Browsing by Author "Jayasena, VSD"

Now showing 1 - 9 of 9

item: Conference-Abstract
Detection of false sharing using machine learning
(2014-06-25) Jayasena, VSD; Amarasinghe, S; Abeyweera, A; Amarasinghe, G; De Silva, H; Rathnayake, S; Meng, X; Liu, Y
False sharing is a major class of performance bugs in parallel applications. Detecting false sharing is difficult as it does not change the program semantics. We introduce an efficient and effective approach for detecting false sharing based on machine learning. We develop a set of mini-programs in which false sharing can be turned on and off. We then run the mini-programs both with and without false sharing, collect a set of hardware performance event counts and use the collected data to train a classifier. We can use the trained classifier to analyze data from arbitrary programs for detection of false sharing. Experiments with the PARSEC and Phoenix benchmarks show that our approach is indeed effective. We detect published false sharing regions in the benchmarks with zero false positives. Our performance penalty is less than 2%. Thus, we believe that this is an effective and practical method for detecting false sharing.
item: Conference-Abstract
Effect of non-work related internet usage on stimulating employee creativity in the software industry
Gunawardena, SIP; Jayasena, VSD
This study investigates the effect of non-work related Internet usage on stimulating employee creativity in the software industry. Drawing from past literature this research proposes six dimensions for measuring creativity stimulation, which include: accessibility to information, intrinsic motivation to execute ideas, curiosity and exploration, independent thinking, collaboration and breaking down technical barriers. A survey was conducted through distribution of a research questionnaire among a stratified random sample of knowledge workers employed in the software industry. The findings of the research were partially consistent with the initial predictions which stated a positive effect of non-work related Internet usage on creativity stimulation. In addition, the research results also provided an exploratory view on the nature of employees’ non-work related Internet usage.
item: Conference-Abstract
Ethical decision making in an information technology context: influences on the bankers in Sri Lanka
Samaranayaka, GPRG; Jayasena, VSD
se of Information Technology in business has revolutionized over the last few decades and more businesses increasingly rely on their computerized information systems to handle their business operations. Studies indicate t at the unethical behaviour of the information workers, who misuse computers and information systems of the organization, has caused serious consequences with billions of losses to the business. In information technology research, much emphasis has recently been given to understand the ethical behaviour of individuals and the factors that could influence such behaviour. This research study aimed to recognize specific factors that might influence the ethical/unethical behaviour intention of the bankers of the Licensed Commercial Banks (LCBs) in Sri Lanka, when they are faced with ethical dilemmas while using IT. A questionnaire instrument primarily based on ethical scenarios in IT was administered among 307 bankers of leading LCBs in Sri Lanka, using simple random sampling. Data was analyzed using univariate, bivariate and multivariate analyses. According to the observations across all scenarios, banker's Attitude and the Personal Normative Beliefs significantly explained the banker s intention to behave ethically/unethically. It is important to note that the significance of the factors that influence the behaviour intention varied along with the ethical scenario faced bv the individual as well as the perceived importance towards the particular ethical scenario in questioni Meaningful code of ethics, ethics training with the use recognized as the recommendations for the bank managements in order to curb the undesired misuse of IT. This research concluded that in addition to the surveillance and control enforced in the LCBs. deterrent should also be adopted to encourage and promote ethical behaviour among the bankers in Sri Lanka. of fictitious ethical scenarios and effective reinforcement mechanisms were measures should also be adopted to encourage and promote ethical behaviour among the bankers in Sri Lanka.
item: Article-Abstract
Facilities offered by the banks and expectations of IT savvy banking customers
Vivekanandan, L; Jayasena, VSD
The banking industry across the globe has witnessed a revolution in the last two decades in its efforts to meet the ever growing customer demands. The advancement of technology has encouraged many organisations to move across markets to position themselves as multinationals and the Sri Lankan banking sector is no different. Continuous analysis of customer demands has driven the banks to rely heavily on technology thereby crowding the markets with many products and services.Sri Lankan banks today are continuing to introduce innovative electronic banking solutions to their customers. The question however is: How much of these electronic banking solutions are actually used by the customers? This study was conducted to identify the consumer response on facilities offered by the bank. A survey was conducted and data collected from 404 banking customers in the Colombo district were analysed. Results of the survey reveal that the topmost facilities demanded from the bank are accurate and secure services. The ATM is ranked as the third and the usage is 80%. The research concludes that customers expect the bank to provide a high level of service more than the electronic-based solutions per se.
item: Conference-Abstract
Fast computation of stock market indices using GPUs
(2014-06-23) Induwara, S; Jayasena, VSD
The stock market index is a valuable financial tool to measure the state of a segment of the stock market. With high input data rates, real-time index computation is a challenging task. It cannot be done in real-time with today's reasonably high-end computers with many CPU cores. Doing so with CPU-based systems will require server farms with lots of computing power and therefore costly. Thus currently this index value is computed periodically (non-real-time). In this paper we describe our attempt in fast index computation using Graphics Processing Units (GPUs), which usually have several hundreds of processing cores and are much less expensive than CPU-based solutions. The computation itself is data parallel and therefore suitable for GPU processing. Preliminary results indicate our approach is promising as we can compute much faster using GPUs than using multi-core CPUs.
item: Conference-Abstract
Identification and characterization of crawlers through analysis of web logs
(2014-06-18) Algiriyage, N; Jayasena, VSD; Dias, G; Perera, A; Dayananda, K; Sharma, K
Web crawlers are software programs that automatically traverse the hyperlink structure of the world-wide web in order to locate and retrieve information. In addition to crawlers from search engines, we observed many other crawlers which may gather business intelligence, confidential information or even execute attacks based on gathered information while camouflaging their identity. Therefore, it is important for a website owner to know who has crawled his site, and what they have done. In this study we have analyzed crawler patterns in web server logs, developed a methodology to identify crawlers and classified them into three categories. To evaluate our methodology we used seven test crawler scenarios. We found that approximately 53.25% of web crawler sessions were from 'known' crawlers and 34.16% exhibit suspicious behavior.
item: Conference-Abstract
Identification and characterization of crawlers through analysis of web logs
Algiriyage, N; Jayasena, VSD; Dias, G; Perera, A; Dayananda, K
Web crawlers are software programs that automatically traverse the hyperlink structure of the world-wide web in order to locate and retrieve information. In addition to crawlers from search engines, we observed many other crawlers which may gather business intelligence, confidential information or even execute attacks based on gathered information while camouflaging their identity. Therefore, it is important for a website owner to know who has crawled his site, and what they have done. In this study we have analyzed crawler patterns in web server logs, developed a methodology to identify crawlers and classified them into three categories. To evaluate our methodology we used seven test crawler scenarios. We found that approximately 53.25% of web crawler sessions were from “known” crawlers and 34.16% exhibit suspicious behavior.
item: Conference-Extended-Abstract
Sport -event management framework (SEMF)
(2007) Anuruddha, KD; Damayanthi, RGI; Gamage, DCGU; Kariyawasam, KINC; Jayasena, VSD; De Silva, T
Managing a sport event is a complex task. The requirement of a software system for managing such events has significantly increased in the recent years. Sport-event management software has short life time of operation. Most of the time, the software development life cycle is also short. Thus building a high quality and reliable system is difficult. Due to these reasons, such software can be quite expensive.
item: Conference-Full-text
A visualization and analysis platform for performance tuning
(IEEE, 2016-04) Eranjith, HMD; Fernando, ID; Fernando, GKS; Soysa, WCM; Jayasena, VSD; Jayasekara, AGBP; Bandara, HMND; Amarasinghe, YWR
With a framework like OpenTuner, one could build domain-specific multi-objective program auto-tuners and gain significant performance improvements. But explaining why and interpreting the results are often hard, mainly due to the large number of parameters and the inability to figure out how each parameter affects the performance improvement. We have a solution that can explain the performance improvements by identifying key parameters while providing better insights on the tuning process. Our tool uses machine learning techniques to identify parameters which account for a significant performance improvement. A user could utilize different methods provided in the tool to further experiment and verify the accuracy of such findings. Further, our tool uses multidimensional scaling to display all the configurations in a two dimensional graph. This interface allows users to analyze the search space closely and identify clusters of configurations with good or bad performance. It also provides real-time information of tuning process which would help users to optimize the tuning process.