Computing of high breakdown regression estimators without sorting on graphics processing units

We present an approach to computing high-breakdown regression estimators in parallel on graphics processing units (GPU).We show that sorting the residuals is not necessary, and it can be substituted by calculating the median. We present and compare various methods to calculate the median and order statistics on GPUs. We introduce an alternative method based on the optimization of a convex function, and showits numerical superiority when calculating the order statistics of very large arrays on GPUs.

A cluster operating system supporting parallel computing

The single factor limiting the harnessing of the enormous computing power of clusters for parallel computing is the lack of appropriate software. Present cluster operating systems are not built to support parallel computing – they do not provide services to manage parallelism. The cluster operating environments that are used to assist the execution of parallel applications do not provide support for both Message Passing (MP) or Distributed Shared Memory (DSM) paradigms. They are only offered as separate components implemented at the user level as library and independent servers. Due to poor operating systems users must deal with computers of a cluster rather than to see this cluster as a single powerful computer. A Single System Image of the cluster is not offered to users. There is a need for an operating system for clusters. We claim and demonstrate that it is possible to develop a cluster operating system that is
able to efficiently manage parallelism, support Message Passing and DSM and offer the Single System Image. In order to substantiate the claim the first version of a cluster operating system, called GENESIS, that manages parallelism and offers the Single System Image has been developed.

Comparing alternate teaching styles to teach computing skills to girls in their English classes

Low female participation rates in computing are a current concern of the education sector. To address this problem an intervention was developed — computing skills were introduced to girls in their English classes using three different teaching styles: peer tutoring, cross-age tutoring and teacher instruction (control). The sample comprised 136 girls from Years 8 and 10 from a single-sex government school. A pre-test post-test quantitative design was used. To describe the students perspectives, qualitative data were collected from six focus groups conducted with 8–10 students — one from each of the six classes. It was predicted that cross-age tutoring would yield more positive effects than peer tutoring which, in turn, would yield more positive effects than traditional teacher instruction as assessed by achievement on class tasks and attitudes towards computing. The hypothesis was not supported by the quantitative analysis, however in the qualitative data cross-age tutoring was appraised more favourably than peer tutoring or teacher instruction. The latter was the least preferred condition due to: (1) inefficiency; (2) difficulty understanding teachers' explanations; and (3) lack of teacher knowledge. Problems with the implementation of the intervention identified in the focus groups were teacher differences, system failures, missed classes, lack of communication, and selection of computing activities. Practical suggestions were provided relevant to the introduction of cross-age tutoring and the use of computers within secondary level English classes.

Fault detection service architecture for grid computing systems

The ability to tolerate failures while effectively exploiting the grid computing resources in an scalable and transparent manner must be an integral part of grid computing infrastructure. Hence, fault-detection service is a necessary prerequisite to fault tolerance and fault recovery in grid computing. To this end, we present an scalable fault detection service architecture. The proposed fault-detection system provides services that monitors user applications, grid middlewares and the dynamically changing state of a collection of distributed resources. It reports summaries of this information to the appropriate agents on demand or instantaneously in the event of failures.

Dynamic parallel job scheduling in multi-cluster computing systems

Job scheduling is a complex problem, yet it is fundamental to sustaining and improving the performance of parallel processing systems. In this paper, we address an on-line parallel job scheduling problem in heterogeneous multi-cluster computing systems. We propose a new space-sharing scheduling policy and show that it performs substantially better than the conventional policies.

Large-scale biological sequence assembly and alignment by using computing grid

Large-scale sequence assembly and alignment are fundamental parts of biological computing. However, most of the large-scale sequence assembly and alignment require intensive computing power and normally take very long time to complete. To speedup the assembly and alignment process, this paper parallelizes the Euler sequence assembly and pair-wise/multiple sequence assembly, two important sequence assembly methods, and takes advantage of Computing Grid which has a colossal computing capacity to meet the large-scale biological computing demand.

Autonomic job scheduling policy for grid computing

Autonomic middleware services will play an important role in the management of resources and distributed workloads in emerging distributed computing environments. In this paper, we address the problem of autonomic grid resource scheduling and propose a scheduling infrastructure that is capable of self-management in the face of dynamic behavior inherent to this kind of systems.

Job scheduling policy for high throughput grid computing

The growing computational power requirements of grand challenge applications has promoted the need for merging high throughput computing and grid computing principles to harness computational resources distributed across multiple organisations. This paper identifies the issues in resource management and scheduling in the emerging high throughput grid computing context. We also survey and study the performance of several space-sharing and time-sharing opportunistic scheduling policies that have been developed for high throughput computing.

Robust parallel job scheduling infrastructure for service-oriented grid computing systems

Recent trends in grid computing development is moving towards a service-oriented architecture. With the momentum gaining for the service-oriented grid computing systems, the issue of deploying support for integrated scheduling and fault-tolerant approaches becomes paramount importance. To this end, we propose a scalable framework that loosely couples the dynamic job scheduling approach with the hybrid replications approach to schedule jobs efficiently while at the same time providing fault-tolerance. The novelty of the proposed framework is that it uses passive replication approach under high system load and active replication approach under low system loads. The switch between these two replication methods is also done dynamically and transparently.

Grid accounting service infrastructure for service-oriented grid computing systems

In this paper, we propose an architecture of accounting and payment services for service-oriented grid computing systems. The proposed accounting and payment services provide the mechanisms for service providers to be paid for authorized use of their resources. It supports the recording of usage data, secure storage of that data, analysis of that data for purposes of billing and so forth. It allows a variety of payment methods, it is scalable, secure, convenient, and reduce the overall cost of payment processing while taking into account requirements of Grid computing systems.

A novel data replication and management protocol for mobile computing systems

Mobile computing has enabled users to seamlessly access databases even when they are on the move. Mobile computing environments require data management approaches that are able to provide complete and highly available access to shared data at any time from any where. In this paper, we propose a novel replicated data protocol for achieving such goal. The proposed scheme replicates data synchronously over stationary sites based on three dimensional grid structure while objects in mobile sites are asynchronously replicated based on commonly visited sites for each user. This combination allows the proposed protocol to operate with less than full connectivity, to easily adapt to changes in group membership and not require all sites to agree to update data objects at any given time, thus giving the technique flexibility in mobile environments. The proposed replication technique is compared with a baseline replication technique and shown to exhibit high availability, fault tolerance and minimal access times of the data and services, which are very important in an environment with low-quality communication links.

Building autonomic clusters : a response to IBM’s autonomic computing challenge

IBM has announced Autonomic Computing as one of its Grand Challenges. A list of characteristics was generated to describe an autonomic computing system. Many researchers and research group have responded positively to the challenge by initiating research around one or two of the characteristics identified by IBM as the requirements for Autonomic Computing. We have addressed the IBMrsquos request to contribute to the challenge by building an autonomic operating system that possesses nearly all IBM characteristics. This paper contains our response to the IBMrsquos Autonomic Computing Challenge.

The design and development of an integrated system for object-oriented finite element computing

Finite Element Method (FEM) is widely used in Science and Engineering since 1960’s. The vast majority of FEM software is procedure-oriented. However, this conventional style of designing FEM software encounters problems in maintenance, reuse, and expansion of the software. Recently the object-oriented finite element method attracts the attention of lots of researchers, and now there is a growing interest in this method. In this paper, the object-oriented finite element (OOFE) is briefly introduced. Then the design and development of an integrated OOFE system is described. A comparison of the integrated OOFE system and a procedure-oriented system shows that our OOFE system has many advantages.