Virtualized service and network functions are deployed on virtual machines (VMs) to realize essential processing to realize service function chaining (SFC). Issues on SFC is SF allocation to a VM and to minimize the response time and number of function instances. In this paper, we propose an SF clustering-based scheduling algorithm, called "SF-clustering for utilizing virtual CPUs" (SFCUV), to solve the SF allocation and SF selection problems simultaneously. Experimental results show that SF-CUV can utilize vCPUs to minimize the response time.

Big data computing in the cloud needs faster processing and better resource provisioning. MapReduce is the framework for computing large scale datasets in cloud environments. Optimization of resource requirement for each job to satisfy a specific objective in MapReduce is an open problem. Many factors, e.g., system side information and requirements of each client must be considered to estimate the appropriate amount of resources. This paper presents a mathematical model for the optimal number of map tasks in MapReduce resource provisioning. This model is to estimate the optimal number of the mappers based on the resource specification and the size of the dataset.

Edge computing can realize a data locality among a cloud and users, and it can be applied to task offloading, i.e., a part of workload on a mobile terminal is moved to an edge or a cloud system to minimize the response time with reducing energy consumption. Mobile workflow jobs have been widely used due to advance of computational power on a mobile terminal. Thus, how to offload or schedule each task in a mobile workflow is one of the current challenging issues.
In this paper, we propose a task scheduling algorithm with task offloading, called priority-based continuous task selection for offloading (PCTSO), to minimize the schedule length with energy consumption at a mobile client being reduced. PCTSO tries to select dependent tasks such that many tasks are offloaded so as to utilize many vCPUs in the edge cloud; in this manner, the degree of parallelism can be maintained. Experimental results of the simulation demonstration that PCTSO outperforms other algorithms in the schedule length and satisfies the energy constraint.

To develop and interoperate smart city applications efficiently, smart city IoT platforms require efficient handling of various types of sensor devices, networking and computing resources, and different domain applications. To address this fact, the authors of this paper proposed a research project named 'Fed4IoT', which is an acronym for federation of IoT and cloud infrastructures, to provide scalable and interoperable smart city applications. The project primarily proposes two key technologies: IoT device virtualization and context-information sharing. In this paper, we introduce the IoT device virtualization (one of key technology in Fed4IoT) that enables efficient utilization of networking and computing resources by applying a micro-service sharing and dynamic resource scaling. In the performance evaluations, we implement three micro services, such as object detection, garbage detection and road damage detection, as smart city applications and dynamically and flexibly allocate computing resources to three micro services by using Docker and Kubernetes. Through the evaluations we validate that the proposed IoT device virtualization achieves the efficient computing resource utilization and reduction of networking resources.

To develop and interoperate smart city applications efficiently, smart city IoT platforms require efficient handling of various types of sensor devices, networking and computing resources, and different domain applications. To address this fact, in this paper, we introduce an IoT device virtualization that enables efficient utilization of computing resources. The proposal applies a micro-service sharing and dynamic resource scaling. In the performance validations, we implement an early prototype using Docker, Kubernetes, and Apache Kafka. Through the preliminary experiment, we confirm that the proposal can improve the application processing time by appropriately sharing and scaling micro services.

Recently, several routing protocols have been proposed anddeveloped for MANETs, such as Optimized Link State Rout-ing (OLSR). OLSR is one of well-known proactive protocolsfor MANETs. Fundamental functions of OLSR are Multi-point Relay (MPR) selection and routing table calculation.MPR selection and routing table calculation based on linkdisconnection prediction are important to improve networkperformance. From such background, in [1], the authors pro-posed and evaluated Multipoint Relay (MPR) selection androuting algorithm based on the link disconnection predictioncaused by the node mobility. In this proposed routing algo-rithm, each node recalculates the routing table assuming thatthe link between nodes is disconnected before the link dis-connection between nodes actually occurs. Therefore, thereis a possibility that routes with longer hop-count are selectedon the routing table.In this paper, we propose MPR selection and routing algo-rithm based on link disconnection prediction and hop-countlimitation. In this proposed routing algorithm, each node re-calculates the routing table assuming that the link betweennodes is disconnected if the link disconnection between nodesis predicted. In addition, the hop-count of each route of therouting table which is constructed by the proposed routing al-gorithm is limited within H hops longer than the hop-countof the original route of OLSR. In experimental evaluation, weshow that OLSR with the proposed MPR selection and rout-ing algorithm outperforms the original OLSR in terms of thenetwork performance.

The near real-time flood map obtained from satellite datais essential data for government measures and flooddamage monitoring and efficient rescue activities. It isnecessary to distinguish between standing water and floodwater from the damage map. Bangladesh is considered tobe among the countries most vulnerable to a variety ofnatural disasters. The major natural hazards whichBangladesh has to confront regularly include floods,cyclones and accompanying storm surge. Flooding fromriver waters overflowing the banks, particularly duringmonsoon, is an annual phenomenon. Flash floods, whichare caused by heavy or excessive rainfall in a short periodover a relatively small area. In this study, spatial-temporalpatterns of continuous water area were examined usingthe patterns observed in metrics calculated for 15-year ofModerate Resolution Imaging Spectrometer (MODIS).Four analytical approached were used; calculation oftemporal Red, Blue, Near Infrared, Short Wave Infraredlayers for selected water area.First, the annual time-series data of MODIS were used tomainly detect floodwaters in the case of the 15-year (from2001 to 2016) flood. Five training samples (water, forest,cultivated area, urban, snow) and a number of differentattributes were tested with sample areas. Three machinelearning methods were used to find the most influentialindex which compose the model for classification offlooded areas.

Network Function Virtualization (NFV) has been widelyspread for software-level network configuration. Such functionalitiesinteract each other to achieve data-flow processing,named as Service Function Chaining (SFC). In SFC,each execution unit, named as Virtualized Network Function(VNF), has dependencies with others. Such configuration isrealized by allocating each VNF to a VM. Conventional approachesmainly focus on how to allocate each VNF to a VM,or how to select a VNF and they are resolved separately. Inthis paper, we propose a VNF clustering algorithm, calledClustering for Minimizing Worst Schedule Length for VNF(CMWSL-VNF) to resolve both VNF allocation and VNFselection simultaneously. CMWSL-VNF can greatly reducethe data communication latency and maintains the degree ofparallelism by the clustering algorithm. Experimental resultsshow that CMWSL-VNF can utilize vCPUs to minimize theresponse time.

In recent years, most of Internet traffic is occupied by video.From such background, Content-Centric Networking (CCN)has been emerged as a new networking paradigm. The mainfeatures of CCN architecture are an in-network caching and acontent-based routing. In [1], the authors proposed and evalu-ated an in-network caching method considering content pop-ularity and playback sequences of content chunks for reduc-ing playback interruption time on video streaming over CCN.However, this in-network caching method has many systemparameters in terms of content popularity and playback se-quences of content chunks and the parameter tuning is difficult.In this paper, we propose an adaptive in-network cachingmethod based on feedback mechanism using the playbackinterruption time of the received data (i.e., content chunks).Experimental evaluations show that the proposed adaptivein-network caching method outperforms in-network cachingmethods proposed in past literature in terms on playback in-terruption time.

Data parallel processing such as MapReduce is becomingmore and more attractive due to its simple and effectiveparadigm for large amount of data analysis. Conventionalapplications on MapReduce have been executed at the manuallyconfigured status, e.g., the number of map/reduce slots,the size of data chunk, and so on. However, there is no guaranteethat such status is optimal for minimizing the responsetime. In this paper, we study how to derive the number ofmap/reduce slots, as well as the size of data chunk by mathematicalanalysis.

Recently, several routing protocols have been proposed anddeveloped for MANETs, such as Optimized Link State Rout-ing (OLSR). OLSR is one of well-known proactive protocolsfor MANETs. Fundamental functions of OLSR are Multi-point Relay (MPR) selection and routing table calculation.MPR selection and routing table calculation based on linkdisconnection prediction are important to improve networkperformance. This is because the reselection of MPR nodesand the recalculation of routing table can be performed beforethe link disconnection among nodes actually occurs if the linkdisconnection time among nodes can be predicted.In this paper, we propose routing algorithm based on thelink disconnection prediction caused by the node mobility.The proposed routing algorithm predicts the link disconnec-tion time among nodes caused by the node mobility and uti-lize the predicted link disconnection time in the recalcula-tion of routing table. In experimental evaluation, we showthat OLSR with the proposed routing algorithm outperformsOLSR with the original routing algorithms in terms of thenetwork performance.

In recent years, most of the internet traffic are occupied by thelarge-capacity contents such as web, voice and video. In orderto solve this problem, Content-Centric Networking (CCN)has been emerged as a new networking paradigm. The mainfeatures of CCN architecture are an in-network caching and acontent-based routing. In [1], the authors proposed and evalu-ated an in-network caching method considering the playbacksequences of content chunks for reducing playback interrup-tion time of the video streaming over CCN. However, this in-network caching method does not consider whether the con-tents and/or chunks are used recently and frequently.In this paper, we propose an in-network caching methodconsidering recency and frequency of content chunks accessand playback sequence of content chunks. Experimental eval-uations show that the proposed in-network caching methodoutperforms in-network caching methods proposed in past lit-erature in terms on playback interruption time.

The final goal of this study is to create data to supportemergency efforts in a disaster affected area by locatingdamaged buildings shortly after the disaster. In this study,prioritizing the practicality of the method for emergencypurposes, we designed a method only to use a singlesatellite image of an affected area, eliminating the use ofcomplex algorithms and auxiliary data. The uniqueness ofour method lies in the application of an object-basedregion segmentation to images and the use of features ofobjects obtained from texture, hierarchical and otherinformation in order to extract damaged buildings. Out of26 features resulting from the analysis of objects, wefound one feature and three combinations of two differentfeatures that are effective in extracting damaged buildings,such as Rectangular fit, Homogeneity, Number of subobjects/Area, and Length of longest of edge/Area.

Satellite data has often been used to investigate damage from natural disasters (e.g., flood, drought, forest fire and volcano eruption) close to real-time. In particular, Terra/MODIS and Aqua/MODIS have been widely used for retrieving information on natural disasters and land cover changes because of their advantage of good ground resolution (maximum 250 m) and daily availability (full daily coverage of the Earth). However, the observation of areas covered by clouds using optical sensors does not work. Although the composite method is widely used to reduce the influence of clouds, clouds are likely to remain in some compositing periods. This paper presents the White Object Index (WOI), used for estimating the cloud fraction of a pixel in satellite imagery. The WOI was obtained from differences in reflectance of visible and short-wave infrared spectral bands using a mixture model. The efficiency of the WOI was tested by comparing the maximum value composite (MVC) method with MOD35 using time series data. As a result, WOI was confirmed to be an efficient tool for estimating the influence of clouds within a pixel.

Many workflow scheduling algorithms for heterogeneous systems have been developed to satisfy multiple requirements such as minimizing schedule length while maximizing throughput. In particular, in list-based scheduling approaches, the schedule length depends on the given nodes as well as the task allocation and ordering policies. This is because the scheduling priority is derived by averaging the execution time and communication time of the given nodes. If the set of nodes can be adjusted before the scheduling tasks, a small schedule length can be achieved. In this paper, we propose a prior node selection algorithm, called lower bound based candidate node selection (LBCNS) to select a subset of given nodes to minimize the schedule length while fairly scheduling each job. Our proposal has two approaches: (i) LBCNS_DEFAULT, which considers the job characteristics and each node's performance, and (ii) priority-based LBCNS, which additionally takes each scheduling priority into account for a dedicated task scheduling algorithm.
The experimental results of extensive simulations show that LBCNS_DEFAULT has the best fairness for scheduling multiple workflow jobs, while priority-based LBCNS achieves the minimum schedule length with the highest efficiency for a single workflow job and multiple workflow jobs. (C) 2017 Elsevier Inc. All rights reserved.

主筆<br />
論文4で提案したタスクスケジューリングアルゴリズムを実環境用に実装し，その有効性を実環境で検証した．具体的にはガウス消去法，高速フーリエ変換のジョブについて，既存のタスクスケジューリングアルゴリズムと応答時間に関して比較した．その結果，いずれの場合も論文4におけるアルゴリズムが最も応答時間が小さくなることを確認した．

Recently, Content-Centric Networking (CCN) has emerged as a new networking paradigm. The main features of CCN architecture are in-network caching and content-based routing. Cache replacement and decision policies have been discussed as important mechanisms of in-network caching in past literature. However, these policies cannot efficiently support streaming. In this paper, we propose an in-network caching method for reducing playback interruption time of on-demand streaming over CCN through efficient use of network resources. The proposed method splits streaming content data into chunks with fixed size and stores these chunks into the cache storage of each CCN router in order to reduce playback interruption time for streaming. Experimental evaluations show that the proposed in-network caching method outperforms in-network caching methods of the original CCN architecture in terms of playback interruption time.

Edge computing is one of current attractive schemes, withproviding additional processing capability to a cloud systemas well as taking a role of intermediate data caching betweencloud and users. This architecture can be applied to taskoffloading, e.g., a part of workload on a mobile terminal ismoved to an edge or a cloud system in order to minimize theresponse time with reducing energy consumption at the mobileterminal. Furthermore, a mobile workflow job, in whicheach task has dependencies with others, has been widely useddue to advance of computational power on a mobile terminal.Thus, how to offload or how to schedule each task in a mobileworkflow is one of current challenging issues.In this paper, we propose a task scheduling with task offloadingalgorithm to minimize the schedule length (responsetime) with reduced energy consumption at a mobile terminal.Experimental results by the simulation show that proposedtask scheduling algorithm with task offloading obtains a lowenergy and effective scheduling results.

Parallelization paradigms for effective execution in a Directed Acyclic Graph (DAG) application have been widely studied in the area of task scheduling. Schedule length can be varied depending on task assignment policies, scheduling policies, and heterogeneity in terms of each processor and each communication bandwidth in a heterogeneous system. One disadvantage of existing task scheduling algorithms is that the schedule length cannot be reduced for a data intensive application. In this paper, we propose a clustering-based task scheduling algorithm called Clustering for Minimizing the Worst Schedule Length (CMWSL) to minimize the schedule length in a large number of heterogeneous processors. First, the proposed method derives the lower bound of the total execution time for each processor by taking both the system and application characteristics into account. As a result, the number of processors used for actual execution is regulated to minimize the Worst Schedule Length (WSL). Then, the actual task assignment and task clustering are performed to minimize the schedule length until the total execution time in a task cluster exceeds the lower bound. Experimental results indicate that CMWSL outperforms both existing list-based and clustering-based task scheduling algorithms in terms of the schedule length and efficiency, especially in data-intensive applications.

主筆
ワークフロー型ジョブのスケジューリングで応答時間を最小化するための計算機集合について，事前選択する基準を提案した．シミュレーションの結果，既存のリストスケジューリングアルゴリズムに本事前選択手法を適用すると，特にデータ集約型ジョブについて応答時間が改善されることが分かった．

Several task clustering heuristics are proposed for allocating tasks in heterogeneous systems to achieve a good response time in data intensive jobs. However, one of the challenging problems is the process in task scheduling after task allocation by task clustering. We propose a task scheduling method after task clustering, leveraging worst schedule length (WSL) as an upper bound of the schedule length. In our proposed method, a task in a WSL sequence is scheduled preferentially to make the WSL smaller. Experimental results by simulation show that the response time is improved in several task clustering heuristics. In particular, our proposed scheduling method with the task clustering outperforms conventional list-based task scheduling methods. Copyright 2016. The Korean Institute of Information Scientists and Engineers.

主筆
応答時間の上限値を最小化するため，割り当て済みタスクの実行順を決めるアルゴリズムを提案した．応答時間の上限値（WSL: Worst Schedule Length)を定義し，WSLを支配する実行経路から優先的にスケジュールする手法を述べた．シミュレーションの結果，ランダムタスクグラフにおいて既存手法よりも応答時間が改善されることが分かった．

主筆
ワークフロージョブをスケジューリングにおいて，各計算機へ公平にタスクを割り当てるための計算機のグルーピング手法を提案した．各計算機の性能と各タスクの特徴（仕事量，転送データサイズ）を考慮し，これらの分布とグループ内の性能分布が同一となるようなグルーピングアルゴリズムを提案した．シミュレーションにより，従来手法よりも公平なタスク割り当てが可能になったことが分かった．

In the last decades, many kinds of task models such as grid and cloud computing have been developed. In such distributed systems, each task is processed by respective processor in multicored computers e.g., household PCs which we can easily harness in recent years. If there is one policy to automatically decide the "best" combination and the number of processors (and computers), we effectively utilize those computational resources, thereby large number of jobs can be executed in parallel. In this paper, we propose a method for mapping of execution units for such environments. The method adopts a remapping technology after processor-execution unit mapping[11] is finished. Experimental comparisons by a simulation show the advantages of the proposed method.

One interesting problem regarding wireless local area network (WLAN) ad-hoc networks is the effective mitigation of hidden nodes. The WLAN standard IEEE 802.11 provides request to send/clear to send (RTS/CTS) as mitigation for the hidden node problem
however, this causes the exposed node problem. The first 802.11 standard provided only two transmission rates, 1 and 2 Mbps, and control frames, such as RTS/CTS assumed to be sent at 1 Mbps. The 802.11 standard has been enhanced several times since then and now it supports multi-rate transmission up to 65 Mbps in the currently popular 802.11n (20 MHz channel, single stream with long guard interval). As a result, the difference in transmission rates and coverages between the data frame and control frame can be very large. However adjusting the RTS/CTS transmission rate to optimize network throughput has not been well investigated. In this paper, we propose a method to decrease the number of exposed nodes by increasing the RTS transmission rate to decrease RTS coverage. Our proposed method, Asymmetric Range by Multi-Rate Control (ARMRC), can decrease or even completely eliminate exposed nodes and improve the entire network throughput. Experimental results by simulation show that the network throughput in the proposed method is higher by 20% to 50% under certain conditions, and the proposed method is found to be effective in equalizing dispersion of throughput among nodes. © 2014. The Korean Institute of Information Scientists and Engineers.

主筆
マルチコアPCが分散した環境において，タスク間のデータが大きい場合は同一プロセッサへ割り当て，通信遅延を削減するためのタスク再マッピング手法を提案した．シミュレーションにより，リストスケジューリングよりも応答時間が改善されることが分かった．

In the last decades, many kinds of task execution models such as grid and cloud computing have been developed. In such distributed systems, each task is processed by respective processor in multi cored computers e. g., household PCs which we can easily harness in recent years. If there is one policy to automatically decide the "best" combination and the number of processors (and computers), we effectively utilize those computational resources, thereby large number of jobs can be executed in parallel. In this paper, we propose a method for mapping of execution units for such environments. T he method adopts a re-mapping technology after processor-execution unit mapping[6] is finished. Experimental comparisons by a simulation show the advantages of the proposed method.

In the last decades, many kinds of task execution models such as grid and cloud computing have been developed. In such distributed systems, each task is processed by respective processor in multi cored computers e. g., household PCs which we can easily harness in recent years. If there is one policy to automatically decide the "best" combination and the number of processors (and computers), we effectively utilize those computational resources, thereby large number of jobs can be executed in parallel. In this paper, we propose a method for mapping of execution units for such environments. T he method adopts a re-mapping technology after processor-execution unit mapping[6] is finished. Experimental comparisons by a simulation show the advantages of the proposed method.

主筆<br />
均一環境向けのタスク集約手法（タスククラスタリング）を不均一環境へ適用した場合の応答時間への影響を，理論的な解析及び実験により明らかにした．その結果，応答時間に関して均一環境・不均一環境との差異が小さくするには，各プロセッサへ割り当てる仕事量に上限を設けることが必要であることが分かった．

主筆
P2P Gridにおいて，独立タスクを割り当てて全体としての応答時間を最小化するためのピアグルーピング手法を提案した．ピア同士のホップ数による処理遅延，帯域幅，プロセッサ速度を考慮して，投入ジョブに応じて適切な計算資源をグループ化する．シミュレーションの結果，ホップ数のみを考慮する場合よりも応答時間が改善できることが分かった．

主筆
不均一分散環境でワークフロージョブをスケジュールする問題において，各計算機へ割り当てる適切な仕事量を理論的に算出し，その有効性を確認した．最悪応答時間（WSL）を小さくするように各タスクを割り当てれば，すスケジュール後の応答時間も小さくできることを示し，シミュレーションにおいても確認できた．

主筆<br />
不均一分散環境におけるワークフロージョブ実行の応答時間を最小化するために，各計算資源に割り当てる実行粒度の決定，及び割り当て基準について提案する．シュミレーションの結果，実行粒度の決定が応答時間の最小化には必須であること，及び提案した実行粒度導出手法が応答時間の最小化に有効であることが分かった．

副筆<br />
伝統産業を若者の文化に生かすことにより，地域活性化につながることを群馬県桐生市での実証実験により認められたことを述べる．その上で，人的形成に必要な要素は何か，及び人的形成の課題点についてもまとめた．

主筆<br />
均一分散環境における，タスク集約（クラスタリング）に基づくタスクスケジューリングを提案した．応答時間の最悪値（WSL: Worst Schedule Length)を最小化するようなタスク割り当てができれば，スケジュール後の応答時間の上限・下限値も最小化されることを証明し，そのためのアルゴリズムを提案し，実験により有効性を示した．

This paper presents the interactive web interface for inquiring about the long-term care insurance system in Japan. In Japan, health care services provided by the long-term care insurance system takes costs depending on the health care level. Although the system has been revised several times, such revision made the system more complex, so that a web-based system by which every people can accurately and easily ask about the long-term care insurance system is needed. Thus, the proposed interface in this paper has two points to achieve the requirement. First point is to show only basic information related to selected options on the web page. The second point is that dependencies (parent-child) among each category for inquiring should be presented on the web page to make the user understand each ranges of the category meaning. The experimental evaluation shows that our proposed presentation method using the web interface has advantage on perception among categories, human error, and user-friendliness. © 2010 IEEE.

In a distributed system, which consists of an unknown number of processors, it is important to derive an appropriate number of processors by which the good schedule length is obtained by a task scheduling. Many task clustering heuristics have been proposed to determine the number of processors and to minimize the schedule length for scheduling a directed acyclic graph (DAG) application. However, those heuristics are not aware of the actual number of existing processors. As a result, the number of processors determined by an existing task clustering may exceed that of actually existing processors. Therefore, conventional approaches adopt merging of each cluster for reducing the number of clusters at the expense of decreasing degree of task parallelism. In this paper, we present a static cluster size determination method, which derives the lower bound of the cluster size with considering the DAG structure and the task size to data size ratio to suppress the schedule length with the small number of processors. Our experimental evaluations by simulations show that the lower bound of each cluster size determined by the proposed method has a good impact on both the schedule length and the processor utilization. © 2010 Springer Science+Business Media LLC.

主筆
均一分散環境において，各計算資源へ割り当てる実行単位の大きさ（粒度）をどのように決めれば応答時間を最小化できるかを，理論的に検討し，様々な値で実験して比較を行った．その結果，お魚頭時間のある上限値が粒度の決め方によっては極小値を持つことが分かった．

Grid computing has been widely spread in pragmatic areas and scientific research areas in order to achieve more powerful computation than ordinary computation by single computer. Globus Toolkit4(GT4), which has been the de facto standard middleware in Grid, gives to us many chances of submitting jobs, especially using WSRF technologies. However, in case of cluster type grid using GT4, rules for job execution depend on a local scheduler(LS) which interacts with WS-GRAM in GT4, so that clients can not reflect their own rules to behaviors in each worker node. In this paper, we formulate those potential problems and solutions based on our proposed architecture in [1], whose main feature is to monitor Resource Property(RP) in WS-GRAM and notify its status to worker nodes. Then we describe how to implement a mechanism for monitoring RP in WS-GRAM to achieve the transparent delivery of client's rules to worker nodes. Our experimental results show that our proposed architecture can be incorporated in WS-GRAM without bottleneck in performance points of view. ©2008 IEEE.

In this paper, we present the architecture of efficient resource mapping between global resources and local resources for Grid environment. Many current Grid infrastructures which adopt WSRF as a communication scheme lay emphasis on only managing resource information from local scheduler. As a result, they don't define how each of local node should manipulate specific submitted job using resource data represented by WS-Resource.
Our proposed architecture is directed to resource mapping based on the collaboration of WSRF and JMX such that local nodes can behave autonomously according to resource changes in a master node. We present methods of the mapping and benefits for WSRF-based Grid environment.

In recent years, Web Service technologies are becoming more and more important communication scheme for Grid environment. WSRF-based resource management components, such as WS-GRAM[4] in GT4[3], lay emphasis on how to make existing procedures such as local scheduler be invoked via HTTP so that they inevitably have two problems described below. (l)If assumed environment is cluster type Grid, each worker node does not have access scheme to any information included in Resource Property of master node. (2)As a result, the contents of information sent to clients is forced by the local scheduler and therefore clients can not control job status in accordance with their own policy. If those two problems are resolved, any client's own resource management policy can be reflected on worker nodes.
In this paper we propose a method for resolving those problems in order to realize the interactive communication between clients and worker nodes for WSRF-based cluster Grid.