 |
||||||
 GF8 and HPDC-12 Report |
||||||
|
||||||
| Overview of GGF8 and HPDC-12 |
||
| GGF8 and HPDC-12 are held June 22-27, 2003 in Seattle, Washington, USA. HPDC-12 is one of the international conferences for presenting the latest research results. HPDC-12 are set to provide two keynote speeches, 5 invited speeches, 25 paper sessions, 19 posters and demonstrations and 4 workshops. GGF is to promote and support the development, deployment, and implementation of Grid technologies and applications from all angles. In GGF8, 45 sessions (21 Working Groups, 14 Research Groups and 10 BoFs) was held. Moreover, it reaches 8 Tutorials and 18 Plenary Programs. It was represented by over 700 researchers in the GGF8. Many persons are interested in standardization of Grid technology. | ||
 |
||
| Motivation of the participation to GGF8 and HPDC-12 |
||
| The participation to GGF8 and HPDC-12 is the opportunity which can know the directivity of the Grid technology and the latest research results. Now, Grid is expected as one of the infrastructures, I think that the participation to GGF is enable to get useful information for research. And I think that the participation to HPDC is enable to get the latest research results by keynote speeches and paper sessiones. | ||
|
||
| Review of the presentation in HPDC-12 |
||
| In this report, I present 3 sessions among 25 paper sessions. | ||
|
||
| 1. Distributed Pagerank for P2P Systems This presentation proposed a fully distributed implementation of Google's highly effective pagerank algorithm for peer to peer systems. To the search engine of Google which is a centralized system, presenters extended so that distributed processing could do pagerank algorithm. A result of simulation for homogeneous environment, presenter evaluates that distributed pagerank can be quickly and low-traffics for keeping the pagerank quality. I am interested in being changed so that pagerank algorithm can be run for distributed computing environments. |
||
|
||
| 2. Squid: Flexible Information Discovery in Decentralized Distributed Systems This presentation described a Peer-to-Peer information discovery system that supports flexible queries using partial keywords and wildcards, and range queries. The information discovery system is called "Squid", it searches the computer corresponding to conditions using Hilbert Space-Filling Curve algorithm. In Grid, the searching of the computer resources requires responsibility and exactness, I think that the searching algorithm applied Squid can be searched resources quickly and exactly. |
||
|
||
| 3. Adaptive Polling of Grid Resource Monitors Using a Slacker Coherence Model This presentation presented that design of resource monitor system using a Slacker coherence model for Grid environments and how to forecast the resource information. Presenter described that forecasting mechanisms of the resource information using lagrange interpolation is better than other forecasting mechanisms. I agree with presenters opinion that it is important to apply the forecasting technique to the surveillance of computer resources. |
||
|
||
| The review of the working group sessions in GGF8 |
||
| In this report, I present 4 working groups in which I participated. | ||
|
||
| Distributed Resource Management Application API (DRMAA) WG The objective of this working group is to facilitate the direct interfacing of application to today's distributed resource management systems by application's builders, portal builders, and independent software venders. In GGF8, this session explained and discussed the draft "Distributed Resource Management Application API Specification 1.0." In future, the working group will discuss the draft "DRMAA Version 2.0". |
||
|
||
| Grid Economic Services Architecture (GESA) WG The purpose of this working group is to define the protocols and service interfaces needed to extensibility support a variety of economic models for the charging of Grid Services in the OGSA. In GGF8, this session discussed the mechanism of Grid Banking System (GBS) and Chargeable Grid Service (CGS). |
||
|
||
| Resource Usage Service (RUS) WG The objective of this working group is to enable the tracking of resource usage within Grid Services deployed within an OGSA environment.In GGF8, this session discussed the mechanism, protocol and operation of resource usage service. I think that the working group has relevance to GESA WG. |
||
|
||
| Network Management (NM) WG The Network Management WG focuses on metrics of interest to grid applications and works in collaboration with, other standards groups. In GGF8, this session followed by the working group's milestones and presented some measurement tools for Grid. And the session discussed some of the UML schema. I'm interested in the measurement tool "MONALISA" which corrected the resource information using the Simple Network Manamement Protocol. |
||
|
||
| Conclusion |
||
| Having participated in GGF8 and HPDC-12, I feel that I can acquire various information. In HPDC-12, I was able to listen to the latest research topics, the opportunity to think my research was obtained. Specially, presented research topics in this report are about resource management, I'm able to confirm the effectiveness of my research. In GGF8, I obtained how to decide the specification for GGF. As for GGF, it is interesting to perform specification decision of Grid which is easy to use for a user. I feel that GGF is trying at construction of the grid environment which is easier to use. In future, I hope that I can present a research topics for myself in HPDC and discuss to decide a specification for Grid in GGF. |
||
|
||
