Byeongsik Kim

Queue length and waiting time analysis of a batch arrival queue with bilevel control

Abstract Scope and Purpose —This paper concerns the analysis of a production system in which an early set-up and N -policy are employed. We model the system by M x /M/1 queue with bilevel control and obtain the queue length and waiting time distributions. In general, with more than two control policies mixed up, obtaining the waiting time distribution is extremely complicated due to the dependencies between the random variables involved. The methodology employed in this paper can be extended to other batch arrival queueing system with more complicated control policies. We consider an M x /M /1 queue with bilevel control. As soon as the system empties, the server becomes idle until m or more customers accumulate, at which time the server starts a set-up. After the set-up, if N or more customers are in the system ( N ⩾ m ) the server immediately start serving the customers in FCFS fashion. If less than N customers are in the system after the set-up, the server remains dormant in the system and wait until N or more customers accumulate. We derive the distributions of the number of customers in the system and the waiting time.

Multi-path utility maximization and multi-path TCP design

The canonical multi-path network utility maximization (NUM) model which is extended directly from the single-path NUM has been studied widely in the literature. Most of the previous approaches do not specify the case of subflows on paths with different characteristics. Moreover, the transport protocol derived from the canonical multi-path NUM exhibits flappiness in the subflows because of the non-strictly convexity of the optimization problem.This paper introduces a modified multi-path NUM model and proposes a novel approach to overcome the mentioned issues. Using Jensens inequality, the multi-path NUM is approximated to a strictly convex and separable problem which can be solved efficiently by dual-based decomposition method. The algorithm successively solving a sequence of approximation problems is proven to converge at the global optimum of the original problem. Moreover, considering the separable form of the approximation utility and the dual-based nature of the proposed algorithm, the reverse engineering frameworks of the current TCPs are used to develop a series of multi-path TCPs that are compatible with corresponding regular single-path TCPs. Propose a modified multi-path NUM model.Solve it by successive approximation approach in fluid model.Design a series of multi-path TCPs.Conduct experiments on mReno and compare it to the current MPTCP.

mReno: a practical multipath congestion control for communication networks

In current networks, end-user devices are usually equipped with several network interfaces. The design of a multipath protocol that can cooperate with current single-path transport protocols is an interesting research field. Most previous works on multipath network utility maximization (NUM) lead to rate-based control protocols. Moreover, these studies do not model a case in which paths from a source may have different characteristics. Thus, these approaches are difficult to deploy to the Internet. In this paper, we introduce a multipath NUM model for a network with both multipath and single-path users. The proposed algorithm converges to a global solution to the multipath NUM. Based on the mathematical framework, we develop a multipath TCP called mReno. Analysis and simulations indicate that mReno is completely compatible with TCP Reno and achieves load-balance, fairness, and performance improvement targets.

An evaluation of federated cloud computing effect with service level

In this paper, we consider a federated cloud computing effect. Cloud computing provides software, platform, and infrastructure as services. Customers pay for the services in a pay-as-you-go model. As an effort of quantifying the effect of cloud computing, this paper evaluates a federated cloud computing network. For the evaluation we studied the simple deterministic demand model and the simple probabilistic demand model. In both cases, the effects of online migration of virtual machines can be clearly identified. Specially, for the probabilistic model we define service level as the proportion of time that the capacity of a data center satisfies the amount of the demand. With the service level concept, the performance of the federation and non-federation can be compared. The models in the paper are basic quantitative ones which can be extended to several other complex models with financial effects.

Integrated management system for distributed micro-datacenters

In this paper, we introduce a Micro-Datacenter (MDC) and Integrated Management System (IMS) that controls and manages the MDCs for supporting several services to users. The MDC is a small and rack-based datacentre system that designed to solve different problems or to handle different workloads that cannot be handled by traditional datacenters. The IMS controls and manages a multitude of MDCs to provide users with several services. And also, we introduce a case study applied to the Korea Advanced Research Network (KOREN).

Implementation of inter-domain path computation element

The Path Computation Element (PCE) Working Group (WG) at the Internet Engineering Task Force (IETF) is chartered to specify a PCE-based architecture for the path computation of inter-domain MPLS/GMPLS-based traffic engineered Label Switched Paths (LSPs). The PCE architecture introduces a special computational entity that will corporate with similar entities to compute the best possible path through single or multiple domains. This paper presents the state-of-the-art in PCE and several topics (e.g. PCE discovery mechanism, inter-domain path computation). In this paper, we describe the details about our PCE implementation.

Content searching scheme using interesting keyword based overlay network

Content-Centric Network (CCN) sends data to avoid duplicated transmission with content name in the network. However, CCN may cause increase in overhead while maintaining contents management table and create storage problem while storing contents information in the node. In this paper, we proposed a new searching scheme using interesting keyword and manage contents table via overlay network. The content providers with high frequency are chosen in the network. Then an overlay network is constructed using content providers. Via overlay network, it is possible to reduce the number of message transmissions and overhead to manage content management table in the network.

PCE implementation and testing by using virtual machines

In this paper, we describe the way to implement and test a PCE on the virtualized environment. The PCE is a special computational entity that will corporate with similar entities to compute the best possible path through single or multiple domains. Generally, in real network, the testing and implementation of network-related works are very difficult and complex but those works are much easier on the virtualized environment using virtual machines and networks. In this paper, we introduce the result of PCE implementation and test environment by using virtual machines and networks of VMware hypervisor, VMware ESX Server.

Routing information collector for Path Computation Element

In this paper, we present the design and the implementation of OSPF-TE routing information collector for MPLS and GMPLS networks. OSPF is commonly used as an IGP in MPLS and GMPLS networks to determine the topology over which LSPs can be established. TE extensions have been recently enabled in OSPF to support SPF tree calculation upon different purposes, thus possibly achieving optimal path computation and helping improve resource utilization efficiency. The IETF has standardized the PCE-based architecture, a good candidate to perform multidomain path computation. In such an architecture, the PCE needs the TED to compute the paths requested by a PCC. In this paper, we explain the approaches of using a OSPF-TE for the building and maintenance of the TED.

Bandwidth broker signaling for service level negotiation over heterogeneous IPv4/IPv6 diffserv networks

The Differentiated Services (DiffServ) is a Quality of Service (QoS) architecture developed by IETF. The basic assumption of this architecture is to achieve scalability in the core. A limited number of services provided by the network and a simplified architecture of the core nodes are key factors of DiffServ network scalability. At interdomain boundaries, Service Level Agreements (SLAs) specify the transit service to be given to each aggregate. In general, SLAs are complex business-related Contracts that cover a wide range of issues, including network availability guarantees, payment models, and other legal and business necessities. Bandwidth Broker (BB) is responsible for ensuring that resources within the DiffServ domain and on links connecting adjacent domains are properly provisioned and not oversubscribed. The BB is conceptually a policy management entity located in a policy-enabled domain. It negotiates SLAs with neighboring domains and ensures compliance of the administered domain with the SLAs contracted, We have defined and implemented a new interdomain protocol (a modification of COPS-SLS for the support of DifiServ over IPv4/IPv6 networks) called a BB-SLS as a service level negotiation protocol. And also, a testbed implementation of the proposed solutions is finally described.