ExaFlooding RD: A Mathematical Model to Support Unstructured Resource Discovery in Distributed Exascale Computing Environments

Ehsan Mousavi Khaneghah | Zohreh Esmaeili Bidhendi

---

URL :   http://research.shahed.ac.ir/WSR/WebPages/Report/PaperView.aspx?PaperID=159541
Date :  2022/04/22
Publish in :    Journal of Grid Computing
DOI :  https://doi.org/10.1007/s10723-022-09608-z
Link :  http://dx.doi.org/10.1007/s10723-022-09608-z
Keywords :Unstructured Resource Discovery, Distributed Exascale System, Request Attribute Set, Dynamic and Interactive Events, Framework.

Abstract :

---

The unstructured resource discovery has capabilities that can be used as a mechanism for creating a responsive structure and scalability in Distributed Exascale Systems for managing dynamic and interactive events. This paper introduces the ExaFlooding RD framework that can manage and control dynamic and interactive events affecting the functionality of the resource discovery in addition to activities related to unstructured resource discovery. For this purpose, while analyzing the functionality function of the unstructured resource discovery and defining the concept of request in the mentioned element, the concept of request in Distributed Exascale Systems has been redefined. Defining the functionality function of the unstructured resource discovery and redefining the concept of request as the pivotal element of the unstructured resource discovery makes it possible to define the elements of the ExaFlooding RD framework and the functionality of this framework. By using the concept of reviewing the status of executing system activities, the requesting process, and dynamic and interactive events, this framework manages resource discovery activities. Our examination indicates that this framework can manage 60 percent of D&I events without any dependency on the computing system performing resource discovery activities.