Geunyoung Park

Using replication and checkpointing for reliable task management in computational Grids

In large-scale Grid computing environments, providing fault-tolerance is required for both scientific computation and file-sharing to increase their reliability. In previous works, several mechanisms were proposed for the Grids or distributed computing systems. However, some of them used only space redundancy (hardware replication), and others used only time redundancy (checkpointing and rollback). For this reason, the existing mechanisms are inefficient in terms of their resource utilization on the Grids. The main goal of ART is reducing the number of replications by using checkpointing and rollback scheme for each replication. In ART, the minimum number of replications is adaptively selected based on analysis of probability of successful execution within the given deadline and reliability requirement of each task. Our simulation results show that ART can significantly reduce the number of replications and improve scalability compared with existing mechanisms.

Energy aware routing with dynamic probability scaling

The goal of energy aware routing algorithms is to increase the lifetime and long-term connectivity of the wireless sensor networks. However, most of those algorithms do not use the newest states of nodes for retrieving routing information. In this paper, we propose an efficient energy-aware routing algorithm for wireless sensor networks. In our algorithm, the energy drain rate and residual energy of each sensor node are used for selecting candidate routes. Information is retrieved with energy awareness per almost every communication. Simulation results show that our approach outperforms the previous works with respect to long term connectivity by as much as 30%.

EAR-RT: energy aware routing with real-time guarantee for wireless sensor networks

Most energy aware routing algorithms focus on the increasing the lifetime and long-term connectivity of the wireless sensor networks. But energy efficiency sacrifices the communication delay between source and sink node. Therefore, many researchers have mentioned of energy and delay trade-offs. But the delay was not the main concern. In this paper we propose EAR-RT, an real-time guaranteed routing protocol for wireless sensor networks without harming energy awareness. Simulation results show that our real-time routing algorithm provides real-time guaranteed delivery while network is stable.

Energy aware routing based on adaptive clustering mechanism for wireless sensor networks

The main goal of research concerning energy aware routing algorithm for wireless sensor network is to increase the lifetime and long-term connectivity of the wireless sensor networks. However, most of energy aware routing algorithms do not take into account the clustering mechanism efficiently. In this paper, we present an efficient energy aware routing algorithm for the wireless sensor networks. In our algorithm, the data aggregation technique and adaptive clustering mechanism are considered for reducing and compacting the cumulative size of packets on the wireless sensor networks. Simulation results show that the energy usage of EAR-ACM is significantly reduced compared with the previous clustering based routing algorithm for the sensor networks.

Adaptive load balancing mechanism for server cluster

Server cluster provides high availability, scalability, and reliability by gathering server nodes into a group. Client requests need to be distributed to each server node fairly to maximize the performance of server cluster. In this paper, we propose an adaptive and efficient load balancing algorithm for the server cluster. The proposed algorithm computes the load of server nodes with the usages of computer resources and their weights. These weights are determined dynamically based on the statistics of the usages. The experimental result shows that the proposed algorithm can prevent the bottleneck of server cluster efficiently compared with existing algorithms. This guarantees its adaptability even though there are changes to the characteristic of service.

Adaptive link recovery scheme for structured peer-to-peer networks under churn

The problem of churn in structured peer-to-peer (P2P) networks is one of important research issues in this area. When churn, continuous coming and going of participating nodes, is arisen, search performance of structured P2Ps is degraded. Under high churn rate, network topology is congested, so link recovery is needed to maintain proper search performance. Generally periodic link recovery is used to recover broken links by churn. In periodic maintenance, link recovery period is important parameter. If it is too high or low compared with churn rate, it is difficult to guarantee consistent QoS. We propose an adaptive link recovery scheme to assure consistent search performance of structured P2Ps. In our scheme, churn rate is estimated by gossip protocol, and then the expected number of failed links, Nf, is derived from it. Each node makes Nf times of attempts to recover invalid links. Simulation results show that our scheme guarantees consistent QoS under various situations.

Impact on the writing granularity for incremental checkpointing

Incremental checkpointing is an cost-efficient fault tolerant technique for long running programs such as genetic algorithms. In this paper, we derive the equations for the writing granularity of incremental checkpointing and find factors associated with the time overhead and disk space for incremental checkpoint. We also verify the applicability of the derived equation and the acceptability of the factors through experiments.