• Smart Media Journal
• /
• v.11 no.2
• /
• pp.70-76
• /
• 2022

IoT application development using a cloud server causes problems such as data transmission and reception delay, network traffic, and cost for real-time processing support in network connected hardware. To solve this problem, edge cloud-based platforms can use neuromorphic hardware to enable fast data transfer. In this paper, we propose a model optimization method for supporting spiking neural networks on FPGA hardware. We focused on auto-adjusting network model parameters optimized for neuromorphic hardware. The proposed method performs optimization to show higher performance based on user requirements for accuracy. As a result of performance analysis, it satisfies all requirements of accuracy and showed higher performance in terms of expected execution time, unlike the naive method supported by the existing open source framework.

• The KIPS Transactions:PartC
• /
• v.10C no.2
• /
• pp.217-232
• /
• 2003

QoS routing can improve network performance while providing support for QoS guarantees. These benefits, however, comes with additional routing costs such as more complex and frequent route computation and the protocol overheads to exchange dynamic network state information. Moreover, little has been done to mininize the impact of the QoS traffic to the best effort traffic or to enhance the routine Performance of the best effort traffic when QoS routing is deployed. In this paper, it is proposed that a sewer based routing mechanism, which supports the QoS routing without incurring the QoS routing protocol overhead for the network state update exchanges and enhances the performance of the best effort traffic without affecting the performance of QoS routing. Simulation results show that the proposed scheme enhances the routing performance for the QoS traffic while reducing the routing protocol overhead. The routing performance of the best traffic is also improved with virtually no impact to the routine performance of the QoS traffic. The proposed scheme is shown to be especially effective when the ratio of QoS traffic is high, that is, when the impact of the QoS traffic to the performance of best effort traffic is significant.

• The Journal of the Korea Contents Association
• /
• v.20 no.8
• /
• pp.368-374
• /
• 2020

Journaling file systems (JFS) manage changes of file systems not yet committed in a data structure known as a journal to restore the file system in the event of an unexpected failure. Extra write operations required for journaling negatively affect the performance of JFS. The high-performance and byte-addressable non-volatile memory (NVM) was expected to easily mitigate these performance problems by providing NVM space as journal storage. However, even with such non-volatile memory technologies, performance problems still arise due to scalability problems inherent in processing transactions of JFS. To solve this problem, we proposes a technique for processing file system transactions for scalable performance. To this end, lock-free data structures are used and multiple I/O requests are allowed to simultaneously be processed on high-performance storage devices with multiple I/O channels. We evaluate the file system with the proposed technique by comparing the original ext4 file system and the recent proposed NVM-based journaling file system on a multi-core server, and experimental results show that our file system has better performance (up-to 2.9/2.3 times) than the original ext4 file system and the recent NVM-based journaling file system, respectively.

• The Journal of Bigdata
• /
• v.4 no.2
• /
• pp.141-158
• /
• 2019

Based on RBV(Resource Based View), IT of SMEs classified into IT resource and capabilities. And We confirmed that capabilities and resources affected each performance. In other words, based on the questionnaire of SMEs and IT professionals, divides capability from the overall IT resource that are possessed by SMEs. Among the four attributes (value, rare, non-substitutability, imperfect imitability) presented by Barney (1991), this study targeted at value and imperfect imitability and investigated how SMEs recognize IT resource and capability. Furthermore, this study tests how IT resource and capability influence corporate performance. The result of this study finds that resources that are needed on "Knowledge-based" are classified into IT capability, otherwise classified into IT resource. Analysis shows that server, DB(database), system administrators, programmers, CIO, BA were capabilities, Desktop PC, PC software, software for salary and accounting management, e-commerce, Homepage, and network inside th enterprise were resources. Secondly, this study reveals that both IT resource and IT capability affected company performance (employee satisfaction, CEO satisfaction). IT is certainly having an impact on corporate performance. In conclusion, resource can be either IT resource or IT capability based on they way of utilization. And both IT resource and IT capability have an influence on corporate performance (employee job satisfaction, CEO satisfaction). Therefore, when considering IT investment, a company can purchase necessary IT resource and actively utilize it to be IT capability, which can have an influence on corporate performance in return.

• Journal of Korea Game Society
• /
• v.4 no.4
• /
• pp.11-18
• /
• 2004

The on-line games in the past were played by only two persons exchanging data based on one-to-one connections, whereas recent ones (e.g. MMORPG: Massively Multi-player Online Role-playings Game) enable tens of thousands of people to be connected simultaneously. Specifically, Korea has established an excellent network infrastructure that can't be found anywhere in the world. Almost every household has a high-speed Internet access. What made this possible was, in part, high density of population that has accelerated the formation of good Internet infrastructure. However, this rapid increase in the use of on-line games may lead to surging traffics exceeding the limited Internet communication capacity so that the connection to the games is unstable or the server fails. expanding the servers though this measure is very costly could solve this problem. To deal with this problem, the present study proposes the load distribution technology that connects in the form of local clustering the game servers divided by their contents used in each on-line game reduces the loads of specific servers using the load balancer, and enhances performance of sewer for their efficient operation. In this paper, a cluster system is proposed where each Game server in the system has different contents service and loads are distributed efficiently using the game server resource information such as CPU utilization. Game sewers having different contents are mutually connected and managed with a network file system to maintain information consistency required to support resource information updates, deletions, and additions. Simulation studies show that our method performs better than other traditional methods. In terms of response time, our method shows shorter latency than RR (Round Robin) and LC (Least Connection) by about 12%, 10% respectively.

• Journal of KIISE:Computing Practices and Letters
• /
• v.11 no.3
• /
• pp.224-234
• /
• 2005

As the network-related technology develops the number of both Internet users and the usage are explosively increasing. The networking traffic is increasing in the campus as the networking system inside universities, following the trend, adds more nodes and various networking services. Nonetheless, the quality of services for users has been degraded. Accordingly, core problems, which can cause troubles for network management, design and expansion of the network, and the cost policy, has appeared. To effectively cope with the problems with analyses a great number of technicians, tools, and budget are needed. However, it is not possible for mid and small-sized colleges to spend such a high expenditure for professional consulting. To reduce the cost and investment creating the optimized environment, the analyses on the replacement of the tools, changing the network structure, and performance analysis about capacity planning of networking is necessary. For this reason, in this paper, framework-based performance management tools are used for all steps that are related to the subject of the analysis for the network management. As the major research method, the current data in detailed categories are collected, processed, and analyzed to provide the solution for the problems. As a result we could manage the network, server, and application more systematically and react efficiently to errors and degrading of performance that affect the networking tasks. Also, with the scientific and organized analyses the overall efficiency is upgraded by optimizing the cost for managing the operation of entire system.

• Journal of KIISE:Databases
• /
• v.31 no.5
• /
• pp.505-515
• /
• 2004

One of the most important issue in location-aware computing is tracking moving objects efficiently. To this end, an efficient protocol which updates location information in a location server is highly needed. In fact, the performance of a location update strategy highly depends on the assumed mobility pattern. In most existing works, however, the mobility issue has been disregarded and too simplified as linear function of time. In this paper, we propose a new mobility model, called state-based mobility model (SMM) to provide more generalized framework for both describing the mobility and updating location information of complexly moving objects. We also introduce the state-based location update protocol (SLUP) based on this mobility model. Using experimental comparison, we illustrate that the proposed technique is many times better in reducing location update cost and the communication bandwidth consumption.

• Journal of Mechanical Science and Technology
• /
• v.20 no.1
• /
• pp.85-93
• /
• 2006

The trend of miniaturization has been applied to the research of rockets to develop prototypes of micro rockets. In this paper, the development of a web-integrated prototyping system for three-dimensional micro rockets, and the results of combustion tests are discussed. The body of rocket was made of 6061 aluminum cylinder by lathe process. The three-dimensional micro nozzles were fabricated on the same aluminum by using micro endmills with ${\phi}100{\mu}m{sim}{\phi}500{\mu}m$ diameter. Two types of micro nozzle were fabricated and compared for performance. The total mass of the rockets was 7.32 g and that of propellant (gun powder) was 0.65 g. The thrust-to-weight ratio was between 1.58 and 1.74, and the flight test with 45 degree launch angle from the ground resulted in $46\;m{\sim}53\;m$ of horizontal flight distance. In addition, ABS housing for the micro machined rocket was fabricated using Fused Deposition Modeling (FDM). A web-based design, fabrication, and test system for micro nozzles was proposed to integrate the distributed hardware resources. Test data was sent to the designer via the same web server for the faster feedback to the rocket designer.

• ETRI Journal
• /
• v.36 no.5
• /
• pp.819-828
• /
• 2014

To solve the increasing popularity of video streaming services over the Internet, recent research activities have addressed the locality of content delivery from a network edge by introducing a storage module into a router. To employ in-network caching and persistent request routing, this paper introduces a hybrid content delivery network (CDN) system combining novel content routers in an underlay together with a traditional CDN server in an overlay. This system first selects the most suitable delivery scheme (that is, multicast or broadcast) for the content in question and then allocates an appropriate number of channels based on a consideration of the content's popularity. The proposed scheme aims to minimize traffic volume and achieve optimal delivery cost, since the most popular content is delivered through broadcast channels and the least popular through multicast channels. The performance of the adaptive scheme is clearly evaluated and compared against both the multicast and broadcast schemes in terms of the optimal in-network caching size and number of unicast channels in a content router to observe the significant impact of our proposed scheme.

• The Transactions of the Korea Information Processing Society
• /
• v.2 no.6
• /
• pp.815-822
• /
• 1995

In the conventional UNIX system, the all device driver realized in the kern el. Hence, whenever we want to add a new device driver or change the existin g device driver, the modification of kernel is unavoidable. Generally, it is very difficult to modify the kernel codes.As a method of overcoming this difficulty, a TTY device driver model of user-level is presented in this paper. The basic concept of this model is providing a dynamic reconfiguration of TTY device driver by realizing a user-level server process for TTY device driver. In order to verify the proper y of this model, a prototype of TTY device driver has been realized in the SunOS and Linux environment and evaluated its performance.