• KIISE Transactions on Computing Practices
• /
• v.22 no.11
• /
• pp.547-558
• /
• 2016

Construction of multi-site Cloud Infrastructure with OpenStack, the most popular cloud open source project with various ICT infrastructure technologies, has operational inefficiency due to installation complexity and geographical limitation. To address this inefficiency, we gradually developed automated installation/configuration tools to automate installation/configuration of Linux and OpenStack by complying with DevOps methodology over a multi-site OpenStack testbed called OF@KOREN SmartX Playground. However, the pre-developed installation tools can be used only for limited cases. Therefore, we designed and developed a prototype of SmartX Provisioning Framework which could conducts Playground-wide provisioning flexibly by following three steps of resource management in Software-defined Infrastructure. We verified the efficiency of the provisioning functionality of the framework by demonstrating an example of automated multi-site cloud construction.

• The KIPS Transactions:PartA
• /
• v.18A no.4
• /
• pp.123-128
• /
• 2011

As the global grid has grown in size, large-scale distributed data analysis schemes have gained momentum. Over the last few years, a number of methods have been introduced for allocating data intensive tasks across distributed and heterogeneous computing platforms. However, these approaches have a limited potential for scaling up computing nodes so that they can serve more tasks simultaneously. This paper tackles the scalability and communication delay for computing nodes. We propose a distributed data node for storing and allocating the data. This paper also provides data provisioning method based on the steady states for minimizing the communication delay between the data source and the computing nodes. The experimental results show that scalability and communication delay can be achieved in our system.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.11
• /
• pp.5271-5289
• /
• 2016

On large-scale data analysis platforms deployed on cloud infrastructures over the Internet, the instability of the data transfer time and the dynamics of the processing rate require a more sophisticated data distribution scheme which maximizes parallel efficiency by achieving the balanced load among participated computing elements and by eliminating the idle time of each computing element. In particular, under the constraints that have the real-time and limited data buffer (in-memory storage) are given, it needs more controllable mechanism to prevent both the overflow and the underflow of the finite buffer. In this paper, we propose an auto regulated data provisioning model based on receiver-driven data pull model. On this model, we provide a synchronized data replenishment mechanism that implicitly avoids the data buffer overflow as well as explicitly regulates the data buffer underflow by adequately adjusting the buffer resilience. To estimate the optimal size of buffer resilience, we exploits an adaptive buffer resilience control scheme that minimizes both data buffer space and idle time of the processing elements based on directly measured sample path analysis. The simulation results show that the proposed scheme provides allowable approximation compared to the numerical results. Also, it is suitably efficient to apply for such a dynamic environment that cannot postulate the stochastic characteristic for the data transfer time, the data processing rate, or even an environment where the fluctuation of the both is presented.

• Journal of KIISE:Computing Practices and Letters
• /
• v.15 no.8
• /
• pp.525-533
• /
• 2009

Staging technique is used to provide files for a job in the Grid. If a staged file has large volume, the start time of the job is delayed and the throughput of job in the Grid may decrease. Therefore, removal of staging overhead helps the Grid operate more efficiently. In this paper, we present two methods for efficient file provisioning to clear the overhead. First, we propose RA-RFT, which extends RFT of Globus Toolkit and enables it to utilize RLS with replica information. RA-RFT can reduce file transfer time by doing partial transfer for each replica in parallel. Second, we suggest Remote Link that uses remote I/O instead of file transfer. Remote link is able to save storage of computational nodes and enables fast file provisioning via prefetching. Through various experiments, we argue that our two methods have an advantage over existing staging techniques.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.12
• /
• pp.5701-5722
• /
• 2018

The continuous increase in the cost of energy production and concerns for environmental sustainability are leading research communities, governments and industries to amass efforts to reduce energy consumption and global $CO_2$ footprint. Players in the information and communication industry are keen on reducing the operational expenditures (OpEx) and maintaining the profitability of cellular networks. Meanwhile, network virtualization has been proposed in this regard as the main enabler for 5G mobile cellular networks. In this paper, we propose a generic framework of slice resource provisioning and customized physical resource allocation for energy-efficiency and quality of service optimization. In resource slicing, we consider user demand and population resources provisioning scheme aiming to satisfy quality of service (QoS). In customized physical resource allocation, we formulate this problem with an integer non-linear programming model, which is solved by a heuristic algorithm based on minimum vertex coverage. The proposed algorithm is compared with the existing approaches, without consideration of slice resource constraints via system-level simulations. From the perspective of infrastructure providers, traffic is scheduled over a limited number of active small-cell base stations (sc-BSs) that significantly reduce the system energy consumption and improve the system's spectral efficiency. From the perspective of virtual network operators and mobile users, the proposed approach can guarantee QoS for mobile users and improve user satisfaction.

• Journal of KIISE
• /
• v.44 no.5
• /
• pp.439-448
• /
• 2017

The workloads of large high-performance computing (HPC) scientific applications are steadily becoming "bursty" due to variable resource demands throughout their execution life-cycles. However, the over-provisioning of virtual resources for optimal performance during execution remains a key challenge in the scheduling of scientific HPC applications. While over-provisioning of virtual resources guarantees peak performance of scientific application in virtualized environments, it results in increased amounts of idle resources that are unavailable for use by other applications. Herein, we proposed a memory resource reconfiguration approach that allows the quick release of idle memory resources for new applications in OS-level virtualized systems, based on the applications resource-usage pattern profile data. We deployed a scientific workflow application in Docker, a light-weight OS-level virtualized system. In the proposed approach, memory allocation is fine-tuned to containers at each stage of the workflows execution life-cycle. Thus, overall memory resource utilization is improved.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.16 no.2
• /
• pp.35-43
• /
• 2020

Cloud computing integrates computing resources such as servers, storage, and networks with virtualization technology to provide suitable services according to user needs. Due to the structural characteristics of sharing physical resources based on virtualization technology, threats to availability can occur, so it is essential to respond to availability threats in cloud computing. Existing over-provisioning method is not suitable because it can generate idle resources and cause under-provisioning to degrade or disconnect service. System resources must be allocated in real-time according to the system load to guarantee the cloud system's availability. Through appropriate management measures, it is necessary to reduce the system load and increase the performance of the system. This paper analyzes the work response time according to the allocation or migration of virtual machines and discusses an efficient resource management method considering the system load.

• Journal of information and communication convergence engineering
• /
• v.12 no.1
• /
• pp.14-18
• /
• 2014

A recent major development in computer technology is the advent of wearable computer systems. Wearable computer systems employ a wireless universal serial bus (WUSB), which refers to a combination of USB with the WiMedia wireless technical specifications. In this study, we focus on an integrated system of WUSB over wireless body area networks (WBANs) for wireless wearable computer systems. However, current WBAN MACs do not have well-defined quality of service (QoS) mapping and resource allocation mechanisms to support multimedia streams with the requested QoS parameters. To solve this problem, we propose a novel QoS-aware time slot allocation method. The proposed method provides fair and adaptive QoS provisioning to isochronous streams according to current traffic loads and their requested QoS parameters by executing a QoS satisfaction algorithm at the WUSB/WBAN host. The simulation results show that the proposed method improves the efficiency of time slot utilization while maximizing QoS provisioning.

• The Journal of Information Systems
• /
• v.14 no.3
• /
• pp.9-16
• /
• 2005

In this paper, the performance of the WebService architecture for QoS guaranteed connection provisioning in inter-AS domain networks has been measured and analyzed for service publish/inquiry, collection of NMSs ASBR details, source routing by ingress NMS in constraint based routing and connection establishment. From the analysis, it has been found that, the connection between inter-AS domain networks can be established within the usual time limits of 3 seconds by the Web Service architecture. Since no standard solutions have been implemented in Interdomain QoS provisioning, this performance analysis assures Web Service architecture as a promising solution and can be easily implemented in the early stages of MPLS network employment.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2002.11a
• /
• pp.844-848
• /
• 2002

In this paper, we present a VoIP service provisioning architecture based on MEGACO/H.248 which is one of the key protocols for VoIP services. MEGACO/H.248 is a media gateway control protocol standardized by both ITU-T and IETF, and many ITSPs, carriers, and vendors currently have a lot of interest in the protocol. MEGACO/H.248 is used by a softswitch a key component of the next generation VoIP network, in order to control various media gateways and provide seamless interworking between PSTN and Yon networks.