• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권6호
• /
• pp.2824-2837
• /
• 2019

Container technologies are widely used in infrastructures to deploy and manage applications in cloud computing environment. As containers are light-weight software, the cluster of cloud applications can easily scale up or down to provide Internet-based services. Container-based applications can well deal with fluctuate workloads by dynamically adjusting physical resources. Current works of scheduling applications often construct applications' performance models with collected historical training data, but these works with static models cannot self-adjust physical resources to meet the dynamic requirements of cloud computing. Thus, we propose a self-adaptive automatic container scheduling framework AutoScale for cloud applications, which uses a feedback-based approach to adjust physical resources by extending, contracting and migrating containers. First, a queue-based performance model for cloud applications is proposed to correlate performance and workloads. Second, a fuzzy Kalman filter is used to adjust the performance model's parameters to accurately predict applications' response time. Third, extension, contraction and migration strategies based on predicted response time are designed to schedule containers at runtime. Furthermore, we have implemented a framework AutoScale with container scheduling strategies. By comparing with current approaches in an experiment environment deployed with typical applications, we observe that AutoScale has advantages in predicting response time, and scheduling containers to guarantee that response time keeps stable in fluctuant workloads.

• 한국인터넷방송통신학회논문지
• /
• 제23권4호
• /
• pp.85-91
• /
• 2023

국산 마이크로서버인 KOSMOS의 활용을 위해 클라우드 컴퓨팅 분야의 실제 응용 서비스 기반 벤치마크 프로그램인 CloudSuite로 성능 평가 결과를 제시한다. CloudSuite는 오프라인 응용과 온라인 응용의 두 가지 부분에서 클라우드 서비스로 제공되는 몇 가지의 구분되는 응용 프로그램을 컨테이너 기반으로 제공하고 있다. KOSMOS의 유사 스펙의 비교군인 다른 마이크로서버와의 성능 비교에서 전 부분에 걸처 KOSMOS가 우수하였으며, 인텔 Xeon CPU 기반 서버와의 비교에서도 일부 오프라인 응용에서는 성능이 더 우수하였다. CloudSuite 오프라인 응용 벤치마크 프로그램인 Graph Analytics 수행 시 KOSMOS의 다수의 노드들을 분산 실행시킨 형상에서 인텔 Xeon CPU 기반 2개의 서버 비교군과 비교하였을 때, 각각 30.3%, 72.3%만큼의 수행시간을 감소시켰다.

• 한국IT서비스학회지
• /
• 제14권4호
• /
• pp.221-236
• /
• 2015

Application software is delivered to customers as a form of service at cloud environment. A cloud service provider is a marketplace between supply side (application providers) and demand side (customers). Cloud service providers have to validate applications to be included in their service portfolio. Not only performance, security, networking, compliances should be checked but also business contract, authentication should be provided. Organization customers are more sensitive to these validation criteria and process. We study the Internet2 NET+, which is a successful cloud marketplace of applications for research and education organizations. This case study shows us three things : (i) a cloud marketplace's application management process : selection, validation, transition to service, customization of applications (ii) what a cloud marketplace has for its infrastructure like authentication, security, access control etc. (iii) what a cloud marketplace has as its governance structure. This case study will provide informative analysis of Internet2 NET, a profit-making vertical and buyer's marketplace (education industry). And we will get some strategic implications for planning and implementing cloud marketplaces.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2018년도 춘계학술대회
• /
• pp.419-421
• /
• 2018

Cloud vehicular networks are a promising paradigm to improve vehicular through distributing computation tasks between remote clouds and local vehicular terminals. Software-Defined Network(SDN) can bring advantages to Intelligent Transportation System(ITS) through its ability to provide flexibility and programmability through a logically centralized controlled cluster that has a full comprehension of view of the network. However, as the SDN paradigm is currently studied in vehicular ad hoc networks(VANETs), adapting it to work on cloud-based vehicular network requires some changes to address particular computation features such as task computation of applications of cloud-based vehicular networks. There has been initial work on briging SDN concepts to vehicular networks to reduce the latency by using the fog computing technology, but most of these studies do not directly tackle the issue of task computation. This paper proposes a Software-Defined Cloud-based vehicular Network called SDCVN framework. In this framework, we study the effectiveness of task computation of applications of cloud-based vehicular networks with vehicular cloud and roadside edge cloud. Considering the edge cloud service migration due to the vehicle mobility, we present an efficient roadside cloud based controller entity scheme where the tasks are adaptively computed through vehicular cloud mode or roadside computing predictive trajectory decision mode. Simulation results show that our proposal demonstrates a stable and low route setup time in case of installing the forwarding rules of the routing applications because the source node needs to contact the controller once to setup the route.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2018년도 춘계학술대회
• /
• pp.238-240
• /
• 2018

Cloud vehicular networks are a promising paradigm to improve vehicular through distributing computation tasks between remote clouds and local vehicular terminals. Software-Defined Network(SDN) can bring advantages to Intelligent Transportation System(ITS) through its ability to provide flexibility and programmability through a logically centralized controlled cluster that has a full comprehension of view of the network. However, as the SDN paradigm is currently studied in vehicular ad hoc networks(VANETs), adapting it to work on cloud-based vehicular network requires some changes to address particular computation features such as task computation of applications of cloud-based vehicular networks. There has been initial work on briging SDN concepts to vehicular networks to reduce the latency by using the fog computing technology, but most of these studies do not directly tackle the issue of task computation. This paper proposes a Software-Defined Cloud-based vehicular Network called SDCVN framework. In this framework, we study the effectiveness of task computation of applications of cloud-based vehicular networks with vehicular cloud and roadside edge cloud. Considering the edge cloud service migration due to the vehicle mobility, we present an efficient roadside cloud based controller entity scheme where the tasks are adaptively computed through vehicular cloud mode or roadside computing predictive trajectory decision mode. Simulation results show that our proposal demonstrates a stable and low route setup time in case of installing the forwarding rules of the routing applications because the source node needs to contact the controller once to setup the route.

• 한국환경과학회지
• /
• 제32권12호
• /
• pp.899-914
• /
• 2023

Various seeding materials for cloud seeding are being used, and sodium chloride powder is one of them, which is commonly used. This study analyzed the experimental results of multi-aircraft cloud seeding in connection with Republic of Korea Air Force (CN235) and KMA/NIMS(Korea Meteorological Administration/National Institute of Meteorological Sciences) Atmospheric Research Aircraft. Powdered sodium chloride was used in CN235 for the first time in South Korea. The analysis of the cloud particle size distributions and radar reflectivity before and after cloud seeding showed that the growth efficiency of powdery seeding material in the cloud is slightly higher than that of hygroscopic flare composition in the distribution of number concentrations by cloud aerosol particle diameter (10 ~ 1000 ㎛). Considering the radar reflectivity, precipitation, and numerical model simulation, the enhanced precipitation due to cloud seeding was calculated to be a maximum of 3.7 mm for 6 hours. The simulated seeding effect area was about 3,695 km², which corresponds to 13,634,550 tons of water. In the precipitation component analysis, as a direct verification method, the ion equivalent concentrations (Na⁺, Cl^-, Ca²⁺) of the seeding material at the Bukgangneung site were found to be about 1000 times higher than those of other non-affected areas between about 1 and 2 hours after seeding. This study suggests the possibility of continuous multi-aircraft cloud seeding experiments to accumulate and increase the amount of precipitation enhancement.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권6호
• /
• pp.2035-2049
• /
• 2015

Applications deployed in cloud receive a huge volume of requests from geographically distributed users whose satisfaction with the cloud service is directly proportional to the efficiency with which the requests are handled. Assignment of user requests based on appropriate load balancing principles significantly improves the performance of such cloud-based applications. To study the behavior of such systems, there is a need for simulation tools that will help the designer to set a test bed and evaluate the performance of the system by experimenting with different load balancing principles. In this paper, a novel architecture for cloud called Request Assignment Simulator (RAS) is proposed. It is a customizable, visual tool that simulates the request assignment process based on load balancing principles with a set of parameters that impact resource utilization. This simulator will help to ascertain the best possible resource allocation technique by facilitating the designer to apply and test different load balancing principles for a given scenario.

• 한국IT서비스학회지
• /
• 제13권3호
• /
• pp.201-219
• /
• 2014

This paper considers service deployment and priority optimization for multiple service-oriented applications sharing reusable services, which are deployed as multiple instances in the cloud. In order to handle variations in the workloads of the multiple applications, service instances of the individual reusable services are dynamically provisioned in the cloud. Also service priorities for each application in a particular reusable service are dynamically adjusted. In this paper, we propose an analytic performance model, based on a queueing network model, to predict the expected sojourn times of multiple service-oriented applications, given the number of service instances and priority disciplines in individual reusable services. We also propose a simple heuristic algorithm to search an optimal number of service instances in the cloud and service priority disciplines for each application in individual reusable services. A numerical example is also presented to demonstrate the applicability of the proposed performance model and algorithm to the proposed optimal decision problem.

• 정보처리학회논문지A
• /
• 제18A권1호
• /
• pp.11-18
• /
• 2011

클라우드 컴퓨팅은 자원 공유를 위한 새로운 컴퓨팅 패러다임이다. 클라우드 서비스를 위해 사용 하는 다양한 어플리케이션들은 워크플로우들로 표현된다. 이러한 워크플로우 어플리케이션은 클라우드의 자원 또는 서비스들에 적절하게 할당되어야 한다. 본 논문에서는 클라우드 컴퓨팅 환경을 고려하여 다중 워크플로우 어플리케이션을 위한 새로운 스케줄링 알고리즘을 제안한다. 제안하는 알고리즘은 비용 기반 랭크 스케줄링 알고리즘으로서 다중 워크플로우 어플리케이션을 고려할 뿐만 아니라, 서비스 평가를 위한 다양한 QoS 메트릭을 고려한다. 실험결과에서 제안한 알고리즘은 다른 알고리즘들에 비해 평균 총 처리시간과 평균 가용성에서 향상된 결과를 보였다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권3호
• /
• pp.1141-1163
• /
• 2019

In an effort to minimize operational expenses and supply users with more scalable services, distributed applications are actually going towards the Cloud. These applications, sent out over multiple environments and machines, are composed by inter-connecting independently developed services and components. The implementation of such programs on the Cloud is difficult and generally carried out either by hand or perhaps by composing personalized scripts. This is extremely error prone plus it has been found that misconfiguration may be the root of huge mistakes. We introduce AutoBot, a flexible platform for modeling, installing and (re)configuring complex distributed cloud-based applications which evolve dynamically in time. AutoBot includes three modules: A simple and new model describing the configuration properties and interdependencies of components; a dynamic protocol for the deployment and configuration ensuring appropriate resolution of these interdependencies; a runtime system that guarantee the proper configuration of the program on many virtual machines and, if necessary, the reconfiguration of the deployed system. This reduces the manual application deployment process that is monotonous and prone to errors. Some validation experiments were conducted on AutoBot in order to ensure that the proposed system works as expected. We also discuss the opportunity of reusing the platform in the transition of applications from Cloud to Fog computing.