Editorial

Special Section: Grid and pervasive computing (selected papers from 2010 International Conference on Grid and Pervasive Computing)

Electronic program guide

When you have 500 channels of cable TV, finding a show you like is like finding a needle in a haystack. An electronic program guide (EPG) is a digital guide for scheduling broadcast television or radio programs with functions that allow a user to navigate, select, and discover contents by criteria such as time, title, channel, and genre [1]. However, providing selection is not enough. Providing recommendation is the desired outcome. To recommend programs, the EPG has to know a user’s viewing

Service discovery in the grid environment

Resources in a grid environment are heterogeneous and distributed. How to locate the necessary and qualified resources to meet a task’s requirements efficiently is very important. Resources can provide services. Therefore, service discovery [3] is the automatic detection of devices and services offered by these devices. There are many existing service discovery protocols such as Apple’s Bonjour technology [4], Simple Service Discovery Protocol (SSDP) as used in Universal Plug and Play (UPnP) [5]

Map-reduce task scheduling in grids

Cloud computing [7] has been a very active research topic recently. To process a large set of data in the cloud or in a cluster of computers, Google introduced the MapReduce software framework [8]. Basically there are many map tasks executed simultaneously using different data from a large data set. The results are then merged by the reduce tasks. Grid can be seen as a cluster of computers dispersed around the world. Therefore, applying the MapReduce paradigm in a grid environment, a key step

Reconfigurable processors in grid computing

In a computer system, a processor is either general-purpose or application-specific. For a general-purpose processor, since it is supposed to be able to perform all kinds of tasks, its design cannot be optimized and its efficiency cannot be accelerated. On the other hand, an application-specific processor is designed to execute a certain task very fast. But, for other kinds of tasks, an application-specific processor is powerless. Reconfigurable processors [10] try to rectify the situation.

Workflow scheduling

Scheduling tasks to processors subject to various constraints is an old and interesting problem. Many of these problems are hard to find an optimal solution for. One of them, the workflow scheduling problem [12], [13], further requires that the tasks have workflow relationships. That is, there is precedence between tasks. Some tasks have to be completed before others. There could be one workflow or many workflows. The fifth paper, Online Scheduling of Workflow Applications in Grid Environment,

Service level guarantee in chaotic Internet

To guarantee that an Internet service will be accomplished on time is not easy. Although some applications allow a SLA (Service Level Agreement) [14], [15] to be specified beforehand, the agreement sometimes is not fulfilled because the application cannot control the Internet that it is run on. Among all the agreements, response time is one of the most difficult attributes to control or guarantee, especially now that we are in the cloud computing age. Most, if not all, applications are on the

Conclusions

The six papers in this special section touch on six different topics. They reflect the diversity and the richness of GPC research. We hope that by this special issue the readers can have a better understanding about the breadth and depth of current research. We hope that they can also trigger further related research and technology improvements in various areas of GPC.

Finally, the guest editors would like to express their sincere gratitude to Prof. Peter Sloot, Editor in Chief of Future