Scheduling of a parallel computation-bound application and sequential applications executing concurrently on a cluster - a case study

Studies have shown that most of the computers in a non-dedicated cluster are often idle or lightly loaded. The underutilized computers in a non-dedicated cluster can be employed to execute parallel applications. The aim of this study is to learn how concurrent execution of a computation-bound and sequential applications influence their execution performance and cluster utilization. The result of the study has demonstrated that a computation-bound parallel application benefits from load balancing, and at the same time sequential applications suffer only an insignificant slowdown of execution. Overall, the utilization of a non-dedicated cluster is improved.

A computation trust model with trust network in multi-agent systems

Trust is a fundamental issue in multi-agent systems, especially when they are applied in e-commence. The computational models of trust play an important role in determining who and how to interact in open and dynamic environments. To this end, a computation trust model is proposed in which the confidence information based on direct prior interactions with the target agent and the reputation information from trust network are used. In this way, agents can autonomously deal with deception and identify trustworthy parties in multi-agent systems. The ontological property of trust is also considered in the model. A case study is provided to show the effectiveness of the proposed model.

Non-smooth optimization methods for computation of the conditional value-at-risk and portfolio optimization

We examine numerical performance of various methods of calculation of the Conditional Value-at-risk (CVaR), and portfolio optimization with respect to this risk measure. We concentrate on the method proposed by Rockafellar and Uryasev in (Rockafellar, R.T. and Uryasev, S., 2000, Optimization of conditional value-at-risk. Journal of Risk, 2, 21-41), which converts this problem to that of convex optimization. We compare the use of linear programming techniques against a non-smooth optimization method of the discrete gradient, and establish the supremacy of the latter. We show that non-smooth optimization can be used efficiently for large portfolio optimization, and also examine parallel execution of this method on computer clusters.

Scheduling of a parallel computation-bound application and sequential applications executing concurrently on a cluster - a case study.

Studies have shown that most of the computers in a non-dedicated cluster are often idle or lightly loaded. The underutilized computers in a non-dedicated cluster can be employed to execute parallel applications. The aim of this study is to learn how concurrent execution of a computation-bound and sequential applications influence their execution performance and cluster utilization. The result of the study has demonstrated that a computation-bound parallel application benefits from load balancing, and at the same time sequential applications suffer only an insignificant slowdown of execution. Overall, the utilization of a non-dedicated cluster is improved.

Symbolic computation and perfect fluids in general relativity

Focuses on two areas within the field of general relativity. Firstly, the history and implications of the long-standing conjecture that general relativistic, shear-free perfect fluids which obey a barotropic equation of state p = p(w) such that w + p = 0, are either non-expanding or non-rotating. Secondly the application of the computer algebra system Maple to the area of tetrad formalisms in general relativity.

Fast computation of trimmed means

We present two methods of calculating trimmed means without sorting the data in O(n) time. The existing method implemented in major statistical packages relies on sorting, which takes O(n log n) time. The proposed algorithm is based on the quickselect algorithm for calculating order statistics with O(n) expected running time. It is an order of magnitude faster than the existing method for large data sets.

Data-driven computation of contact dynamics during two-point manipulation of deformable objects

This study represents a preliminary step towards data-driven computation of contact dynamics during manipulation of deformable objects at two points of contact. A modeling approach is proposed that characterizes the individual interaction at both points and the mutual effects of the two interactions on each other via a set of parameters. Both global as well as local coordinate systems are tested for encoding the contact mechanics. Artificial neural networks are trained on simulated data to capture the object behavior. A comparison of test data with the output of the trained system reveals a mean squared error percentage between 1% and 3% for simple interactions.

Computation and construction of vault geometry prototypes

Physical models and scaled prototypes of architecture play an important role in design. They enable architects and designers to investigate the formal, functional, and material attributes of the design. Understanding digital processes of realizing scaled prototypes is a significant problem confronting design practice. This paper reports on three approaches to the translation of Gaussian surface models into scaled physical prototype models. Based on the geometry of Eladio Dieste’s Gaussian Vaults, the paper reports on the aspects encountered in the process of digital to physical construction using scaled prototypes. The primary focus of the paper is on computing the design geometry, investigating methods for preparing the geometry for fabrication and physical construction. Three different approaches in the translation from digital to physical models are investigated: rapid prototyping, two-dimensional surface models in paper and structural component models using CNC fabrication. The three approaches identify a body of knowledge in the design and prototyping of Gaussian vaults. Finally the paper discusses the digital to fabrication translation processes with regards to the characteristics, benefits and limitations of the three approaches of prototyping the ruled surface geometry of Gaussian Vaults. The results of each of three fabrication processes allowed for a better understanding of the digital to physical translation process. The use of rapid prototyping permits the production of form models that provide a representation of the physical characteristics such as size, shape and proportion of the Gaussian Vault.

First of all, I split the 24 : strategies for mental computation

In this article, the author sets out some goals and classroom activities for the teaching of mental computation. The author also discusses the importance of allowing children to help each other and explains that there is benefit in children listening to mathematical strategies given by other children.

First of all I split the 24 : strategies for mental computation

Goals for mental computation are presented along with some ideas for teaching mental computation in the classroom. The understanding of number and the operations that come from developing mental computation strategies are useful for algebra and children should be encouraged to use more efficient and diverse mental strategies.

The mechanics of computation and human thought

Thousands of students are preparing for chemistry examinations in June. An unresolved debate is whether they should be permitted to use graphics and programmable calculators in those examinations. Some educators have not only advocated the use of graphics calculators, but have also pointed to the Danish system in which students are permitted to use computers in senior school examinations.

In some Australian jurisdictions, graphics calculators are permitted in year 12 mathematics examinations, but not in chemistry examinations. The reasoning is that information or methods of solving numerical chemical problems can be stored in the memory of graphics calculators, giving some students an unfair advantage. This means that chemistry students either have to learn how to use (and buy!) two types of calculators or, if they only have one calculator, are disadvantaged in using non-programmable calculators in mathematics examinations.

The use of technology (or its lack thereof) can limit how and what students learn. “The mechanics of computation and human thought” is an allusion to Asimov’s short story, “A Feeling of Power” in which, overuse of technology has caused people to forget how to do simple arithmetic. In our current assessment system, the insistence that students must be able to do simple chemical calculations has lead to underuse of available technology. The misperception is that the ability to do calculations is linked to understanding of concepts.

Graphics calculators, programmable calculators and computers are tools. Instead of banning or limiting technology, we should take the opportunity to rethink what is being assessed and how it is assessed. It is the proper use of technology, by combining the mechanics of computation and human thought to deepen understanding and to ask probing questions that truly leads to a feeling of power.

Minimizing bounded uncertainty impact on scheduling with earliest start and due-date constraints via interval computation

Bounded uncertainty is a major challenge to real life scheduling as it increases the risk and cost depending on the objective function. Bounded uncertainty provides limited information about its nature. It provides only the upper and the lower bounds without information in between, in contrast to probability distributions and fuzzymembership functions. Bratley algorithm is usually used for scheduling with the constraints of earliest start and due-date. It is formulated as . The proposed research uses interval computation to minimize the impact of bounded uncertainty of processing times on Bratley’s algorithm. It minimizes the uncertainty of the estimate of the objective function. The proposed concept is to do the calculations on the interval values and approximate the end result instead of approximating each interval then doing numerical calculations. This methodology gives a more certain estimate of the objective function.