• The Journal of Information Technology
• /
• v.4 no.2
• /
• pp.25-36
• /
• 2001

In this paper, we present the development of distributed dual real-time kernel system. This paper proposed that real-time applications should be split into small and simple parts with real-time constraints, Following this concept we have designed to preserve the properties of both hard real-time kernel and general kernel. To satisfy these properties, we designed real-time kernel and general kernel, that have their different properties. In real-time tasks, interrupt processing should be un. In general kernel, non real-time tasks or general tasks are run. We compared the results of this study for performance of the proposal real-time kernel with both RT Linux 0.5a and QNX 4.23A, that is, of interrupt latency scheduling precision and message passing.

• Journal of KIISE
• /
• v.42 no.11
• /
• pp.1332-1338
• /
• 2015

Recently, growing attention has been paid to multi-GPU rendering to support real-time high-quality rendering at high resolution. In order to attain high performance in real-time multi-GPU rendering, great care needs to be taken to reduce the overhead of data transfer among GPUs and frame composition. This paper presents a novel multi-GPU algorithm that greatly enhances split frame rendering with implicit query-based synchronization. In order to support implicit synchronization in frame composition, we further present a message queue-based scheduling algorithm. We carried out an experiment to evaluate our algorithm, and found that our algorithm improved rendering performance up to 200% more than previously existing algorithms.

• Journal of KIISE:Computer Systems and Theory
• /
• v.31 no.8
• /
• pp.432-444
• /
• 2004

In this paper, we construct the TMO-based object group model on distributed environment, and analyze and evaluate the executability for distributed real-time service of our object group model by developing the distributed real-time application simulator applying the model. The Time-triggered Message-triggered Object(TMO) is a real-time server object having real-time property itself. The TMO-based object group is defined as a set of objects which logically reconfigured the physically distributed one or more TMOs on network by a given distributed application. For supporting group management of the server objects, the TMO-based object group we suggested provides the functions which register and withdraw the solver objects as a group member to an arbitrary object group, and also provides the functions which insert and delete the access rights of server objects from clients. Also, our model was designed and implemented to support the appropriate object selection and dynamic binding service for a single TMO as well as the duplicated TMOs, and to support the real-time scheduling service for the clients which are requesting the service. Finally, we developed the Defence System against Invading Enemy Planes(DSIEP) simulator as a practical example of distributed real-time application by applying our model, and evaluated the adaptability of distributed service strategies for the group components and the executability of real-time services that the TMO-based object group model provides.

• Journal of KIISE:Computer Systems and Theory
• /
• v.27 no.1
• /
• pp.23-33
• /
• 2000

We propose a model which overlaps communication with computation for efficient communication in the data-parallel programming paradigm. The overlapping model divides a given loop partition into several sub-partitions to obtain computation which can be overlapped with communication. A loop partition sometimes refers to other data partitions, but not all iterations in the loop partition require non-local data. So, a loop partition may be divided into a set of loop iterations which require non-local data, and a set of loop iterations which do not. Each loop sub-partition is dynamically scheduled depending on associated message arrival, The experimental results for a few benchmarks in IBM SP2 show enhanced performance in our overlapping model.

• The Journal of the Korea Contents Association
• /
• v.14 no.5
• /
• pp.8-17
• /
• 2014

Petascale systems(so called supercomputers) have been mainly used for supporting communication-intensive and tightly-coupled parallel computations based on message passing interfaces such as MPI(HPC: High-Performance Computing). On the other hand, computing paradigms such as High-Throughput Computing(HTC) mainly target compute-intensive (relatively low I/O requirements) applications consisting of many loosely-coupled tasks(there is no communication needed between them). In Korea, recently emerging applications from various scientific fields such as pharmaceutical domain, high-energy physics, and nuclear physics require a very large amount of computing power that cannot be supported by a single type of computing resources. In this paper, we present our HTCaaS(High-Throughput Computing as a Service) which can leverage national distributed computing resources in Korea to support these challenging HTC applications and describe the details of our system architecture, job execution scenario and case studies of various scientific applications.

• Journal of KIISE
• /
• v.43 no.6
• /
• pp.613-620
• /
• 2016

This paper presents a MOHA(Many-Task Computing on Hadoop) framework which aims to effectively apply the Many-Task Computing(MTC) technologies originally developed for high-performance processing of many tasks, to the existing Big Data processing platform Hadoop. We present basic concepts, motivation, preliminary results of PoC based on distributed message queue, and future research directions of MOHA. MTC applications may have relatively low I/O requirements per task. However, a very large number of tasks should be efficiently processed with potentially heavy inter-communications based on files. Therefore, MTC applications can show another pattern of data-intensive workloads compared to existing Hadoop applications, typically based on relatively large data block sizes. Through an effective convergence of MTC and Big Data technologies, we can introduce a new MOHA framework which can support the large-scale scientific applications along with the Hadoop ecosystem, which is evolving into a multi-application platform.

• Journal of KIISE:Information Networking
• /
• v.32 no.5
• /
• pp.633-640
• /
• 2005

This paper proposes and analyzes a message scheduling scheme and corresponding bandwidth allocation method for the hard real-time communication on dual standard 802.11 Wireless LANs. By making the superframeof one network precede that of the other by half, the dual network architecture can minimize the effect of deferred beacon and reduce the worst case waiting timeby half. The effect of deferred beacon is formalized and directly considered to decide the polling schedule of PCF phase. Simulation results executed via ns-2 show that the proposed scheme can improve the schedulability by 3$36\%$ for real-time messages and give $9\%$ more bandwidth to non-real-time messages for the given stream sets, compared with the network whose bandwidth is just doubled with the same MAC.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.46 no.11
• /
• pp.92-100
• /
• 2009

In this paper, we analyze the hardware architecture of Low Density Parity Check (LDPC) decoder using Log Likelihood Ration-Belief Propagation (LLR-BP) decoding algorithm. Various design issues that affect the decoding performance and the hardware complexity are discussed and the tradeoffs between the hardware complexity and the performance are analyzed. The message data for passing error probability is quantized to 7 bits and among them the fractional part is 4 bits. To maintain the decoding performance, the integer and fractional parts for the intrinsic information is 2 bits and 4 bits respectively. We discuss the alternate implementation of $\Psi$(x) function using piecewise linear approximation. Also, we improve the hardware complexity and the decoding time by applying overlapped scheduling.

• Journal of IKEEE
• /
• v.19 no.3
• /
• pp.385-391
• /
• 2015

In this paper, we analyze the communication performance in a time synchronous multi-channel environment and deal with a packet collision avoidance technique to improve it based on IEEE1609.4 for increasing the efficiency of the control channel IEEE802.11p WAVE communication system. In previous works, they tried to solve this problem by message scheduling method on application layer software or changing the value of the random back-off optionally Contention Window. In this paper, we propose a method for adjusting the Channel Guard Interval for packet collision avoidance. The performance was evaluated by the actual vehicle test. The result was confirmed performance over 90% PDR(Packet Delivery Ratio).

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.45 no.6
• /
• pp.18-26
• /
• 2008

In this paper, dynamic GTS scheduling method based on IEEE 802.15.4 is proposed for wireless control system considering reliability and real-time property. The proposed methods can guarantee a transmission of real-time periodic and sporadic data within the limited time frame in factory environment. The superframe of IEEE 802.15.4 is used for the dynamic transmission method of real-time mixed traffic (periodic data, sporadic data, and non real-time message). By separating CFP and CAP properly, the periodic, sporadic, and non real-time messages are transmitted effectively and guarantee real-time transmission within a deadline. The simulation results show the improvement of real-time performance of periodic and sporadic data at the same time.