USE OF HIGH PERFORMANCE COMPUTING TECHNOLOGIES AND SCRIPT RUN MEDIATOR MIDDLEWARE FOR EDUCATIONAL PROCESS IN LIEPAJA UNIVERSITY

Abstract. In this paper the authors describe a solution for implementation of the SRM (ScriptRunMediatorCSP) middleware to resolve trusting and secure connection problems to HPC (High Performance Computing) cluster resources in the Liepaja University. The main idea is to develop a specific middleware tool called SRM, which would allow connection to the parallel computing cluster system, namely to the main server masternode, to provide the easiest solution for using HPC technologies in educational processes. Cluster computing resources could be used for solution and analysis of various scientific problems applying mathematical methods and high performance technological solutions for learning and performing parallel computations in the educational process thus increasing the students' theoretical knowledge in practical applications. Keywords: cluster, HPC technologies, Lazarus, SRM middleware, parallel calculations, parallel compilers. Introduction Rapid development of High Performance Computing (HPC) systems in the field of IT is determined by the necessity to solve society and science problems more effectively, for instance, to optimize gas and oil research and acquisition, to forecast climate changes, and also to introduce innovation products and services in industrial and financial fields. Therefore, it is important to provide the HPC technological environment, which is implemented with the help of parallel data processing and computation technologies, applications of cluster networks and information technologies, in order to promote involvement of students as young experts to real problem solution in the field of scientific research. In the academic and research field, HPC utilizes multiple processors to perform concurrent tasks in parallel for solving large cutting-edge research computation problems. At this moment authors provide centralized high-performance computing resources and support to Liepaja University students and researchers in large-scale computing disciplines with the use of advanced computing cluster hardware infrastructure, software, tools and programming techniques.The goal of centralized supercomputing centres and conventional distributed computing systems, such as grids and clusters, and their support service is to enhance research capabilities and effectiveness. The current efforts in the grid computing research focus on a design of the new grid schedulers, that can efficiently optimize the standard scheduling objectives, such as makespan, flowtime and resource utilization, but also can fulfil the security requirements of the grid users and can minimize the energy consumed by all of the system components. These schedulers should also capture the complexity of the whole system and provide meaningful measures for a wide range of grid applications and services. Therefore, energy-efficient and security-aware scheduling in CGs (Computational Grids) becomes a complex research and engineering endeavour mainly due to different priorities and preferences of the grid users and resource owners Several parallel computation clusters with various configurations have been created and tested, where the special middleware SRM (ScriptRunMediatorCSP (ClusterSecurityPackage)) was used in order to implement solutions of direct, indirect and optimisation problems using data parallel processing technologies -computation automation software. Within the research a Beowulf cluster is used, and the operating system adopted is Linux, which is widely available; software tools for message passing such as MP