2020 21st Asia-Pacific Network Operations and Management Symposium (APNOMS), 2020
As the number of mobile devices increases and the concept of 5G networks becomes popular, various... more As the number of mobile devices increases and the concept of 5G networks becomes popular, various network infrastructures and services emerge. Also, more users request user-specific network services within limited network resources. Under this complex situation, in order to provide a guaranteed QoS level to users, it requires to consider various factors which affect the performance of network service, and dynamic network provisioning is required. In this paper, we propose a dynamic network provisioning system with AI-enabled path planning, which uses the side channel information such as disaster events, maintenance events, and distribution of users. In this system, we design a user request handler that understands user-specific QoS and spatio-temporal requirements in order to maximize the utilization of a given network resource. Also, this system utilizes side-channel information to optimizing the network provisioning in a realtime manner.
Abstract. The Web is rapidly increasing its reach beyond the desktop to various devices and the t... more Abstract. The Web is rapidly increasing its reach beyond the desktop to various devices and the transcoding proxy is appeared to support web services efficiently. Recently, the cooperative transcoding proxy archi-tecture is proposed to improve the system performance to cope with the scalability problem of a stand-alone transcoding proxy. However, because of the multiple versions, the communication protocol of the cooperative caches is very complex and causes additional delay to find best version for a requested object. In this paper, we propose efficient cooperative transcoding proxy archi-tecture which uses the content-aware caching. The main purpose of the proposed system is simplifying the communication protocol of cooper-ative caches. We associates a home proxy for each URL and the home proxy is responsible for transcoding and maintaining multiple version of an URL. This mechanism reduces the amount of messages exchanged and communication latency involved. To prevent the hot-spo...
In modern agriculture development, a huge amount of data is generated from many devices (e.g., se... more In modern agriculture development, a huge amount of data is generated from many devices (e.g., sensors, robotic drones, etc.) and services (e.g., weather service, market price service, etc.). Leveraging these multiple forms of data might provide many benefits to farmers and agribusiness. However, these massive data are often complex and heterogeneity (e.g., structured, semi-structured, and unstructured), so it is very difficult to comprehend them. Due to these factors, big data analytics and visualization techniques play an important role in human understanding. This paper describes how we propose that big data can be efficiently used for creating agriculture services and supporting decision-making systems. We first discuss the challenges in agriculture big data. We then present a comprehensive study related to the techniques for collecting, storing, analyzing, and visualizing of big data. We show that our proposed techniques can be used to answer important questions in agriculture ...
Evaluating Regular Path Queries (RPQs) have been of interest since they were used as a powerful w... more Evaluating Regular Path Queries (RPQs) have been of interest since they were used as a powerful way to explore paths and patterns in graph databases. Traditional automata-based approaches are restricted in the graph size and/or highly complex queries, which causes a high evaluation cost (e.g., memory space and response time) on large graphs. Recently, although using the approach based on the threshold rare label for large graphs has been achieving some success, they could not often guarantee the minimum searching cost. Alternatively, the Unit-Subquery Cost Matrix (USCM) has been studied and obtained the viability of the usage of subqueries. Nevertheless, this method has an issue, which is, it does not cumulate the cost among subqueries that causes the long response time on a large graph. In order to overcome this issue, this paper proposes a method for estimating joining cost of subqueries to accelerate the USCM based parallel evaluation of RPQs on a large graph, namely USCM-Join. T...
Image processing techniques play an increasingly important role in many aspects of our daily life... more Image processing techniques play an increasingly important role in many aspects of our daily life. For example, it has been shown to improve agricultural productivity in a number of ways such as plant pest detecting or fruit grading. However, massive quantities of images generated in real-time through multi-devices such as remote sensors during monitoring plant growth lead to the challenges of big data. Meanwhile, most current image processing systems are designed for small-scale and local computation, and they do not scale well to handle big data problems with their large requirements for computational resources and storage. In this paper, we have proposed an IPABigData (Image Processing Algorithm BigData) platform which provides algorithms to support large-scale image processing in agriculture based on Hadoop framework. Hadoop provides a parallel computation model MapReduce and Hadoop distributed file system (HDFS) module. It can also handle parallel pipelines, which are frequentl...
Expert system has played an important role in human’s life for a long time. In order to establish... more Expert system has played an important role in human’s life for a long time. In order to establish an expert system, knowledge acquisition is the process which should be considered first. Regarding to this step, two of the most demanding requests are: how to discover the knowledge from a massive dataset, and how to organize them in a temporary database properly in order to be reviewed again. In this paper, we propose a design of a subsystem, which acquires the knowledge from Big Data sources and stores them following the rule form. Our work is conducted with the support of a data mining technique called association rules mining, which is executed through parallel distributed environments: given data under file form as input, outputs which includes rules representing knowledge with their corresponding certainty factor are gotten. Then all of these rules are stored in a relational database, which could be queried out by knowledge engineers to review them again before encoding them into...
While web proxy caching is a widely deployed technique, the performance of a proxy cache is limit... more While web proxy caching is a widely deployed technique, the performance of a proxy cache is limited by the local storage. Some studies have addressed this limitation by using the residual resources of clients via a p2p method and have achieved a very high hit rate. However, these approaches treat web objects as homogeneous objects and there is no consideration of various web characteristics. Consequently, the byte hit rate of the system is limited, external bandwidth is wasted, and perceived user latency is increased. The present paper suggests an efficient p2p based web caching technique that manages objects with different policies so as to exploit the characteristics of web objects, such as size and temporal locality. Small objects are stored alone whereas large objects are stored by dividing them into numerous small blocks, which are distributed in clients. On a proxy cache, header blocks of large objects take the place of objects themselves and smaller objects are cached. This t...
Nowadays, Distributed Denial of Service (DDoS) attacks have gained increasing popularity and have... more Nowadays, Distributed Denial of Service (DDoS) attacks have gained increasing popularity and have been a major factor in a number of massive cyber-attacks. It could easily exhaust the computing and communicating resources of a victim within a short period of time. Therefore, we have to find the method to detect and prevent the DDoS attack. Recently, there have been some researches that provide the methods to resolve above problem, but it still gets some limitations such as low performance of detecting and preventing, scope of method, most of them just use on cloud server instead of network, and the reliability in the network. In this paper, we propose solutions for (1) handling multiple DDoS attacks from multiple IP address and (2) handling the suspicious attacks in the network. For the first solution, we assume that there are multiple attacks from many sources at a times, it should be handled to avoid the conflict when we setup the preventing rule to switches. In the other, there a...
Recently, during disasters occurrence, dealing with emergencies has been handled well by the earl... more Recently, during disasters occurrence, dealing with emergencies has been handled well by the early transmission of disaster relating notifications on social media networks (e.g., Twitter or Facebook). Intuitively, with their characteristics (e.g., real-time, mobility) and big communities whose users could be regarded as volunteers, social networks are proved to be a crucial role for disasters response. However, the amount of data transmitted during disasters is an obstacle for filtering informative messages; because the messages are diversity, large and very noise. This large volume of data could be seen as Social Big Data (SBD). In this paper, we proposed a big data platform for collecting and analyzing disaster data from SBD. Firstly, we designed a collecting module; which could rapidly extract disaster information from the Twitter; by big data frameworks supporting streaming data on distributed system; such as Kafka and Spark. Secondly, we developed an analyzing module which lear...
Recently, the software-defined networks (SDNs) are the most promising architecture which are main... more Recently, the software-defined networks (SDNs) are the most promising architecture which are mainly composed of the Openflow switches and the Flowvisor. The SDN provides virtualized separate networks as slices on a physical network according to the customer needs. The Flowvisor can provide customers to have their own controller and let the controller handle flow packets in its slice -a virtual network. In this paper we proposes a policy based slice embedding and have deployed a prototype in a virtualized testbed on KREONET - a Korean R&E network. The slice embedding system is designed as a part of aggregation management system to administer our testbed as interact with Flowvisor and slices can be automatically assigned by network resource provisioning with decision of Policy based Engineering (PBE) module in the system. Several points are considered to make policies for slicing so that slice embedding service for provisioning virtualized networks could be automatically deployable. K...
The 9th International Conference on Smart Media and Applications, 2020
Propositional satisfiability solving (SAT) has been one of the most successful automated reasonin... more Propositional satisfiability solving (SAT) has been one of the most successful automated reasoning methods in the last decade in computer science by solving a wide range of both industrial and academic problems. One of the most widely used constraint during the process of translating a practical problem into a SAT instance is the at-most-one (AMO) constraint. Many studies on SAT encodings of the AMO constraint have been reported in the literature, however thorough comparisons of these encodings are largely missing. This paper not only provide such comparisons, but also discusses the similarity of the SAT encodings of the AMO constraint and SAT encodings of more general CSPs. More specifically, the paper empirically evaluates the most well-known AMO encodings on three state-of-the-art SAT solvers of a number of benchmarks and provides some guidelines for efficient SAT encodings of the AMO constraint leading.
The 9th International Conference on Smart Media and Applications, 2020
Traffic on the Internet is growing exponentially around the globe, and SDN (Software Defined Netw... more Traffic on the Internet is growing exponentially around the globe, and SDN (Software Defined Network) becomes very attractive to network research communities. The trend switching from a hardware-centered network to a software-centric network allows that networks become more flexible and efficient. SDN is getting more and more attention, especially with the recent advent of cloud and 5G technology, and many researchers are making various attempts to improve SDN performance. Also, the application of artificial intelligence technology, which applied to multiple fields, has attracted many researchers, and a great deal of research is underway. Specifically, in many fields, including the network industry, the reinforcement learning technique has gained attention. The SDN field also reflected the recent increase in research that relates reinforcement learning. In this paper, we would briefly introduce SDN and reinforcement learning technology by identifying and analyzing recent studies which use reinforcement learning for efficient routing in SDN. Implementing RL techniques in SDN routing shows that enforcing routing to dynamic networks becomes more efficient, and it can provide excellent levels of QoS while optimizing resource utilization.
2021 22nd Asia-Pacific Network Operations and Management Symposium (APNOMS), 2021
Because of the activation of IoT (Internet of Things) devices due to the rapid development of rec... more Because of the activation of IoT (Internet of Things) devices due to the rapid development of recent communication technology, network traffic is currently fluctuating and increasing explosively. As existing network resource management policies are not sophisticated enough to cope with network conditions that change constantly, resource utilization can be lowered and costs can be higher. With the recent advances in deep learning techniques, network operators can manage networks intelligently. For the intelligent network, there is a technique which predict the state of network links. However, when the scale of the network increases, overall network management can be complicated. In addition, as the models of link state prediction are affected by the states of adjacent links, it is necessary to consider the spatio-temporal characteristics between links. In this paper, we propose a GCN(Graph Convolutional Neural Network)-GRU(Gated Recurrent Unit) based link state prediction technique. The proposed GCN-GRU model predicts network traffic by considering the spatio-temporal characteristics of each link state such as bandwidth, delay, and packet loss rate. Through extensive experiments on actual network traffic, the proposed GCN-GRU based link state prediction technique has shown to achieve 1.5% lower a mean absolute percentage error (MAPE) compared to a LSTM (Long Short term Memory) based link state prediction technique.
2020 21st Asia-Pacific Network Operations and Management Symposium (APNOMS), 2020
As the number of mobile devices increases and the concept of 5G networks becomes popular, various... more As the number of mobile devices increases and the concept of 5G networks becomes popular, various network infrastructures and services emerge. Also, more users request user-specific network services within limited network resources. Under this complex situation, in order to provide a guaranteed QoS level to users, it requires to consider various factors which affect the performance of network service, and dynamic network provisioning is required. In this paper, we propose a dynamic network provisioning system with AI-enabled path planning, which uses the side channel information such as disaster events, maintenance events, and distribution of users. In this system, we design a user request handler that understands user-specific QoS and spatio-temporal requirements in order to maximize the utilization of a given network resource. Also, this system utilizes side-channel information to optimizing the network provisioning in a realtime manner.
Abstract. The Web is rapidly increasing its reach beyond the desktop to various devices and the t... more Abstract. The Web is rapidly increasing its reach beyond the desktop to various devices and the transcoding proxy is appeared to support web services efficiently. Recently, the cooperative transcoding proxy archi-tecture is proposed to improve the system performance to cope with the scalability problem of a stand-alone transcoding proxy. However, because of the multiple versions, the communication protocol of the cooperative caches is very complex and causes additional delay to find best version for a requested object. In this paper, we propose efficient cooperative transcoding proxy archi-tecture which uses the content-aware caching. The main purpose of the proposed system is simplifying the communication protocol of cooper-ative caches. We associates a home proxy for each URL and the home proxy is responsible for transcoding and maintaining multiple version of an URL. This mechanism reduces the amount of messages exchanged and communication latency involved. To prevent the hot-spo...
In modern agriculture development, a huge amount of data is generated from many devices (e.g., se... more In modern agriculture development, a huge amount of data is generated from many devices (e.g., sensors, robotic drones, etc.) and services (e.g., weather service, market price service, etc.). Leveraging these multiple forms of data might provide many benefits to farmers and agribusiness. However, these massive data are often complex and heterogeneity (e.g., structured, semi-structured, and unstructured), so it is very difficult to comprehend them. Due to these factors, big data analytics and visualization techniques play an important role in human understanding. This paper describes how we propose that big data can be efficiently used for creating agriculture services and supporting decision-making systems. We first discuss the challenges in agriculture big data. We then present a comprehensive study related to the techniques for collecting, storing, analyzing, and visualizing of big data. We show that our proposed techniques can be used to answer important questions in agriculture ...
Evaluating Regular Path Queries (RPQs) have been of interest since they were used as a powerful w... more Evaluating Regular Path Queries (RPQs) have been of interest since they were used as a powerful way to explore paths and patterns in graph databases. Traditional automata-based approaches are restricted in the graph size and/or highly complex queries, which causes a high evaluation cost (e.g., memory space and response time) on large graphs. Recently, although using the approach based on the threshold rare label for large graphs has been achieving some success, they could not often guarantee the minimum searching cost. Alternatively, the Unit-Subquery Cost Matrix (USCM) has been studied and obtained the viability of the usage of subqueries. Nevertheless, this method has an issue, which is, it does not cumulate the cost among subqueries that causes the long response time on a large graph. In order to overcome this issue, this paper proposes a method for estimating joining cost of subqueries to accelerate the USCM based parallel evaluation of RPQs on a large graph, namely USCM-Join. T...
Image processing techniques play an increasingly important role in many aspects of our daily life... more Image processing techniques play an increasingly important role in many aspects of our daily life. For example, it has been shown to improve agricultural productivity in a number of ways such as plant pest detecting or fruit grading. However, massive quantities of images generated in real-time through multi-devices such as remote sensors during monitoring plant growth lead to the challenges of big data. Meanwhile, most current image processing systems are designed for small-scale and local computation, and they do not scale well to handle big data problems with their large requirements for computational resources and storage. In this paper, we have proposed an IPABigData (Image Processing Algorithm BigData) platform which provides algorithms to support large-scale image processing in agriculture based on Hadoop framework. Hadoop provides a parallel computation model MapReduce and Hadoop distributed file system (HDFS) module. It can also handle parallel pipelines, which are frequentl...
Expert system has played an important role in human’s life for a long time. In order to establish... more Expert system has played an important role in human’s life for a long time. In order to establish an expert system, knowledge acquisition is the process which should be considered first. Regarding to this step, two of the most demanding requests are: how to discover the knowledge from a massive dataset, and how to organize them in a temporary database properly in order to be reviewed again. In this paper, we propose a design of a subsystem, which acquires the knowledge from Big Data sources and stores them following the rule form. Our work is conducted with the support of a data mining technique called association rules mining, which is executed through parallel distributed environments: given data under file form as input, outputs which includes rules representing knowledge with their corresponding certainty factor are gotten. Then all of these rules are stored in a relational database, which could be queried out by knowledge engineers to review them again before encoding them into...
While web proxy caching is a widely deployed technique, the performance of a proxy cache is limit... more While web proxy caching is a widely deployed technique, the performance of a proxy cache is limited by the local storage. Some studies have addressed this limitation by using the residual resources of clients via a p2p method and have achieved a very high hit rate. However, these approaches treat web objects as homogeneous objects and there is no consideration of various web characteristics. Consequently, the byte hit rate of the system is limited, external bandwidth is wasted, and perceived user latency is increased. The present paper suggests an efficient p2p based web caching technique that manages objects with different policies so as to exploit the characteristics of web objects, such as size and temporal locality. Small objects are stored alone whereas large objects are stored by dividing them into numerous small blocks, which are distributed in clients. On a proxy cache, header blocks of large objects take the place of objects themselves and smaller objects are cached. This t...
Nowadays, Distributed Denial of Service (DDoS) attacks have gained increasing popularity and have... more Nowadays, Distributed Denial of Service (DDoS) attacks have gained increasing popularity and have been a major factor in a number of massive cyber-attacks. It could easily exhaust the computing and communicating resources of a victim within a short period of time. Therefore, we have to find the method to detect and prevent the DDoS attack. Recently, there have been some researches that provide the methods to resolve above problem, but it still gets some limitations such as low performance of detecting and preventing, scope of method, most of them just use on cloud server instead of network, and the reliability in the network. In this paper, we propose solutions for (1) handling multiple DDoS attacks from multiple IP address and (2) handling the suspicious attacks in the network. For the first solution, we assume that there are multiple attacks from many sources at a times, it should be handled to avoid the conflict when we setup the preventing rule to switches. In the other, there a...
Recently, during disasters occurrence, dealing with emergencies has been handled well by the earl... more Recently, during disasters occurrence, dealing with emergencies has been handled well by the early transmission of disaster relating notifications on social media networks (e.g., Twitter or Facebook). Intuitively, with their characteristics (e.g., real-time, mobility) and big communities whose users could be regarded as volunteers, social networks are proved to be a crucial role for disasters response. However, the amount of data transmitted during disasters is an obstacle for filtering informative messages; because the messages are diversity, large and very noise. This large volume of data could be seen as Social Big Data (SBD). In this paper, we proposed a big data platform for collecting and analyzing disaster data from SBD. Firstly, we designed a collecting module; which could rapidly extract disaster information from the Twitter; by big data frameworks supporting streaming data on distributed system; such as Kafka and Spark. Secondly, we developed an analyzing module which lear...
Recently, the software-defined networks (SDNs) are the most promising architecture which are main... more Recently, the software-defined networks (SDNs) are the most promising architecture which are mainly composed of the Openflow switches and the Flowvisor. The SDN provides virtualized separate networks as slices on a physical network according to the customer needs. The Flowvisor can provide customers to have their own controller and let the controller handle flow packets in its slice -a virtual network. In this paper we proposes a policy based slice embedding and have deployed a prototype in a virtualized testbed on KREONET - a Korean R&E network. The slice embedding system is designed as a part of aggregation management system to administer our testbed as interact with Flowvisor and slices can be automatically assigned by network resource provisioning with decision of Policy based Engineering (PBE) module in the system. Several points are considered to make policies for slicing so that slice embedding service for provisioning virtualized networks could be automatically deployable. K...
The 9th International Conference on Smart Media and Applications, 2020
Propositional satisfiability solving (SAT) has been one of the most successful automated reasonin... more Propositional satisfiability solving (SAT) has been one of the most successful automated reasoning methods in the last decade in computer science by solving a wide range of both industrial and academic problems. One of the most widely used constraint during the process of translating a practical problem into a SAT instance is the at-most-one (AMO) constraint. Many studies on SAT encodings of the AMO constraint have been reported in the literature, however thorough comparisons of these encodings are largely missing. This paper not only provide such comparisons, but also discusses the similarity of the SAT encodings of the AMO constraint and SAT encodings of more general CSPs. More specifically, the paper empirically evaluates the most well-known AMO encodings on three state-of-the-art SAT solvers of a number of benchmarks and provides some guidelines for efficient SAT encodings of the AMO constraint leading.
The 9th International Conference on Smart Media and Applications, 2020
Traffic on the Internet is growing exponentially around the globe, and SDN (Software Defined Netw... more Traffic on the Internet is growing exponentially around the globe, and SDN (Software Defined Network) becomes very attractive to network research communities. The trend switching from a hardware-centered network to a software-centric network allows that networks become more flexible and efficient. SDN is getting more and more attention, especially with the recent advent of cloud and 5G technology, and many researchers are making various attempts to improve SDN performance. Also, the application of artificial intelligence technology, which applied to multiple fields, has attracted many researchers, and a great deal of research is underway. Specifically, in many fields, including the network industry, the reinforcement learning technique has gained attention. The SDN field also reflected the recent increase in research that relates reinforcement learning. In this paper, we would briefly introduce SDN and reinforcement learning technology by identifying and analyzing recent studies which use reinforcement learning for efficient routing in SDN. Implementing RL techniques in SDN routing shows that enforcing routing to dynamic networks becomes more efficient, and it can provide excellent levels of QoS while optimizing resource utilization.
2021 22nd Asia-Pacific Network Operations and Management Symposium (APNOMS), 2021
Because of the activation of IoT (Internet of Things) devices due to the rapid development of rec... more Because of the activation of IoT (Internet of Things) devices due to the rapid development of recent communication technology, network traffic is currently fluctuating and increasing explosively. As existing network resource management policies are not sophisticated enough to cope with network conditions that change constantly, resource utilization can be lowered and costs can be higher. With the recent advances in deep learning techniques, network operators can manage networks intelligently. For the intelligent network, there is a technique which predict the state of network links. However, when the scale of the network increases, overall network management can be complicated. In addition, as the models of link state prediction are affected by the states of adjacent links, it is necessary to consider the spatio-temporal characteristics between links. In this paper, we propose a GCN(Graph Convolutional Neural Network)-GRU(Gated Recurrent Unit) based link state prediction technique. The proposed GCN-GRU model predicts network traffic by considering the spatio-temporal characteristics of each link state such as bandwidth, delay, and packet loss rate. Through extensive experiments on actual network traffic, the proposed GCN-GRU based link state prediction technique has shown to achieve 1.5% lower a mean absolute percentage error (MAPE) compared to a LSTM (Long Short term Memory) based link state prediction technique.
Uploads