Abstract: Fair share objective has been included into the goaloriented
parallel computer job scheduling policy recently. However,
the previous work only presented the overall scheduling performance.
Thus, the per-user performance of the policy is still lacking. In this
work, the details of per-user fair share performance under the
Tradeoff-fs(Tx:avgX) policy will be further evaluated. A basic fair
share priority backfill policy namely RelShare(1d) is also studied.
The performance of all policies is collected using an event-driven
simulator with three real job traces as input. The experimental results
show that the high demand users are usually benefited under most
policies because their jobs are large or they have a lot of jobs. In the
large job case, one job executed may result in over-share during that
period. In the other case, the jobs may be backfilled for
performances. However, the users with a mixture of jobs may suffer
because if the smaller jobs are executing the priority of the remaining
jobs from the same user will be lower. Further analysis does not show
any significant impact of users with a lot of jobs or users with a large
runtime approximation error.
Abstract: The NGN (Next Generation Network), which can
provide advanced multimedia services over an all-IP based network, has been the subject of much attention for years. While there have
been tremendous efforts to develop its architecture and protocols, especially for IMS, which is a key technology of the NGN, it is far
from being widely deployed. However, efforts to create an advanced
signaling infrastructure realizing many requirements have resulted in a
large number of functional components and interactions between those
components. Thus, the carriers are trying to explore effective ways to
deploy IMS while offering value-added services. As one such
approach, we have proposed a self-organizing IMS. A self-organizing
IMS enables IMS functional components and corresponding physical
nodes to adapt dynamically and automatically based on situation such
as network load and available system resources while continuing IMS
operation. To realize this, service continuity for users is an important
requirement when a reconfiguration occurs during operation. In this
paper, we propose a mechanism that will provide service continuity to
users and focus on the implementation and describe performance
evaluation in terms of number of control signaling and processing time
during reconfiguration
Abstract: In this study a clustering technique has been implemented which is K-Means like with hierarchical initial set (HKM). The goal of this study is to prove that clustering document sets do enhancement precision on information retrieval systems, since it was proved by Bellot & El-Beze on French language. A comparison is made between the traditional information retrieval system and the clustered one. Also the effect of increasing number of clusters on precision is studied. The indexing technique is Term Frequency * Inverse Document Frequency (TF * IDF). It has been found that the effect of Hierarchical K-Means Like clustering (HKM) with 3 clusters over 242 Arabic abstract documents from the Saudi Arabian National Computer Conference has significant results compared with traditional information retrieval system without clustering. Additionally it has been found that it is not necessary to increase the number of clusters to improve precision more.
Abstract: This paper proposes an innovative approach for the Connection Admission Control (CAC) problem. Starting from an abstract network modelling, the CAC problem is formulated in a technology independent fashion allowing the proposed concepts to be applied to any wireless and wired domain. The proposed CAC is decoupled from the other Resource Management procedures, but cooperates with them in order to guarantee the desired QoS requirements. Moreover, it is based on suitable performance measurements which, by using proper predictors, allow to forecast the domain dynamics in the next future. Finally, the proposed CAC control scheme is based on a feedback loop aiming at maximizing a suitable performance index accounting for the domain throughput, whilst respecting a set of constraints accounting for the QoS requirements.