• Journal of KIISE
• /
• v.43 no.4
• /
• pp.450-459
• /
• 2016

Every year, RDFS data tends further toward scalability; hence, the manner of SPARQL processing needs to be changed for fast query. The query processing method of SPARQL has been studied using a scalable distributed processing framework. Current studies indicate that the query engine based on the scalable distributed processing framework i.e., Hadoop(MapReduce) is not suitable for real-time processing because of the repetitive tasks; in addition, it is difficult to construct a query engine based on an In-memory Distributed Query engine, because distributed structure on the low-level is required to be considered. In this paper, we proposed a method to construct a query engine for improving the speed of the query process with the mass triple data. The query engine processes the query of SPARQL using the SparkSQL, which is an In-memory based, distributed query processing framework. SparkSQL is a high-level distributed query engine that facilitates existing SQL statement. In order to process the SPARQL query, after generating the Algebra Tree using Jena, the Algebra Tree is required to be translated to Spark Algebra Tree for application in the Spark system, and construction of the system that generated the SparkSQL query. Furthermore, we proposed the design of triple property table based on DataFrame for more efficient query processing in the Spark system. Finally, we verified the validity through comparative evaluation with the query engine, which is the existing distributed processing framework.

• Journal of Platform Technology
• /
• v.11 no.4
• /
• pp.71-79
• /
• 2023

In this paper, we analyze the characteristics of machine learning workloads and, based on them, propose a distributed in-memory caching technique to improve the performance of machine learning workloads. The core of machine learning workload is model training, and model training is a computationally intensive task. Performing machine learning workloads in a Kubernetes-based cloud environment in which the computing framework and storage are separated can effectively allocate resources, but delays can occur because IO must be performed through network communication. In this paper, we propose a distributed in-memory caching technique to improve the performance of machine learning workloads performed in such an environment. In particular, we propose a new method of precaching data required for machine learning workloads into the distributed in-memory cache by considering Kubflow pipelines, a Kubernetes-based machine learning pipeline management tool.

• International Journal of Contents
• /
• v.12 no.1
• /
• pp.60-64
• /
• 2016

As new applications of cloud data management system (CDMS) such as online games, cooperation edit, social network, and so on, are increasing, transaction processing capabilities for CDMS are required. Several transaction processing methods for cloud data management system (CDMS) have been proposed. However, existing transaction processing methods have some problems. Some of them provide limited transaction processing capabilities. Some of them are hard to be integrated with existing CDMSs. In this paper, we proposed a new concurrency control method to support transaction processing capability for CDMS to solve these problems. The proposed method was designed and implemented based on Spark, an in-memory distributed processing framework. It uses RDD (Resilient Distributed Dataset) model to provide fault tolerant to data in the main memory. In our proposed method, database stored in CDMS is loaded to main memory managed by Spark. The loaded data set is then transformed to RDD. In addition, we proposed a multi-version concurrency control method through immutable characteristics of RDD. Finally, we performed experiments to show the feasibility of the proposed method.

• Journal of Electrical Engineering and Technology
• /
• v.6 no.4
• /
• pp.573-579
• /
• 2011

To meet the scalability and performance requirements of data analyses, which often involve voluminous data, efficient parallel or concurrent algorithms and frameworks are essential. We present a high-performance Korean morphological analyzer which employs the MapReduce framework on the graphics processing unit (GPU). MapReduce is a programming framework introduced by Google to aid the development of web search applications on a large number of central processing units (CPUs). GPUs are designed as a special-purpose co-processor. Their programming interfaces are typically formulated for graphics applications. Compared to CPUs, GPUs have greater computation power and memory bandwidth; however, GPUs are more difficult to program because of the design of their architectures. The performance of the Korean morphological analyzer using the MapReduce framework on the GPU is evaluated in comparison with the CPU-based model. The proposed Korean Morphological analyzer shows promising scalable performance on distributed computing with the GPU.

• Journal of KIISE
• /
• v.42 no.9
• /
• pp.1109-1116
• /
• 2015

In this paper, we propose a query processing approach that uses the Spark functional programming and distributed memory system to solve the computational overhead of SPARQL. In the semantic web, RDF ontology data is produced at large scale, and the main challenge for the semantic web is to query and manipulate such a large ontology with a high throughput. The most existing studies on SPARQL have focused on deploying the Hadoop MapReduce framework, and although approaches based on Hadoop MapReduce have shown promising results, they achieve a low level of throughput due to the underlying distributed file processes. Therefore, in order to speed up the query processes, we suggest query- processing methods that are based on memory caching in distributed memory system. Our approach is also integrated with a clause unification method for propagating between the clauses that exploits Spark join, map and filter methods along with caching. In our experiments, we have achieved a high level of performance relative to other approaches. In particular, our performance was nearly similar to that of Sempala, which has been considered to be the fastest query processing system.

• Journal of KIISE
• /
• v.45 no.2
• /
• pp.113-125
• /
• 2018

Recently, with the advent of the Big Data era, we have gained the capability of acquiring vast amounts of knowledge from various fields. The collected knowledge is expressed by well-formed formula and in particular, OWL, a standard language of ontology, is a typical form of well-formed formula. The symbolic reasoning is actively being studied using large amounts of ontology data for extracting intrinsic information. However, most studies of this reasoning support the restricted rule expression based on Description Logic and they have limited applicability to the real world. Moreover, knowledge management for inaccurate information is required, since knowledge inferred from the wrong information will also generate more incorrect information based on the dependencies between the inference rules. Therefore, this paper suggests that the SWAT, knowledge management system should be combined with the SWRL (Semantic Web Rule Language) reasoning based on ATMS (Assumption-based Truth Maintenance System). Moreover, this system was constructed by combining with SWRL reasoning and ATMS for managing large ontology data based on the distributed In-memory framework. Based on this, the ATMS monitoring system allows users to easily detect and correct wrong knowledge. We used the LUBM (Lehigh University Benchmark) dataset for evaluating the suggested method which is managing the knowledge through the retraction of the wrong SWRL inference data on large data.

• Journal of KIISE
• /
• v.45 no.3
• /
• pp.224-233
• /
• 2018

Recently, there are many of studies on SWRL reasoning engine based on user-defined rules in a distributed environment using a large-scale ontology. Unlike the schema based axiom rules, efficient inference orders cannot be defined in SWRL rules. There is also a large volumet of network shuffled data produced by unnecessary iterative processes. To solve these problems, in this study, we propose a method that uses Map-Reduce algorithm and distributed in-memory framework to deduce multiple rules simultaneously and minimizes the volume data shuffling occurring between distributed machines in the cluster. For the experiment, we use WiseKB ontology composed of 200 million triples and 36 user-defined rules. We found that the proposed reasoner makes inferences in 16 minutes and is 2.7 times faster than previous reasoning systems that used LUBM benchmark dataset.

• Journal of KIISE
• /
• v.43 no.9
• /
• pp.963-973
• /
• 2016

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.

• Journal of Advanced Navigation Technology
• /
• v.18 no.4
• /
• pp.401-408
• /
• 2014

Hadoop is composed with MapReduce programming model for distributed processing and HDFS distributed file system. Hadoop is suitable framework for big data processing, but processing of mass small files have many problems. The processing of mass small file in hadoop have problems to created one mapper per one file, and it have problems to needed many memory for store of meta information of file. This paper have comparison evaluation processing method of mass small file with various method in hadoop platform. The processing of general compression format is inadequate because of processing by one mapper regardless of data size. The processing of sequence and hadoop archive file is removed memory problem of namenode by compress and combine of small file. Hadoop archive file is faster then sequence file about combine time of small file. The processing using CombineFileInputFormat class is needed not combine of small file, and it have similar speed big data processing method.

• KIISE Transactions on Computing Practices
• /
• v.24 no.3
• /
• pp.105-112
• /
• 2018

As data size increase, a single database is not enough to serve current volume of tasks. Since data is partitioned and stored into multiple databases, analysis should also support parallelism in order to increase efficiency. However, traditional analysis requires data to be transferred out of database into nodes where analytic service is performed and user is required to know both database and analytic framework. In this paper, we propose an efficient way to perform K-means clustering algorithm inside the distributed column-based database and relational database. We also suggest an efficient way to optimize K-means algorithm within relational database.