Catch up on stories from the past week (and beyond) at the Slashdot story archive

 


Forgot your password?
Close
typodupeerror
×
News

Grid Computing Meets Web Services? 66

Posted by Hemos from the combine-'em-together dept.
jgeelan writes "According to an article in the current issue of Web Services Journal, by ex-IBM, ex-Vitria Technology, ex-IONA middleware maven Dirk Hamstra, the open source initiative known as OGSA, the Open Grid Services Architecture, is poised to bring utility-based computing a step closer. "The combination of Web services and grid computing," Hamstra writes, "virtualizes networked resources using common computing standards, making them accessible to a larger audience." Amazing what a little R&D money from IBM, a prime grid-computing mover, can achieve."
This discussion has been archived. No new comments can be posted.

Grid Computing Meets Web Services? More

Grid Computing Meets Web Services?

Comments Filter:
  • "Open Grid Services Architecture, is poised to bring utility-based computing a step closer"

    And a lot cheaper .. unlike "other" utilities.

  • Finally a relevant Post (Score:1, Funny)

    by Anonymous Coward
    Grid Computing = a Beowulf Cluster.

    • Re: (Score:2, Informative)

      by account_deleted ( 4530225 )
      Comment removed based on user account deletion
      • Grid computing and clustering technologies are on opposite ends of the parallel computing scale.

        Actually the clustering technologies are in the middle of the scale. Symmetric multiprocessing with shared memory is the most tightly coupled end of the scale, then come the clusters, then the Grid technologies at the other end.

        Each calls for a different style of application development too. In systems where IPC is really expensive, you want to minimise it as much as possible. Not all apps that are written to run on a Beowulf cluster will necessarily port straight over to a grid framework. However, for apps that can be made to run well on a grid, the potential computing power available is far, far greater.

        Yes, the development strategies are certainly different. However, often the Grid technologies can be used to provide a way to access the clusters instead of distributing the whole software on several machines. In that case you usually need only relatively small changes to existing software.

        The benefit in this kind of approach can be that the authentication, authorization and encryption services for the connection and data transfer are provided by the Grid framework. For instance you can use the Globus Java CoG kit [globus.org] to authenticate in "Globus style" if you prefer that to the options Java natively offers. (Mobile Analyzer [wikihip.cern.ch] developed in our group at Helsinki Institute of Physics does that.)

        Currently it is often still a bit unconvenient (mappings between Grid credentials and local user accounts etc.) but as these services develop users probably will have access to many more machines than they have now, because they don't need an account in each box. Then they can run their job (which is not necessarily parallel at all) where they like or run the job on their desktop but access data in an external database using their Grid account.

        The computer or cluster for the job can also be selected automatically. The NorduGRID group has implemented this kind of system which connects several clusters in Nordic countries, they have a status monitor on on their website [nordugrid.org].

        AJT
  • So Microsoft's .NET will run on Sony's PS3? Truly the best of both worlds...
  • Now Saddam Hussein can run his nuclear bomb simulations without having to violate US export laws.
  • It's good to see practical web services leveraged through Open Source initiatives. Certainly, this is not unexpected, but it's good to see meacurable results, given the unbelievable type we've been subjected to over the past several years.

    Positive results of OSS efforts are always gradifying to see.

    --CTH

  • Other companies are already working on this ... (Score:4, Informative)

    by DanEsparza ( 208103 ) on Wednesday September 11, 2002 @01:31AM (#4235175) Homepage
    Just wanted to point out that other companies are already working on this concept, too. 'The Mind Electric' [themindelectric.com] has a very nice set of existing Java based web-services tools.

    They are apparently expanding this toolkit to a 'grid service platform' called 'Gaia' detailed here [themindelectric.com].

    From the website:

    GAIA is a service-oriented grid-computing platform that connects producers and consumers of services and data while shielding them from issues like fail over, load balancing and clustering. GAIA can connect and control web services hosted on any combination of platforms, and uses a P2P architecture for reliability and scalability.

    GAIA is based on simple yet powerful concepts, can run on machines ranging from enterprise servers to wireless PDAs, and has native implementations for Java and Microsoft .NET.

  • Why does the grid computing need to meet Web services? Where do we need web services again? I don't see a whole lot of people using them at all. All the hype seems to have died down, there's no migration to web services visible anywhere.
    • Yes, indeed. If web services are the answer, what was the question?

      It's telling how few response there have been to this (now the 3rd on /. so it's not new) story, even taking into account the time of day in the US.

      I can't imagine the thinking behind things like SOAP - "XML is a good markup language - let's use it as an on-the-wire protocol!". Reminds me of the humorous "implementing TCP/IP in XML" article a while back. People shouldn't write these articles, as other people get silly ideas from them!
  • (Here's a good article that briefly describes some more of the benefits that the OGSA initiative provides and how it will eventually enable the cooperation and interaction between all types of distributed computing systems. It's unlikely that the marketplace will remain dominated by only a single distributed computing vendor's (commercial or non-commercial), but at least OGSA will provide a means for equal interoperability and features such as load-balancing between different vendors.)

    ---

    Check out the entire article in whole at:
    http://www.supercomputingonline.com/article.p hp?si d=2236

    SCO: How do Web services play a role in the future of the Grid?

    MANDYAM: The promise of grid computing is to integrate a variety of systems into a virtual supercomputer capable of aggregating resources such as, processor cycles and storage within a large network consisting of one or more organizations. This far-reaching vision experienced a setback, however, by a lack of interoperability standards among grid computing technologies that were being used in the individual organizations. The advent of Web services has revived the vision by allowing grids to be specified as services that can interoperate with each other.

    The interoperability benefits in grid computing come in two areas: one is in the area of tying together heterogeneous resources managed by different grid technologies. The development of Web services-based standards will specify XML-based languages for these technologies to talk to each other. Secondly, the interoperability will extend to existing infrastructures enterprises have in place today. Web services-based interfaces will permit companies to integrate grid computing frameworks more easily into their environments because Web services-based development is designed to be easier and faster than traditional methods. Interfaces described in WSDL will provide the flexibility for companies to build higher-level, Web services-based applications that can also be discovered and shared.

    A grid service is built on concepts and technologies from the Grid and Web services communities, such as the W3C and GGF/OGSA, and will be the basis of Web services influence on grid computing. Its architecture defines standard mechanisms for creating, naming, and discovering grid service instances: providing location transparency, network and platform heterogeneity.

    The Open Grid Services Architecture (OGSA) is in the process of defining a set of Web Services Description Language (WSDL) interfaces and for creating, managing and securely accessing large computational grids.

    SCO: So would you say that it's critical for companies with Grid solutions to plan for Web services in their architecture?

    MANDYAM: Yes, there is unanimous support for a fusion of Web services with Grid services in the grid computing space, and both major and small vendors are embracing the OGSA standards. Fortunately for companies, Web services will make grid solutions easier to implement and more powerful. Vendors will be to make their product architectures compliant.

    Large and progressive enterprises that are planning to deploy enterprise-wide grid solutions should definitely review and follow the standards being specified by the OGSA. In some cases, it may be necessary for enterprises to actively participate in GGF/OGSA to incorporate business critical features into OGSA standards.

    SCO: Does the advent of Web services lead to the potential for a worldwide, Internet grid?

    MANDYAM: As enterprises begin advertising their grid service capabilities using OGSA-based WSDL interfaces, there will be a strong desire among government organizations and private enterprises to build a worldwide Internet grid. However, such a grid would require a lot of cooperation among the various IT departments to develop and implement usage and access control policies. Web services will be secondary to enhancing this potential, as the important milestone will be a general agreement on standards for interoperability. Web services seem to be fueling that through support for the OGSA.

    SCO: On Monday UD announced the availability of the MetaProcessor platform 3.0. Please tell the readers about it.

    MANDYAM: The MetaProcessor platform v3.0 is United Devices' latest version of its computing platform for building grids that harness underutilized compute resources on Windows and Linux. The platform allows companies to effectively incorporate their desktops into enterprise grids capable of delivering a high performance computing engine.

    The platform builds on its recognized ease of manageability and scalability to add important features that make the product easier to integrate and use. It offers a SOAP-based interface and WSDL specification which companies can use to more easily integrate the grid into their infrastructure and to package research applications with Web services. The platform adds the ability to submit and run simple batch jobs that are not as computationally demanding via command line, as opposed to data-intensive high throughput jobs that the platform was originally designed for. Lastly, getting applications onto the platform has also become easier with a simplification of the migration process for application developers. No API's are necessary to access the platform's robust security and network optimization features such as data encryption and compression, and applications can be run without source code modification whatsoever.

  • Real Americans (Score:1, Offtopic)

    by ari_j ( 90255 )
    I just want to say, no matter in how off-topic a way, that I'm proud to see that the first story posted on Slashdot on September 11, 2002, had absolutely nothing to do with last year's terrorist attacks. Folks, we're not terrified, and because of that, we've already won the war, regardless of the battles yet to be fought.

  • Web services? (Score:2, Funny)

    by dpt ( 165990 )
    <german accent>
    I will now announce the following uses for web services in alphabetical order:

    * Making your broadband connection perform like a dial-up (thanks to XML).

    That is all, thank you.
    </german accent>
    • To the anonymous moderator:

      Can you *please* read the fucking moderator guidelines? I was not off-topic, and if you disagree with my *opinion*, you should reply, not moderate. Who knows, I might enjoy a debate, and we all might learn something. Unless you can't hold up your end, of course - but that would explain your preference for mod-ing down, I guess.

      Go on, mod me down some more, I don't care, but it won't detract from the fact that you are an intellectual coward.

  • grid computing sites (Score:3, Informative)

    by jukal ( 523582 ) on Wednesday September 11, 2002 @02:24AM (#4235310) Journal
    Here is [cyberian.org] a little collection of grid computing related companies, organisations and projects. If there is something crucial missing, let me know :)
    • You might consider adding the ConCert project:

      http://www.cs.cmu.edu/~concert/

      We aim to provide a framework for grid computing with certified code (run native code without needing to trust anyone) and strong language support.
    • > If there is something crucial missing, let me know :)

      Missing? How about Avaki, The Legion Project, Sun Grid Engine, and a host of other companies that have working product out there but didn't prescribe to Ian Foster's Globus project as the "One and Only True (Grid) Way".

      It's a real pity that (from what I've seen) the OGSA is anything but open, and seems intent that only one implementation shall ever be called "Grid" and it is Globus.
  • I don't know much about grid computing, but if you need a platform for enabling Web Services, check out the free download in my sig:

    Enterprise Web Services [webmethods.com]

    -brian

    • Advertising on /.?

      Nah, I'll just stick with SOAP::Lite for Perl, as it does SOAP (the full spec, despite the lite tag) and XML-RPC, is free as in speech *and* beer, and will integrate with everything from databases through to HTTP and more esoteric stuff like LDAP with zero effort as it's Perl, and there's a CPAN module for *everything*.

      SOAP's so, well, simple, I don't see how people expect to make money off "web services platforms". At least for Python and Perl, it's done already! Game over, man :)

  • Now this is becoming interesting... Part of my job is administering a group of scientist who run CPU-intensive models. We traditionally do it two different ways:
    • run it on your own PC while you work at other things, hopping that it won't crash for the next 3 to 5 days or
    • buy a cluster or parallel machine (currently expensive SGIs) and let the OS handle the Fortran code and make the best use of CPUs it can.
    The third way is the Seti@home way: write a distributed client and spread it on plenty of machine. Ever since I heard of seti@home I've been wondering why no-one had come with the following way: a distributed client, but distributed only on your own organisation's PCs. Come on, we buy computers to use the CPU ourself, not to run other people's programs. But our PCs are just glorified typewriters while our mainframes are totally saturated. Why not run our stuff on the secretaries 2GHz PCs while they type away ?

    Maybe this should have been an ask.slashdot question a long time ago, but is there a very simple toolkit that will allow one to write a distributed app and put it on whatever PCs you have available ? Requirements are: load balancing, automated distribution and result gathering, ability to have several jobs per machine, access to common resources like a central RAID disk, crash recovery. No need for interprocess communication. Possibility to run any kind of exe would be awesome (yes, we still use fortran, sigh), or at least only link a special library to your prog and that's it.

    I've heard of Jini here on /., but it looked very very basic to say the least and at the time there weren't any examples of code.

  • Good article and associated links!

    As a consultant, I added SOAP, WSDL, and UDDI support to a commercial Java app server product.

    However, for most purposes, there are some very easy to use tools for publishing (if that is the correct word) web services:

    • for Java: Tomcat, Jakarta Axis (for SOAP and WSDL), and the jUDDI open source project (for UDDI)
    • for Python: SOAP.py for client and server SOAP support
    With the cost of dedicated managed servers going down, I think that there are lots of good opportunities for developers to sell web services for applications that require up-to-date data, etc. For, example, a word processor, obviously, using a remote server based application is stupid, but for applications like web search (e.g., Google SOAP support), stock quotes, specialized data processing, etc., it does make good sense to implement as web services.

    -Mark

  • Imagine a Beowulf cluster of these!

Slashdot Top Deals

Your own mileage may vary.

Close