• The Journal of Society for e-Business Studies
• /
• v.11 no.1
• /
• pp.1-23
• /
• 2006

As data on the Web is more and more in XML due to proliferation of Web applications such as e-Commerce, it is strongly required to rapidly process XML queries. One of such techniques is XML query caching. For frequently submitted queries, their results could be cached in order to guarantee fast response for the same queries. In this paper, we propose techniques for XML query performance improvement whereby the set of node identifiers(NIS) for an XML query is cached. NIS is most commonly employed as a format of XML query result,, consisting of the identifiers of the XML elements that comprise the query result. With NIS, it is suitable to meet the Web applications data retrieval requirements because reconstruction and/or modification of query results and integration of multiple query results can be efficiently done. Incremental refresh of NIS against its source updates can also be efficiently done. When the query result is requested in XML, however, materialization of NIS is needed by retrieving the source XML elements through their identifiers. In this paper, we consider three different types of NISs. proposing the algorithms of their creation, materialization, and incremental refresh. All of them were implemented using an RDBMS. Through a detailed set of performance experiments, we showed the efficiency of the proposed XML query caching techniques.

• Smart Media Journal
• /
• v.4 no.4
• /
• pp.120-129
• /
• 2015

Explosion effect on CG (Computer Graphic) a visual effect on which a higher degree of technological difficulty is required with a variety of effect elements such as Fire, Smoke, Flame, Dust, Debris, etc. integrated on it. As skills for CG software have been advanced, solutions loaded with functions of various fluid simulation have been developed. So more realistic special effects came to be available. However, in Korea, it depends just on CG program functions. Besides, enough R&D's concerned have not been followed up. Accordingly, this study is aimed at offering a production method that may effectively implement more realistic explosion effects under experimentations. To begin with, the study derives problems through a precedent study of the implementation of existing explosion effects. Then to solve them, experimental studies are performed depending on four solutions. There are accesses to the four solutions: first, Numerous Turbulent Flow, a method to allow an attribute of turbulent air in the stage of fluid simulation; second, Cache Retiming Solution produced in script; third, Multiple Volume Container based on cached data; and fourth, RGB Lighting Pipeline, a method to enhance the completion of the result from the stage of composition. Characteristics of effects applied in each stage and consecutive connections of them proved the effective implementation of more realistic explosion effects. This study may not only suppose the production method for efficient explosion effects differentiated from the previous ones but also be utilized as basic data for relevant researches.

• Journal of IKEEE
• /
• v.25 no.4
• /
• pp.664-671
• /
• 2021

This paper proposes a GPGPU structure that can reduce the number of operations and memory access by effectively applying a data reuse method to a convolutional neural network(CNN). Convolution is a two-dimensional operation using kernel and input data, and the operation is performed by sliding the kernel. In this case, a reuse method using an internal register is proposed instead of loading kernel from a cache memory until the convolution operation is completed. The serial operation method was applied to the convolution to increase the effect of data reuse by using the principle of GPGPU in which instructions are executed by the SIMT method. In this paper, for register-based data reuse, the kernel was fixed at 4×4 and GPGPU was designed considering the warp size and register bank to effectively support it. To verify the performance of the designed GPGPU on the CNN, we implemented it as an FPGA and then ran LeNet and measured the performance on AlexNet by comparison using TensorFlow. As a result of the measurement, 1-iteration learning speed based on AlexNet is 0.468sec and the inference speed is 0.135sec.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.11
• /
• pp.4268-4290
• /
• 2015

Future mobile communications face enormous challenges as traditional voice services are replaced with increasing mobile multimedia and data services. To address the vast data traffic volume and the requirement of user Quality of Experience (QoE) in the next generation mobile networks, it is imperative to develop efficient content distribution technique, aiming at significantly reducing redundant data transmissions and improving content delivery performance. On the other hand, in recent years cloud computing as a promising new content-centric paradigm is exploited to fulfil the multimedia requirements by provisioning data and computing resources on demand. In this paper, we propose a cooperative caching framework which implements State based Content Distribution (SCD) algorithm for future mobile networks. In our proposed framework, cloud service providers deploy a plurality of cloudlets in the network forming a Distributed Cloud Service Network (DCSN), and pre-allocate content services in local cloudlets to avoid redundant content transmissions. We use content popularity and content state which is determined by content requests, editorial updates and new arrivals to formulate a content distribution optimization model. Data contents are deployed in local cloudlets according to the optimal solution to achieve the lowest average content delivery latency. We use simulation experiments to validate the effectiveness of our proposed framework. Numerical results show that the proposed framework can significantly improve content cache hit rate, reduce content delivery latency and outbound traffic volume in comparison with known existing caching strategies.

• Journal of KIISE:Computing Practices and Letters
• /
• v.5 no.6
• /
• pp.700-707
• /
• 1999

본 논문에서는 GIS 관련 연산을 실시간에 효율적으로 처리하기 위한 분산공유메모리 기반 병렬처리 시스템을 제안한다. 본 논문의 분산공유메모리 시스템은 메시지전달 방식의 분산메모리 MIMD 컴퓨터 상에 소프트웨어 기반 분산공유메모리 모듈을 탑재함으로써 구현되었다. 또한 GIS 연산의 기본이 되는 공간 객체를 공유의 기본 단위로 설정하고, GIS 데이타의 특성을 반영하여 읽기전용 공유데이타 타입을 추가하였으며, 네트워크 오버헤드를 줄이기 위하여 복수의 객체를 한번에 읽어오는 bulk access가 가능하도록 하였다. 본 시스템에서는 GIS 데이타의 효율적인 분배를 위하여 부하균등화 기법으로 guided self scheduling을 사용하였다. 실험결과 본 시스템은 네트워크 캐쉬의 효율적인 활용을 통하여 소프트웨어 기반 분산메모리 시스템의 오버헤드에도 불구하고 MPI 기반 메시지전달 방식에 비하여 향상된 성능을 얻을 수 있었다.Abstract In this paper, we propose a distributed shared memory(DSM) based parallel processing system to process GIS related computations efficiently in real time. The system is based on a software DSM module implemented on top of a distributed MIMD computer. In the DSM system, spatial object, which is a fundamental structure to represent GIS data, is used as a basic unit for sharing, and a read-only shared data type is added to reflect the characteristics of GIS data. In addition, a bulk access to multiple shared data is made possible to reduce the network overhead. A guided self scheduling method is devised for efficient load balancing in distributing GIS data to parallel processors. The experimental results show that the DSM system performs better than an MPI based message-passing system through the efficient utilization of network cache in spite of the system's software overhead.

• Journal of Convergence for Information Technology
• /
• v.12 no.1
• /
• pp.7-14
• /
• 2022

Unlike general maps, the underground geospatial Information is a system made to view underground information in a 3D shape. This system is managed by a tile maps to lighten the data. But there are various underground structures in the basement, and the structures are made of 3D data, so the data size is large. Therefore, when a client mobile program requests a tile map, the service server fetches the requested tile map from the DB server and transmits ti to the client, but there is a transmission delay time problem. In this paper, we design the tile cache method to improve the request response speed for the tile map data provided to the client in the mobile underground geospatial information system. We propose a method in which a service server predicts and prefetchs the next tile map while the client is viewing tile map, and stores the prefetching data in the memory of client mobile terminal. Then, the transmission delay time problem can be solved.

• The KIPS Transactions:PartB
• /
• v.9B no.3
• /
• pp.303-310
• /
• 2002

In this paper, we propose a proxy caching scheme that stores a portion of a continuous media object or an entire object on the Internet. The proxy stores the initial fraction of a continuous media object and determines the optimal size of the initial fraction of the continuous media object to be cached based on the object popularity. Under the proposed scheme, the initial latency of most clients and the data transferred from a remote server can be reduced and limited cache storage space can be utilized efficiently. Considering the characteristics of continuous media, we also propose the novel popularity for the continuous media objects based on the amount of the data of each object played by the clients. Finally, we have performed trace-driven simulations to evaluate our caching scheme and the popularity for the continuous media objects. Through these simulations, we have verified that our caching scheme, PPC outperforms other well-known caching schemes in terms of BHR, DSR and replacement and that popularity for the continuous media objects based on the amount of the playback data can enhance the performance of caching scheme.

• The KIPS Transactions:PartA
• /
• v.10A no.5
• /
• pp.453-462
• /
• 2003

IP address lookup of router searches and decide proper output link using destination address of IP packet that arrie into router. The IP address lookup is essential part in te development of high-speed router needed to high-speed backbone network as one of bottleneck of router performance. This paper introduces DTC data structure that can support gigabit IP address lookup by dynamic trie compression technique that just uses small memory in conventional Pentium CPU. When make a forwarding table by trie compression, the DTC can dynamically select a size of data structure with considering correlation between table's size and searching speed. Also, when compress the prefix trie, DTC makes IP address lookup on the forwarding table of a search on the high speed SRAM cache by minimizing the size of data structure reflecting the structure of the trie. In the experiment result, the DTC data structure recorded performance of maximum $12.5{\times}10^5$ LPS (lookup per second) in conventional Pentium CPU through a dynamic building of most suitable compression over variety of routing tables.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.8 no.5
• /
• pp.187-195
• /
• 2008

In MANET, it is important to efficiently manage data objects and traffic to improve network throughput and response time. This paper proposes a connection-degree based data object search and replication scheme in mobile ad-hoc networks. In this method, information about the path data object providers is stored at the cache of a cluster header at which lots of queries arrive, so that, to reduce the load of the cluster header, replicas can be created at its neighboring nodes. The method proposes an algorithm that picks up a cluster header among mobile nodes and makes it possible to search for and utilize adaptive and up-to-date information in MANET. The method is expected to be effective since it enables access to data objects in spite of broken links among mobile nodes with an enhancement in network response time of searching and a decrease in communication costs. The efficiency of this system was verified via simulation.

• Journal of KIISE:Computing Practices and Letters
• /
• v.14 no.5
• /
• pp.537-541
• /
• 2008

Log-Structured File System(LFS) is a disk based file system that is optimized for improving the write performance. LFS gathers dirty data in memory as long as possible, and flushes all dirty data sequentially at once. In a real system, however, maintaining dirty data in memory should be flushed into a disk to meet file system consistency issues even if more memory is still available. This synchronizations increase the cleaner overhead of LFS and make LFS to write down more metadata into a disk. In this paper, by adapting Non-volatile RAM(NV-RAM) we modifies LFS and virtual memory subsystem to guarantee that LFS could gather enough dirty data in the memory and reduce small disk writes. By doing so, we improves the performance of LFS by around 2.5 times than the original LFS.