1 / 11

State and Events for Web Services: A comparison of five WS-Resource and WS-notification implementations. Questions

State and Events for Web Services: A comparison of five WS-Resource and WS-notification implementations. Questions. Presenter: Jonatan Alava Agnostic: Oscar Valdivia.

metta
Télécharger la présentation

State and Events for Web Services: A comparison of five WS-Resource and WS-notification implementations. Questions

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. State and Events for Web Services: A comparison of five WS-Resource and WS-notification implementations.Questions Presenter: Jonatan Alava Agnostic: Oscar Valdivia

  2. 1. Are these five implementations (GT4-java, GT4-C, pyGridWare, WSRF:Lite and WSRF.Net) the most popular and complete implementations? How Apache WSRF, could be compare to them?

  3. 2. Could we state that WSRF and WSN are in essence the observer design pattern? If not, why? If yes what are the differences?

  4. 3. Any room for improvement in the presented “Generic service hosting environment architecture”? If there is any, what newfeatures will have to be taken in consideration?

  5. 4. The paper compares the five different implementations for the perspectives of architecture, functionality, performance and standards compliance. Could this comparison be improved? (Presentation, add another perspective like usability…) 4.1 Transport and Soap Processing 4.2 Security Issues 4.3 Web Service Dispatch and container 4.4 Persistence 4.5 Finding/Discovering WS-Resources 4.6 Lifetime Management 4.7 Programming and Tooling 4.8 WS Notification 4.9 Authorization 5 Performance Evaluation 6 Interoperability

  6. 5. Could you mention some real-life situations where transport-level security, and SecureConversation, SecureMessage protocols are used. Could you implement the functionality of one by using another? If I’m a perl programmer (WSRF:Lite does not implement SecureMessage or SecureConversation, only transport layer security), could I create all the applications that other implementations can create?

  7. 6. Persistence is used in the paper as “the state of a WS-resource after one invocation should be the same as its state before the next”, so persistence could be implemented by holding the states in memory. So if a system that implements persistence in memory is completely shutdown, it looses its persisted states and we are talking of a problem of “fault-tolerance” (e.g. backups). But if we use the term persistence as the characteristic of data that outlives the execution of the program that created it, we could say that some of these implementations may not have persistence! What is your opinion about this use of terms?

  8. 7. The paper gives a vague description of GT4-C finding/discovering WS-resources features. Could you expand on this explanation? “GT4-C defines interfaces for manipulating resources and provides a default implementation of those interfaces. Alternative implementations can be provided that,for example,do not require that any resource and state information be maintained within the hosting environment.” • What kind of interfaces? • Do they provide a key to find services?

  9. 8. In the lifetime management comparison, the paper cites the various methods to remove resources in the different implementations (deletion tasks, database updates, etc). Therefore giving particular preconditions, could we state what implementation has the best removal mechanism of expired resources? • Is WSRF::Lite the worst when the state is store in a file or database, because it may leave resources that are never discarded? • Are GT4-Java and PyGridWare the best?

  10. 9. Ignoring the programming language preferences is it fair to choose one programming interface’s implementation as the best or the worst? If not, do we have a useful comparison? • Best in term of features:GT4-C? • Best in terms of easy of use:WSRF.Net, pyGridAware? • Worst in terms of easy of use:GT4-C • What can you say about WSRF:Lite?

  11. 10. In the performance comparison with no security GT4-Java and WSRF.net were “comparable”, but are the current settings of heap parameter in the JVM (Xms64MB and Xmx256MB) adequate giving the available memory in the servers? Four identically configured machines: • Dual(2x)AMD Opteron 240 -1.4GHz w/1MB L2 Cache, • 2GB (4x512MB)PC2700 DDR333 Reg. ECC, • 1x Seagate 120GB EIDE 7200 RPM, 8MB cache. • Two machines ran Windows Server 2003 and were used only WSRF.NET. • Two machines ran redhat8.0 (Linux kernel2.6.9-1.667smp) and were used for other implementations. • GT-4 Java testsSun JVM 1.4.2_04-b05 was used. The JVM was started with “-Xms64m -Xmx256M” options • GT4-Cgcc compiler was used with O3 optimizations.

More Related