Emir Imamagic

Grid infrastructure monitoring system based on Nagios

Monitoring in distributed environment such as a grid is crucial for normal operation of all subsystems. Constant gathering of information enables efficient security auditing, failure detection, maintenance, job scheduling, accounting, resource performance tuning, debugging, etc. In this paper we focus on monitoring of resources in the grid with the purpose of failure detection, notifications and automatic recovery. We introduce our system based on open source monitoring framework Nagios that achieves these functionalities. We describe grid specific features we implemented in order to achieve efficient grid monitoring system, namely sensors for various grid services, advanced sensor hierarchy and certificate-based authorization on web interface. Finally, we give overview of the implementation of our system for monitoring EGEE grid infrastructure.

CRO-GRID grid monitoring architecture

Grid is distributed, heterogeneous environment comprised of various resources. Monitoring of grid resources is essential for realizing functionalities such as failure detection, application performance analysis, debugging, execution management, etc. Today many mature monitoring systems exist, but there are no complete solutions and functionalities of existing systems often overlap. In this article, we propose methodology of integration of various existing systems in order to provide rich information base for applications, users and administrators. Furthermore, we extend this system with following functionalities: additional sensors, data archiving, integrated web interface and tools for installation, configuration and maintenance of monitoring components.

Grid data and replica management system

Computer grids are often used for conducting intensive scientific or technological calculations. Those calculations often require a specific mechanism for manipulating data. Besides demands for data manipulation built in the executions there is also the demand for storing data about data – location of data, accessibility and so on. This data, called metadata is also managed by the grids data management system. In this paper we give an overview of the grid data management problem. We then explain the demands specific for our CROGRID. Based on those specific demands we describe a system for data management which brings certain new concepts, such as combining forward reservation and dynamical replication.

Towards new energy efficiency limits of High Performance Clusters

In recent years performance of High Performance Computing Clusters took precedence over their power consumption. However, costs of energy and demand for ecologically acceptable IT solutions are higher than ever before, therefore a need for HPC clusters with acceptable power consumption becomes increasingly important. Consequently, the Green500 list, which takes into account both performance and power consumption of HPC clusters, almost reached the popularity of the Top500 list. Interestingly, the Green500 list is not an opponent to Top500 list; its core idea is to complement the Top500. Therefore, the Top500 list still serves as the basis for the Green500 list, and its numbers regarding measured HPL performance, are a basis for calculating the Green500 list. Indeed, the Green500 is the Top500 list ordered by HPL measured performance per Watt. Rmax numbers gained from High Performance Linpack benchmarks serve as performance input parameters, and total power consumed during execution of HPL on a certain HPC clusters is a power consumption parameter. The critical question remains: how to measure the consumed power correctly? This paper proposes that if it is not possible to measure the consumed power, one can still use maximum power consumption numbers rated from hardware vendors to find at least the lower bound green efficiency of HPC clusters. The main idea behind this approach is that Rmax values found on Top500 list never achieve Rpeak theoretical values, and that even most efficient HPL benchmark can never utilize computing nodes at their maximum. Furthermore by comparing MFLOPS/W results we gained with those found on Green500 list, we noted the excellent efficiency of the new HPC Isabella cluster recently powered on at University Computing Centre in Zagreb, ranking in just behind University of North Carolina KillDevil Top500 super cluster.

Integration of authorization for grid

Today there are numerous authorization systems both for protecting and managing individual and distributed computer systems. Systems for authentication and authorization for distributed systems work only on software they were developed for. A complete system of authentication and authorization that works with different grid middleware is not yet developed. What can be done is to either implement a solution which is most adequate for a given distributed system and its purposes or try to make an adequate solution by combining multiple systems. In this paper we present problems and challenges when integrating authorization solutions of different grid systems. We present our experiences in enabling different systems to work together and describe extensions which we made in order to enable multiple authorization systems to work together

CRO-GRID security management system

Computer grids are briefly described as highly distributed systems of computer resources which cooperate together towards creating a reliable, scalable dynamically changeable system for providing resources and services to users. Dynamical nature of grid creates additional security challenges. This paper gives an overview of security demands which a grid security system must overcome. First, an explanation of the grid specific security demands is given. Then, an overview of the available grid security systems, first the concepts they use and later technologies that implement those concepts is given. Furthermore, we use these concepts to define our own architecture for enabling security in a grid environment and enhancements to usual grid security mechanisms. We conclude with a description of future steps and how these steps will improve our system.