Report of Contributions
Total Page:16
File Type:pdf, Size:1020Kb
5th EGEE User Forum Report of Contributions https://indico.cern.ch/e/69338 5th EGEE User F … / Report of Contributions Distributed parametric optimizatio … Contribution ID: 1 Type: Oral Distributed parametric optimization using the Geneva library Wednesday, 14 April 2010 14:00 (20 minutes) The Geneva library implements parallel and distributed parametric optimization algorithms, capa- ble of running on devices ranging from multi-core systems over clusters all the way to Grids and Clouds. The generic design makes Grid-resources available to user groups that have formerly not been exposed to such environments. Detailed analysis Geneva (’Grid-enabled evolutionary algorithms’) is a software library which enables users to solve large-scale parametric optimization problems on devices ranging from multi-core systems over clusters to Grids and Clouds. The generic design makes Geneva applicable to problem domains from a wide range of industrial and scientific application scenarios. From a user’s perspective, parallel and multi-threaded execution can be achieved just as easily as serial execution on a single CPU-core, and does not require the user’s evaluation functions to be aware of the parallel environ- ment. Performance and extensibility are at the core of the C++-based, object-oriented design. The software has been shown to run in parallel with 1000 clients on a Linux cluster, each contributing a fraction of the overall solution. Given suitably complex optimization problems, scalability is al- most linear. The code is available as Open Source, allowing customization under the terms ofthe Affero GPL v3. Conclusions and Future Work The optimization environment will be further expanded to become a clearing house ofdifferent optimization algorithms, all based on the same data structures and using the existing framework for Grid-aware parallelization. Impact The generic design makes Grid technologies accessible to users ranging from scientific toindustrial application domains. It thus has a strong potential to bring new user groups to the Grid. Keywords Optimization Grid Cloud URL for further information http://www.gemfony.com Justification for delivering demo and/or technical requirements (for demos) 1 5th EGEE User F … / Report of Contributions Distributed parametric optimizatio … The topic is sufficiently complex that it cannot be presented as a poster. Given sufficient network connectivity, a number of exa-catching demos can be presented. Primary author: Dr BERLICH, Ruediger (Steinbuch Centre for Computing, Karlsruhe Institute of Technology) Presenter: Dr BERLICH, Ruediger (Steinbuch Centre for Computing, Karlsruhe Institute of Technol- ogy) Session Classification: Computer Science Track Classification: Experiences from application porting and deployment September 27, 2021 Page 2 5th EGEE User F … / Report of Contributions Eliminating and preventing Grid S … Contribution ID: 2 Type: Poster Eliminating and preventing Grid Security Vulnerabilities to reduce security risk Monday, 12 April 2010 17:00 (3 minutes) The EGEE Grid Security Vulnerability Group was formed “to incrementally make the Gridmore secure and thus provide better availability and sustainability of the deployed infrastructure”. The aim is to eliminate vulnerabilities from the Grid and prevent new ones from being introduced, thus reducing the risk of security incidents. This poster alerts users and developers to both the activities of the this group and problems that may be caused by vulnerabilities. It is also intended to inform what they should do to avoid introducing vulnerabilities and report any their find. Impact At the time of writing, over 70 Grid Security Vulnerability bugs have been fixed since the activity started, and several pieces of less secure middleware have been taken out of use as more secure software comes into use. As far as we are aware, no major Grid Security incident within EGEE or collaborating projects has occured due to a vulnerability in Grid Middleware. This could be due to the fact that wehave successfully eliminated many vulnerabilities from the middleware, and successful in preventing new ones. It could also be due to hackers being less alert to Grids that other systems, or that the user community is particularly honest. Keywords Grid Security Vulnerability Risk URL for further information http://www.gridpp.ac.uk/gsvg/ Detailed analysis A system for handling Grid Security vulnerabilities was setup in 2006 at the beginning of EGEE- II. A Risk Assessment Team (RAT) was formed which currently has13 members from 10 different institutes to carry out this process, which is: - anyone may report an issue - the RAT investigates the issue - if valid the RAT carries out a Risk assessment putting each issue into 1 of 4 risk categories - Extremely Critical, High, Moderate, or Low. - a target date for resolution is set according to the risk - an advisory is issued when the issue is fixed, or on the target date Certain types of problem occur quite frequently - such as vulnerabilities resulting from incorrect file permissions, or failure to sanitize user input,. Suggestions for how to prevent new vulnerabil- ities entering the infrastructure will be made. September 27, 2021 Page 3 5th EGEE User F … / Report of Contributions Eliminating and preventing Grid S … Some members of the group are currently participating in an ‘Overall Security Risk Assessment’ which looks at high level risks to the infrastructure, data, and various parties. A summary the reasons for doing this, the strategy, and the outcome (assuming it is available in time) will be included on the poster. Conclusions and Future Work A good enthusiastic team has been handling Grid Security Vulnerabilities over the last 4 years, and many vulnerabilities have been eliminated from the deployed infrastructure. New types of vulnerabilities continue to be found and hackers get ever more ingenious in their quest to gain access to sites. This work therefore needs to continue into the future as grid technology increases it’s profile and hackers become more alert to this type of system and find new ways ofexploiting systems. Primary author: Dr CORNWALL, Linda Ann (Particle Physics-Rutherford Appleton Laborato- ry-STFC - Science &) Presenter: Dr CORNWALL, Linda Ann (Particle Physics-Rutherford Appleton Laboratory-STFC - Science &) Session Classification: Poster session Track Classification: National and international activities and collaborations September 27, 2021 Page 4 5th EGEE User F … / Report of Contributions ATLAS Distributed Computing in … Contribution ID: 4 Type: Oral ATLAS Distributed Computing in Spain and Portugal: from data challenges to real data Wednesday, 14 April 2010 12:00 (20 minutes) The ATLAS PIC cloud is composed of the Iberian sites: PIC (Tier-1), IFAE, IFIC, UAM, LIP-Lisbon, LIP-Coimbra and NCG-INGRID-PT (Tier-2s) and is finalising preparations for the LHC data taking. To achieve readiness for data taking, all sites has been involved in the ATLAS Distributed Comput- ing activities since early 2006: simulated event production, reprocessing, data and user analysis challenges and cosmic data taking. The evolution of the computing and operations activities from data challenges to the real data is described in this talk giving experiment and site experiences. Detailed analysis After the deployment of the required services for the experiment, the Iberian sites nowfacethe challenges associated with real data taking, and that is to provide a reliable service for data analysis for the world-wide scientific ATLAS community and particularity to the Iberian ATLAS scientists. The WLCG grid has been used so far by WLCG institute specialists and few scientists, butnowit must be a daily common “tool” for every single LHC scientist with the first real data of the LHC appearing at the sites. The ATLAS institutes committed to ATLAS in the Iberian cloud have been preparing sites for the data taking since 2006, and achieving good results in all the wide activities that WLCG and ATLAS have been conducting. In this talk we will an overview of the ATLAS ac- tivities in Spain and Portugal since 2006, following the evolution of the different tools/frameworks that have been adopted by the experiment and the sites while awaiting the LHC start-up. Conclusions and Future Work It has been a long road to have the first LHC data being analyzed at the sites. By the timeof the 2010 Users Forum, sites should be flooded with new users willing to run their jobs. Thereal challenge is coming soon, when all of this people will be really grid-aware. The work until now has been exhaustive with the Iberian sites participated in all distributed computing activities, and we are now ready to face the data processing challenges associated with real LHC data taking. Impact ATLAS distributed computing activities are reliant on the EGEE structure, the critical services abeing the catalogues (LFC), the transfers engine (FTS) and the Storage and Computing Elements that are crucial for the scientific community. Also ATLAS has important services running atevery ATLAS cloud, such as the pilot factories and Frontier/squid servers which were recently deployed. All of these mechanisms have to be in place and in high availability to ensure the correct flow of data, and the execution of MonteCarlo/data processing jobs and user analysis jobs. In this talk we present the status and operations model of the ATLAS PIC cloud, following the evolution of ATLAS distributed computing and the improvement of the quality of service, which resulted