• Annual Conference of KIPS
• /
• 2012.11a
• /
• pp.268-270
• /
• 2012

DRAM 가격의 하락으로 인메모리 컴퓨팅에 대한 연구 및 개발이 다시 활발해지고 있으나 효율적인 메모리 시스템 구성을 위한 연구는 아직 부족한 실정이다. 이에 본 논문은 64 비트 멀티프로세서와 대용량의 메모리로 구성되는 인메모리 컴퓨팅 시스템을 모델링하고, 메모리 크기 및 채널 개수에 따른 시스템의 성능을 시뮬레이션 하였다. 그리고 처리된 트랜잭션의 수를 성능평가의 기준으로 하여 메모리의 크기와 채널 개수에 따른 비용을 고려한 최적의 인메모리 컴퓨팅 메모리 시스템 구조를 제안하였다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2019.07a
• /
• pp.325-326
• /
• 2019

본 논문은 퍼스널 컴퓨팅 환경의 성능 향상을 위한 인피니밴드 네트워크 기반 인메모리 스토리지 시스템의 구조를 제안한다. 성능평가를 위해 100Gbit/s을 지원하는 MCX455A-ECAT 한 쌍을 MCP1600-E02A 케이블로 직결한 x86-64 architecture의 인피니밴드 네트워크를 구성하고 Target 시스템에 iSCSI Extensions for RDMA(iSER)을 적용한 RAM disk를 생성하였다. CentOS virt-manager에서 생성한 Initiator 시스템의 Windows 가상 머신에는 Target 시스템의 RAM disk를 VirtIO 방식으로 연결한다. 이 구조는 시스템 종료 후 초기화되는 종래 RAM disk의 일반적 특성을 개선한다. 마지막으로 스토리지 성능평가를 통해 향후 출시될 PCI Express 4.0 이상의 시스템과 퍼스널 컴퓨팅 스토리지 성능 향상 측면에서 해당 구조의 적합성을 보인다.

• Journal of the Korean Data and Information Science Society
• /
• v.26 no.5
• /
• pp.1035-1045
• /
• 2015

Recent studies in Big Data Analysis are showing promising results, utilizing the main memory for rapid data processing. In-memory computing technology can be highly advantageous when used with high-performing servers having tens of gigabytes of RAM with multi-core processors. The constraint in network in these infrastructure can be lessen by combining in-memory technology with distributed parallel processing. This paper discusses the research in the aforementioned concept applying to a test taxi hailing application without disregard to its underlying RDBMS structure. The application of IMDG technology in the application's backend API without restructuring the database schema yields 6 to 9 times increase in performance in data processing and throughput. Specifically, the change in throughput is very small even with increase in data load processing.

• KIISE Transactions on Computing Practices
• /
• v.23 no.8
• /
• pp.466-472
• /
• 2017

Hardware Transactional Memory (HTM) has greatly changed the parallel programming paradigm for transaction processing. Since Intel has recently proposed Transactional Synchronization Extension (TSX), a number of studies based on HTM have been conducted. However, the existing studies support conflict prediction for a single cause of the transaction processing and provide a standardized TSX environment for all workloads. To solve the problems, we propose an efficient hardware transactional memory scheme for processing transactions in multi-core in-memory environment. First, the proposed scheme determines whether to use Software Transactional Memory (STM) or the serial execution as a fallback path of HTM by using a prediction matrix to collect the information of previously executed transactions. Second, the proposed scheme performs efficient transaction processing according to the characteristic of a given workload by providing a retry policy based on machine learning algorithms. Finally, through the experimental performance evaluation using Stanford transactional applications for multi-processing (STAMP), the proposed scheme shows 10~20% better performance than the existing schemes.

• Journal of KIISE
• /
• v.42 no.8
• /
• pp.998-1009
• /
• 2015

In recent years, there has been a growing interest in RDFS Inference to build a rich knowledge base. However, it is difficult to improve the inference performance with large data by using a single machine. Therefore, researchers are investigating the development of a RDFS inference engine for a distributed computing environment. However, the existing inference engines cannot process data in real-time, are difficult to implement, and are vulnerable to repetitive tasks. In order to overcome these problems, we propose a method to construct an in-memory distributed inference engine that uses a parallel graph structure. In general, the ontology based on a triple structure possesses a graph structure. Thus, it is intuitive to design a graph structure-based inference engine. Moreover, the RDFS inference rule can be implemented by utilizing the operator of the graph structure, and we can thus design the inference engine according to the graph structure, and not the structure of the data table. In this study, we evaluate the proposed inference engine by using the LUBM1000 and LUBM3000 data to test the speed of the inference. The results of our experiment indicate that the proposed in-memory distributed inference engine achieved a performance of about 10 times faster than an in-storage inference engine.

• The Korean Journal of Applied Statistics
• /
• v.29 no.6
• /
• pp.1077-1094
• /
• 2016

Apache Spark is a fast and general-purpose cluster computing package. It provides a new abstraction named resilient distributed dataset, which is capable of support for fault tolerance while keeping data in memory. This type of abstraction results in a significant speedup compared to legacy large-scale data framework, MapReduce. In particular, Spark framework is suitable for iterative machine learning applications such as logistic regression and K-means clustering, and interactive data querying. Spark also supports high level libraries for various applications such as machine learning, streaming data processing, database querying and graph data mining thanks to its versatility. In this work, we introduce the concept and programming model of Spark as well as show some implementations of simple statistical computing applications. We also review the machine learning package MLlib, and the R language interface SparkR.

• KIISE Transactions on Computing Practices
• /
• v.24 no.1
• /
• pp.24-32
• /
• 2018

This paper proposes a Big Data system for energy Big Data which is aggregated in real-time from industrial and public sources. The constructed Big Data system is based on Hadoop and the Spark framework is simultaneously applied on Big Data processing, which supports in-memory distributed computing. In the paper, we focus on Big Data, in the form of heat energy for district heating, and deal with methodologies for storing, managing, processing and analyzing aggregated Big Data in real-time while considering properties of energy input and output. At present, the Big Data influx is stored and managed in accordance with the designed relational database schema inside the system and the stored Big Data is processed and analyzed as to set objectives. The paper exemplifies a number of heat demand plants, concerned with district heating, as industrial sources of heat energy Big Data gathered in real-time as well as the proposed system.

• Journal of KIISE
• /
• v.44 no.10
• /
• pp.1005-1018
• /
• 2017

Recently, virtualization has become an essential component of cloud computing due to its various strengths, including maximizing server resource utilization, easy-to-maintain software, and enhanced data protection. However, since virtualization allows sharing physical resources among the VMs, the system performance can be deteriorated due to device contentions. In this paper, we first investigate the I/O overhead based on the number of VMs on the same server platform and analyze the block I/O process of the KVM hypervisor. We also propose an in-memory block cache mechanism, called QBic, to overcome I/O virtualization latency. QBic is capable of monitoring the block I/O process of the hypervisor and stores the data with a high access frequency in the cache. As a result, QBic provides a fast response for VMs and reduces the I/O contention to physical devices. Finally, we present a performance measurement of QBic to verify its effectiveness.

• The Journal of Korean Institute of Next Generation Computing
• /
• v.13 no.5
• /
• pp.7-18
• /
• 2017

Recently, the number of SNS user has rapidly increased due to smart device industry development and also the amount of generated data is exponentially increasing. In the twitter, Text data generated by user is a key issue to research because it involves events, accidents, reputations of products, and brand images. Twitter has become a channel for users to receive and exchange information. An important characteristic of Twitter is its realtime. Earthquakes, floods and suicides event among the various events should be analyzed rapidly for immediately applying to events. It is necessary to collect tweets related to the event in order to analyze the events. But it is difficult to find all tweets related to the event using normal keywords. In order to solve such a mentioned above, this paper proposes A Generation and Matching Method of Normal-Transient Dictionary for realtime topic detection. Normal dictionaries consist of general keywords(event: suicide-death-loop, death, die, hang oneself, etc) related to events. Whereas transient dictionaries consist of transient keywords(event: suicide-names and information of celebrities, information of social issues) related to events. Experimental results show that matching method using two dictionary finds more tweets related to the event than a simple keyword search.