Vol. 43, No. 9,
Sep. 2016
Digital Library
Priority-based Hint Management Scheme for Improving Page Sharing Opportunity of Virtual Machines
Yeji Nam, Minho Lee, Dongwoo Lee, Young Ik Eom
Most data centers attempt to consolidate servers using virtualization technology to efficiently utilize limited physical resources. Moreover, virtualized systems have commonly adopted contents-based page sharing mechanism for page deduplication among virtual machines (VMs). However, previous page sharing schemes are limited by the inability to effectively manage accumulated hints which mean sharable pages in stack. In this paper, we propose a priority-based hint management scheme to efficiently manage accumulated hints, which are sent from guest to host for improving page sharing opportunity in virtualized systems. Experimental results show that our scheme removes pages with low sharing potential, as compared with the previous schemes, by efficiently managing the accumulated pages.
Transformation Method for a State Machine to Increase Code Coverage
YoungDong Yoon, HyunJae Choi, HeungSeok Chae
Model-based testing is a technique for performing the test by using a model that represents the behavior of the system as a system specification. Industrial domains such as automotive, military/aerospace, medical, railway and nuclear power generation require model-based testing and code coverage-based testing to improve the quality of software. Despite the fact that both model-based testing and code coverage-based testing are required, difficulty in achieving a high coverage using model-based testing caused by the abstraction level difference between the test model and the source code, results in the need for performing model-based testing separately. In this study, to overcome the limitations of the existing model-based testing, we proposed the state machine transformation method to effectively improve the code coverage using the protocol state machine, one of the typical modeling methods is used as the test model in model-based testing, as the test model. In addition, we performed a case study of both systems and analyzed the effectiveness of the proposed method.
Distributed In-Memory based Large Scale RDFS Reasoning and Query Processing Engine for the Population of Temporal/Spatial Information of Media Ontology
Wan-Gon Lee, Nam-Gee Lee, MyungJoong Jeon, Young-Tack Park
Providing a semantic knowledge system using media ontologies requires not only conventional axiom reasoning but also knowledge extension based on various types of reasoning. In particular, spatio-temporal information can be used in a variety of artificial intelligence applications and the importance of spatio-temporal reasoning and expression is continuously increasing. In this paper, we append the LOD data related to the public address system to large-scale media ontologies in order to utilize spatial inference in reasoning. We propose an RDFS/Spatial inference system by utilizing distributed memory-based framework for reasoning about large-scale ontologies annotated with spatial information. In addition, we describe a distributed spatio-temporal SPARQL parallel query processing method designed for large scale ontology data annotated with spatio-temporal information. In order to evaluate the performance of our system, we conducted experiments using LUBM and BSBM data sets for ontology reasoning and query processing benchmark.
Mutagen4J : Effective Mutant Generation Tool for Java Programs
Yiru Jeon, Yunho Kim, Shin Hong, Moonzoo Kim
Mutation analysis (or software mutation analysis) generates variants of a target program by injecting systematic code changes to the target program, and utilizes the variants to analyze the target program behaviors. Effective mutation analyses require adequate mutation operators that generate diverse variants for use in the analysis. However, the current mutation analysis tools for Java programs have limitations, since they support only limited types of mutation operators and do not support recent language features such as Java8. In this study, we present Mutagen4J, a new mutant generation tool for Java programs. Mutagen4J additionally supports mutation operators recently shown to generate various mutants and fully supports recent Java language features. The experimental results show that Mutagen4J generates useful mutants for analyses 2.3 times more than the existing Java mutation tools used for the study.
Automatic Product Review Helpfulness Estimation based on Review Information Types
Many available online product reviews for any given product makes it difficult for a consumer to locate the helpful reviews. The purpose of this study was to investigate automatic helpfulness evaluation of online product reviews according to review information types based on the target of information. The underlying assumption was that consumers find reviews containing specific information related to the product itself or the reliability of reviewers more helpful than peripheral information, such as shipping or customer service. Therefore, each sentence was categorized by given information types, which reduced the semantic space of review sentences. Subsequently, we extracted specific information from sentences by using a topic-based representation of the sentences and a clustering algorithm. Review ranking experiments indicated more effective results than other comparable approaches.
Fault Localization Method by Utilizing Memory Update Information and Memory Partitioning based on Memory Map
Kwanhyo Kim, Ki-Yong Choi, Jung-Won Lee
In recent years, the cost of automotive ECU (Electronic Control Unit) has accounted for more than 30% of total car production cost. However, the complexity of testing and debugging an automotive ECU is increasing because automobile manufacturers outsource automotive ECU production. Therefore, a large amount of cost and time are spent to localize faults during testing an automotive ECU. In order to solve these problems, we propose a fault localization method in memory for developers who run the integration testing of automotive ECU. In this method, memory is partitioned by utilizing memory map, and fault-suspiciousness for each partition is calculated by utilizing memory update information. Then, the fault-suspicious region for partitions is decided based on calculated fault-suspiciousness. The preliminary result indicated that the proposed method reduced the fault-suspicious region to 15.01(%) of memory size.
Feature Expansion based on LDA Word Distribution for Performance Improvement of Informal Document Classification
Hokyung Lee, Seon Yang, Youngjoong Ko
Data such as Twitter, Facebook, and customer reviews belong to the informal document group, whereas, newspapers that have grammar correction step belong to the formal document group. Finding consistent rules or patterns in informal documents is difficult, as compared to formal documents. Hence, there is a need for additional approaches to improve informal document analysis. In this study, we classified Twitter data, a representative informal document, into ten categories. To improve performance, we revised and expanded features based on LDA(Latent Dirichlet allocation) word distribution. Using LDA top-ranked words, the other words were separated or bundled, and the feature set was thus expanded repeatedly. Finally, we conducted document classification with the expanded features. Experimental results indicated that the proposed method improved the micro-averaged F1-score of 7.11%p, as compared to the results before the feature expansion step.
A Test Scenario Generation Method from Activity Diagram with Concurrency
Seungchan Back, Hyorin Choi, Byungjeong Lee, Jung-Won Lee
Currently, software testing is becoming increasingly important in the industrial field and a large body of research supports the improvement of efficient software testing. Model-based testing is generally used to formalize user requirement data for test design. Complex system that includes loop and concurrency has a high probability of path explosion problem. Specially, as threads are added to concurrency, test scenarios have also increased exponentially. However, it is difficult to solve this problem using existing techniques. In this paper, we propose novel path-search technique that focuses on behavioral features of concurrency path in order to avoid path explosion problem. A system that contains concurrent paths is represented by activity diagram in case study section. Efficiency of our study is shown through comparison with several generated test scenarios of other studies. The result indicate that our approach is efficient for finding faults in loop and concurrency with fewer test scenario.
Search Space Reduction by Vertical-Decomposition of a Grid Map
Yewon Jung, Juyoung Lee, Kyeonah Yu
Path-finding on a grid map is a problem generally addressed in the fields of robotics, intelligent agents, and computer games. As technology advances, virtual game worlds tend to be represented more accurately and more realistically, resulting in an excessive increase in the number of grid tiles and in path-search time. In this study, we propose a path-finding algorithm that allows a prompt response to real-time queries by constructing a reduced state space and by precomputing all possible paths in an offline preprocessing stage. In the preprocessing stage, we vertically decompose free space on the grid map, construct a connectivity graph where nodes are the decomposed regions, and store paths between all pairs of nodes in matrix form. In the real-time query stage, we first find the nodes containing the query points and then retrieve the corresponding stored path. The proposed method is simulated for a set of maps that has been used as a benchmark for grid-based path finding. The simulation results show that the state space and the search time decrease significantly.
Fast and All-Purpose Area-Based Imagery Registration Using ConvNets
Together with machine-learning frameworks, area-based imagery registration techniques can be easily applied to diverse types of image pairs without predefined features and feature descriptors. However, feature detectors are often used to quickly identify candidate image patch pairs, limiting the applicability of these registration techniques. In this paper, we propose a ConvNet (Convolutional Network) “Dart“ that provides not only the matching metric between patches, but also information about their distance, which are helpful in reducing the search space of the corresponding patch pairs. In addition, we propose a ConvNet “Fad“ to identify the patches that are difficult for Dart to improve the accuracy of registration. These two networks were successfully implemented using Deep Learning with the help of a number of training instances generated from a few registered image pairs, and were successfully applied to solve a simple image registration problem, suggesting that this line of research is promising.
Tweet Entity Linking Method based on User Similarity for Entity Disambiguation
SeoHyun Kim, YoungDuk Seo, Doo-Kwon Baik
Web based entity linking cannot be applied in tweet entity linking because twitter documents are shorter in comparison to web documents. Therefore, tweet entity linking uses the information of users or groups. However, data sparseness problem is occurred due to the users with the inadequate number of twitter experience data; in addition, a negative impact on the accuracy of the linking result for users is possible when using the information of unrelated groups. To solve the data sparseness problem, we consider three features including the meanings from single tweets, the users’ own tweet set and the sets of other users’ tweets. Furthermore, we improve the performance and the accuracy of the tweet entity linking by assigning a weight to the information of users with a high similarity. Through a comparative experiment using actual twitter data, we verify that the proposed tweet entity linking has higher performance and accuracy than existing methods, and has a correlation with solving the data sparseness problem and improved linking accuracy for use of information of high similarity users.
A Key Management System for Cloud Services Based on Proxy Server Using Self-Creating Algorithm
A key role in cloud computing systems that is becoming an issue is implementing a database on untrusted cloud servers requiring the complexity of key management. This study proposes a key management system using Self Proxy Servers to minimize key executions and improve the performance of cloud services by generating Self-Creating Algorithms where the data owner is not directly concerned with related keys when a user sends an encrypted database a query. The Self Proxy Server supports active and autonomous key managements as a distributed server if any trouble should arise from a cloud key server and for an efficient cloud key management. Therefore, the key management system provides secure cloud services by supporting confidentiality of a cloud server database.
Tile Partitioning-based HEVC Parallel Decoding Optimization for Asymmetric Multicore Processor
Yeongil Ryu, Hyun-Joon Roh, Eun-Seok Ryu
Recently, there is an emerging need for parallel UHD video processing, and the usage of computing systems that have an asymmetric processor such as ARM big.LITTLE is actively increasing. Thus, a new parallel UHD video processing method that is optimized for the asymmetric multicore systems is needed. This paper proposes a novel HEVC tile partitioning method for parallel processing by analyzing the computational power of asymmetric multicores. The proposed method analyzes (1) the computing power of asymmetric multicores and (2) the regression model of computational complexity per video resolution. Finally, the model (3) determines the optimal HEVC tile resolution for each core and partitions/allocates the tiles to suitable cores. The proposed method minimizes the gap in the decoding time between the fastest CPU core and the slowest CPU core. Experimental results with the 4K UHD official test sequences show average 20% improvement in the decoding speedup on the ARM asymmetric multicore system.
Search
Journal of KIISE
- ISSN : 2383-630X(Print)
- ISSN : 2383-6296(Electronic)
- KCI Accredited Journal
Editorial Office
- Tel. +82-2-588-9240
- Fax. +82-2-521-1352
- E-mail. chwoo@kiise.or.kr