Some Observations on Reuse Types, Technologies, Practices and Problems
A.K. Tripathi* and Manjari Gupta*
Department of Computer Engineering, Institute of Technology, Banaras Hindu University, Varanasi
Reuse promises increased productivity and enhanced product quality . A good deal of research in this area has gone into identifying types of possible reuse types and technologies for the purpose . The experience of the industry shows that organizations are not comfortable in practicing reuse . This paper attempts to identify the reasons (technical) impeding reuse oriented software development. The paper starts with giving introductory remarks about reuse itself, its types etc . for describing the reuse scenario and promises . The research works reported in the literature normally talk about organizational issues whereas this paper considers the technical issues to present the clearer picture . Finally we give our observations and propositions for making reuse practice a regular part of the development process . The basic observation put forth by us is the idea that reuses needs to be inculcated in the development process ( merely encouraging people to practice reuse does not suffice ) . The time now has come to propose “Reuse Oriented Software Process” concept .
Computational Study on Malarial Parasites
M. Jain* , G. C. Sharma* , Sudheer Kumar Sharma* and S. C. Kushwaha*
* Department of Mathematics, I.B.S. Khandari, Agra
The present paper is focused on the dynamics of malaria transmission by considering the aspects of non-random host choice, which may affect quantitative and qualitative aspects of the interaction of human hosts and mosquito vectors . The study incorporates biting preference by mosquitoes with respect to infection in their host , which is evident from a few laboratory experiments . In determining levels of malarial infection, it is showed that mosquitoes are preferentially attracted to infected host even when biting does not occur. The increasing - preference and switching models have been developed to understand the equilibrium and stability .
Cache Disk Drive for Linux
Dr. (Mrs.) Pushpa R.Suri and (Ms) Pooja Aggarwal
* Department of Computer Science & Applications, Kurukshetra University, Kurukshetra, Haryana,(Ind)
A process of using some portion of RAM to be used as Cache memory for improving the I/O performance of the system is presented in this paper. This additional cache helps in reducing the number of disk accesses especially for I / O incentive applications and hence increases the Throughput thereby decreasing the Response time . The objective is to increase the I/O performance at least by 10% . Whole of the coding is done in ‘C' and the product runs successfully on 80486 or higher processor with memory 16 MB or more . This product is meant for Linux Red Hat 6.2 having kernel version 2.2.12-20 and supports all kind of data base applications and kind of file systems supported by Linux .
Traffic Characterization and Prioritized Channel Schemes for Wireless in Local Loop (WLL)
Madhu Jain* , Vidushi Sharma** & Kriti Priya***
* Depertment of Mathematics, Institute of Basic Science, Dr. B. R. Ahmedkar University, Agra 282002, India
** Institute of Technology and Science, Ghaziabad – 201005 (India)
*** Graduate School of Business and Administration, Greater Noida (India)
The paper focuses on the performance prediction of wireless local loop system in terms of blocking probabilities of the calls. We develop a traffic model based on queue theoretic approach. For solution purpose, product type solution is employed. Analysis is brought about by comparing various situations of prioritizing the handoff calls by reserving the channel, allowing new calls to be entertained in this reserved area if the handoff calls are less and further offering facilities to handle more handoff voice calls by subrating the reserved channel. All of the above situations are computed analytically as well as using Artificial Neural Network (ANNs), which is the fastest growing technique of soft computing. A feed forward two layer ANN is considered for obtaining the blocking probabilities. The back propagation algorithm is used for training the ANN. All the above situations are compared using numerical illustrations .
Realization Of Fast Switching, Low Power And Less Space Consuming Logic Circuits Using Single Electron Devices – A Case Study
Prasanna Kumar Sahu , Anup Kumar Biswas , Subir Kumar Sarkar
Single electron tunneling circuits hold out the promise of being the successor to present day technologies. Single electron devices are the attractive candidates for the post CMOS VLSI and hybrid ICs due to their unique principle of operation, quantized nature of carrier transport, ultra small size and low power dissipation. All these properties can be incorporated to realize fast switching, low power and less space consuming logic circuits. In the present work, multiplier circuits are proposed to realize using single electron circuits. Two different schemes: combinational logic based repetitive addition method and sequential logic dominated Booth's technique have been proposed. Each category of the multiplier circuit is proposed to realize separately with both the conventional logic circuit and binary decision diagram based single electron logic circuits substantiating the present schemes. It is found that the execution time of a particular multiplication by single electron device based circuit is significantly lesser compared to conventional logic based circuit. The single electron device based circuit is approximately three times faster that its conventional logic version (for repetitive addition scheme) whereas the same is approximately four time faster compared to conventional logic based Booth's technique.
Measuring the Maintainabilityof Object Oriented Systems
Malviya , A. K.* and Dutta , M**
* M.G. Kashi Vidyapith, Varanasi
**Agra College, Agra
This paper aims to reanalyze the data (Li and Henry,1993) in an effort to find early indicator of predictability of maintainability of Object Oriented Systems. For this purpose, the maintainability of the system was analyzed at the (a) analysis/design level, and (b) implementation levels by employing multiple linear regression and MLP neural network model. Results revealed that the employed models while providing comparability and explanatory bases of the observed major trends in tune with the models, provided early indices for predictability of maintainability of object oriented system .
Controllable Multi Server Queue With Balking and Additional Servers
Madhu Jain* and Pankaj Sharma*
* Department of Mathematics, Institute of Basic Science, Khandari, Agra, India
In this paper, we consider multi server controllable queueing model with balking. To reduce the balking behaviour of the customers, there is provision of additional removable servers. These additional servers are facilitated / removed at a pre specified threshold levels of the queue size. The inter-arrival and service times are interdependent and follow exponential distributions. Chapman Kolomogrove equations governing the model in the steady state are constructed by equating the in-flow rate to out-flow rate for each state. Queue size distribution is obtained by employing recursive approach. Various steady-state characteristics namely average number of customers in the system, expected waiting time, etc. are derived by using queue size distribution .
Three D imension al Reengineering
Vinayak Srivastava*
* Department of Computer Engineering, Institute of Technology, Banaras Hindu University, Varanasi, India
Traditional reengineering applies reverse engineering to existing system code to extract design and requirements. Forward engineering is then used to develop the replacement system. Due to limited resources, many organizations, NASA included, are looking at the use of COTS software packages as a means of decreasing development time and costs. This paper briefly describes traditional reengineering, then describes the earlier work in this area i.e. hybrid re-engineering and then discusses the process of three dimensional reengineering which covers the aspect of reuse in software reengineering projects as mentioned in the earlier work of hybrid reengineering. Three-dimensional reengineering uses a combination of reverse engineering, reuse and forward engineering to replace the existing system. This paper discusses the advantages, potential risks and metrics for a three dimensional mode of software reengineering. Again the reverse engineering is not only to find hidden facts; but also to extract the reusable components.
Hypertext Quality Measurement
P. K. Suri* , Rakash Kumar* and R. K. Chauhan*
* Kurukshetra University, Kurukshetra
Hypertext has given a new dimension to the Information Retrieval (IR) science. The navigation in and retrieval from hypertext is quiet easy due to their flexible structure. But this flexibility is the main cause of ‘Lost in hyperspace problem'. The main source of this problem is superfluous cross-reference links. In the first part of this paper, there is a brief discussion of lost in hyperspace problem, techniques to tackle this problem and significance of software metrics. In the 2 nd part of this paper, a hypertext quality matric LCD (Limit of Converted in Distance) is developed which is an indicator of limit of connectedness in hypertext .
|
