

Major functions of Ninf-G include (1) remote process invocation, (2) information services, and (3) communication between Ninf-G Client and Servers. Ninf-G4 is able to utilize various middleware for remote process invocation, however Ninf-G4 utilize the Globus Toolkit for information services and communication between Ninf-G Client and Servers.
On the other hand, Ninf-G5 does not assume specific Grid middleware as prerequisites, that is, unlike the past versions of Ninf-G (e.g. Ninf-G2, Ninf-G4), Ninf-G5 works in non Globus Toolkit
environments. Ninf-G5 is able to utilize various middleware not only for remote process invocation but also for information services and communication between Ninf-G Client and Servers. Ninf-G5 is appropriate for a single system as well as non-Globus Grid environments. It is expected to provide high performance for task parallel applications from a single system to Grid.

Here are compatibility issues between Ninf-G5 and Ninf-G4. The GridRPC API and Ninf-G API implemented by Ninf-G5 is compatible with Ninf-G4 except the two small issues (details are described in CHANGES file).
- Ninf-G Client configuration file for Ninf-G4 is not compatible with Ninf-G5.
- Due to protocol changes, Ninf-G4 client cannot communicate with
- Ninf-G5 executables and vice versa.

If you have any questions, comments, please send emails to
ninf@apgrid.org or ninf-users@apgrid.org .

http://ninf.apgrid.org/
http://www.apgrid.org/

tag: , , , , , , ,

Sunday, October 28, 2007

What has Betty Cooker and Gridka got in Common, They Both use " ... in box " solution

Creating a cake became much easier in the late 40s when Betty Crocker released cake mix in a box.

Do you ever wish there was an equivalent for computing grids? Now there is, almost.

An approach known as “grid in a box” is making it possible to gather all the ingredients required to make grid computing more affordable and accessible for participating grid centers.

“The idea of ‘grid in a box’ is to put all needed grid services on one piece of hardware,” says Oliver Oberst, Forschungszentrum Karlsruhe. “Instead of having several machines working together to host the infrastructure of a grid site, there are several virtual machines working on one computer—the ‘box.’”

Traditionally, building a grid site with gLite—the middleware designed by Enabling Grids for E-sciencE and used predominately in Europe—required multiple different grid services to be installed, each on a different machine.
Continue reading at International Science Grid.....

Friday, September 07, 2007

3Tera is showcases the AppLogic 2.1 release at the Office 2.0 Conference 2007


SAN FRANCISCO--(BUSINESS WIRE)--3Tera, Inc., the leading innovator of grid computing and utility computing services for web applications, announced today at the Office 2.0 Conference in San Francisco, CA the commercial availability of AppLogic 2.1. The new 2.1 release of the award winning AppLogic grid operating system adds comprehensive Application Monitoring and support for multiple CPUs per appliance. SaaS and Web 2.0 companies can benefit from greater scalability, improved resource utilization, unprecedented visibility and control over application performance.

Utility Computing or Cloud Computing is quickly gaining popularity with online service companies, said Peter Nickolov, president and COO of 3Tera. Our latest version provides unprecedented control of applications and virtual private data center management for production environments, allowing Web 2.0 and SaaS companies to grow and self manage their services using only a browser.

The product we are announcing today has undergone more testing in beta then any previous release of AppLogic, said Bert Armijo, VP of Marketing and Product Development at 3Tera. AppLogic 2.1 allows for greater scalability and control of the infrastructure, assuring users of their ability to grow.

Weve been using the new AppLogic for a month. It had a huge impact on enhancing the manageability of our application, especially for scaling Apache and MySQL, said Joost Schreve, founder and CEO of EveryTrail, Inc., a Web 2.0 startup building an online platform for visualizing travel experiences by mapping and describing geographical locations. Taking advantage of the monitoring capabilities in the new release helped us easily identify elements that needed more resources. We were able to increase the performance and scale our applications easily.

3Tera is showcasing the AppLogic 2.1 release at the Office 2.0 Conference in San Francisco, CA. Office 2.0 is held at the St. Regis Hotel, September 5 7. For more information on Office 2.0, including the conference agenda, visit www.o2con.com/index.jspa.

Wednesday, August 15, 2007

Grid Computing Term may fade?, Long Live Grid

I read an interesting article on computerworld today. I understand what it is trying to say but I don't by the theme of the article. But it is a good article, extensive, I think you should read it too, if you are interested in Grid computing. Let the author know what you think. Link is at the end.
August 14, 2007
(Computerworld) -- Depending on who describes it, grid computing has grown from its roots in high-performance computing into an enterprise technology that provides for shared resources. Or it's an overhyped, meaningless term that will soon disappear in the wake of advances in virtualization and utility computing.

Arguments abound on what constitutes grid computing. But at its core, grid is really an enabling technology that provides on-demand access to computing resources -- including systems, storage and networking -- and data, regardless of location. And because that sounds so similar to how most vendors currently define virtualization, some analysts say the term grid computing may not stick.

The concept behind the technology will likely live on, though, as customers tap into compute power available on underutilized servers, primarily through virtualization. William Fellows, an analyst at The 451 Group and author of a report entitled "Grid Computing: State of the Market" (download PDF), maintains that the term will likely be both more significant and less used in 2007. "Grid computing will be more relevant as grids are used to support far more than high-performance computing tasks, but less used as vendors seek to be associated with far more activity, far higher up the stack, than grid computing."

Grid computing: Term may fade, but features will live on