Tomoya Enokido

A flexible group communication protocol with hybrid clocks

In distributed applications, a group of multiple peer processes are required to cooperate with each other. Messages sent by peer processes have to be delivered in some order like causal order and total order. In this paper, we discuss a scalable group communication protocol for a group of multiple peers in peer-to-peer (P2P) overlay networks. Due to the message overhead O(n) for the number n of peer processes, the vector clock cannot be used to causally deliver messages. In this paper, we consider a group where every member peer process can use a physical clock which is synchronized with GPS time server in the network time protocol (NTP). The physical clock can be used to temporally order messages while the linear clock can be used to causally order messages. Even if each physical clock is synchronized with a time server, every physical clock does not show the same accurate time. The accuracy of the physical clock depends on distance, i.e. number of routers and traffic in a network. We discuss how the physical and linear types of clocks can be used to temporally and causally order messages in terms of delay time between peer processes and accuracy of each physical clock. In this paper, we discuss a group communication protocol to deliver messages by taking usage of physical clock and linear clock.

Laxity-based process allocation algorithms to reduce power consumptions in peer-to-peer overlay networks

Information systems are composed of various types of computers interconnected in networks. In addition, information systems are being shifted from the traditional client-server model to the peer-to-peer (P2P) model. The P2P systems are scalable and fully distributed without any centralized coordination. It is getting more significant to discuss how to reduce the total electric power consumption of computers in information systems in addition to developing distributed algorithms to minimize the computation time. In this paper, we do not discuss the micro level like the hardware specification of each computer. We discuss a model to show the relation of the computation and the total power consumption of multiple peer computers to perform Web types of processes at macro level. We also discuss algorithms for allocating a process to a computer so that the deadline constraint is satisfied and the total power consumption is reduced.

The Improved Transmission Energy Consumption Laxity-Based Algorithm with Parallel Data Transmission

Various kinds of approaches are proposed to reduce the total electric energy consumption of a server cluster system equipped with virtual machines. In order to design and implement energy-efficient server cluster systems, processing load of virtual machines to perform application processes has to be balanced with one another to not only reduce the total electric energy consumption of a server cluster but also satisfy application’s performance requirements like turn-around time for each client. In this paper, the ITECLB-P (improved transmission energy consumption laxity based algorithm with parallel data transmission) algorithm is proposed to allocate communication processes to virtual machines in a server cluster so that the total electric energy consumption of a server cluster and the average turn-around time of each client can be reduced. In the evaluation, we show the total electric energy consumption of a server cluster and the average turn-around time of each client can be reduced in the ITECLB-P algorithm compared with the improved transmission energy consumption laxity based (ITECLB) algorithm.

Group Migration of Virtual Machines in a Server Cluster

In this paper, we propose an algorithm to migrate a group of virtual machines among servers in a cluster so that the total electric energy consumption of the host and guest servers can be reduced. We propose a pair of static SM(v) and dynamic DM(v) algorithms where at most v (\(\ge 0\)) virtual machines migrate from a host server to a guest server. In the DM(v) algorithm, virtual machines are dynamically resumed and suspended as the number of application processes increases and decreases. In the evaluation, we show the total electric energy consumption of servers can be reduced in the DM(v) algorithm compared with other algorithms.

Evaluation of Object-Based Information Flow Control in P2PPS Systems

In the P2PPSO (P2P (peer-to-peer) type of topic-based PS (publish/subscribe) with object concept) model, each peer process (peer) exchanges objects by publishing and subscribing event messages which are characterized by topics with no centralized coordinator. Information in a peer \(p_i\) illegally flows to a peer \(p_j\) if an event message \(e_i\) published by the peer \(p_i\) carries an object on some topics into the target peer \(p_j\), which is not allowed to subscribe the topics. In our previous studies, a TOBS (topics of objects-based synchronization) protocol is proposed to prevent illegal information flow in the P2PPSO system. In the TOBS protocol, an illegal object whose topics a target peer is not allowed to subscribe is not delivered to the target peer. In this paper, we evaluate the TOBS protocol in terms of the number of event messages including at least one illegal object. In the evaluation, we show the number of event messages including illegal objects in the TOBS protocol.

A Synchronization Protocol to Prevent Illegal Information Flow Based on Maximal Roles in the Role-Based Access Control Model

In access control models, a transaction is allowed to read and write an object only if access rights to read and write the object are granted, respectively. Suppose a transaction \(T_1\) reads data d from a file object \(f_1\) and then writes the data d to another file object \(f_2\). Here, another transaction \(T_2\) can get the data d by reading the file object \(f_2\) even if \(T_2\) is not granted a read right on the file object \(f_1\). Here, information in the file object \(f_1\) flows to the transaction \(T_2\) via the file object \(f_2\). We have to prevent illegal information flow to occur by transactions manipulating objects. The role-based access control (RBAC) model is widely used in various applications like database systems. In our previous studies, the legally precedent relation from a role to a role is defined. However, even if the legal condition is satisfied, there is case illegal information flow occurs. In this paper, we redefine legal and illegal precedent relations among roles. In order to check if a collection A of roles illegally precedes a collection B of roles, we introduce a new condition which uses maximal roles of A and B.

A Multi-layered Communication Model with Hybrid Clocks in P2P Overlay Networks

A group of multiple peers are required to cooperate with each other in distributed peer-to-peer (P2P) applications. Here, messages have to be causally delivered to every peer. In our approach to realizing a scalable P2P group, messages are ordered by taking advantage of linear time (LT) and physical time (PT) since message length is O(1). Here, each peer has to hold information on clock accuracy of every peer and minimum delay time between every pair of peers. In a scalable group, it is difficult, maybe impossible for each peer to hold the information on every peer due to storage overhead. It is also difficult for each peer to directly send messages to every peer due to communication overheads. In this paper, we newly discuss a multi-layered model to realize a scalable P2P group. A group is hierarchically decomposed into subgroups, where every peer is in a leaf peer. We discuss how each peer to order messages by taking advantage of PT and LT in a multilayered group. We evaluate the multi-layered group protocol in terms of the size of group information and delay time among peers compared with a flat group.

Flexible group protocol for distributed systems

The group protocol has to support applications with enough quality (QoS) of service in change of QoS supported by the underlying network and QoS requirements of applications. A flexible group service is supported for applications by cooperation of multiple autonomous agents. Each agent dynamically and autonomously takes a class of each protocol function like retransmission in change of QoS supported by networks and required by applications.