Guest editorial: Parallel and Distributed Computing, Applications and Technologies

Parallel and distributed computing has been under many years of development, coupling with different research and application trends such as grid computing, cloud computing, green computing, etc. Nowadays the theory, design, analysis, evaluation and application of parallel and distributed computing systems are still burgeoning, to suit the increasing requirements for high efficiency and energy saving in global economy. The objective of this special section is to publish and overview recent trends in the interdisciplinary area of parallel and distributed computing, applications and technologies. This special section is based on the presentations at the 13th International Conference on Parallel and Distributed Computing, Applications and Technologies (PDCAT 2012).

Manamiary Bruno Andriamiarina, Dominique Méry and Neeraj Kumar Singh in their paper “Revisiting Snapshot Algorithms by Refinement-based Techniques” demonstrate semantical relationships existing between various snapshot algorithms with the refinement of models. They employ an abstract initial specification of the snapshot problem and enrich this specification gradually by a progressive and incremental refinement. They also propose an architecture based on the correction-by-construction paradigm, for conceiving algorithms dedicated to observation of global states of distributed systems.

Xiuqiao Li, Limin Xiao, Ke Xie, Bin Dong, Li Ruan and Dongmei Liu in their paper “Towards Hybrid Client-side Cache Management in Network-based File Systems” propose a hybrid client-side caching scheme (HCCache) to avoid performance degradation caused by the block interleaving problem and increase the cache space efficiency by customizing content addressable levels for files with different sizes. Two metrics of cache efficiency are also defined to correctly reflect cache performance. The simulation results demonstrate that HCCache can improve the I/O performance of small files in terms of bandwidth and latency compared with the block-indexed cache.

Yining Zhao and Alan Wood in their paper “Duplication Problem in Treaty Systems: Causes and Solutions” provide the formal definitions of treaty operations for access control in distributed environments, and discuss the causes of the duplication problem, and how treaty operations can affect this. They also propose three models of treaty systems that aim to solve the duplication problem, and evaluate their performance and scalability.

Hui Tian, Yingpeng Sang, Hong Shen and Chunyue Zhou in their paper “Probability-Model based Network Traffic Matrix Estimation” propose firstly a simple probability model to estimate the traffic matrix for a large-scale network, and then generalize the model by including random traffic data. Traffic matrix estimation is then conducted under these two models by two minimization methods. They also make experiments to show that the Normalized Root Mean Square Errors of these estimates under their model assumption are very small, and for a large-scale network, the traffic matrix estimation methods also perform well.

We sincerely thank all the reviewers and authors for their contributions to this special section. Special thanks are given to Prof. Mirjana Ivanović, the EiC of ComSIS, for providing us the chance for publishing this special section and valuable comments and support in the whole process.

Guest Editors
Hong Shen, Yingpeng Sang