Migration to WebDAV in Belle II Experiment Pardi, Silvio; Hara, Takanori; Hernandez Villanueva, Michel ...
EPJ Web of Conferences,
2024, Volume:
295
Journal Article, Conference Proceeding
Peer reviewed
Open access
The usage of WebDAV protocol has become more and more popular within the physics experiments using grid middleware in the last decade, and today it represents a valid alternative to the GridFTP ...currently supported at best-effort level after the retirement of Globus Toolkit. Belle II experiment established the adoption of WebDAV protocol as the main protocol for data access and third-party-copy transfers, without relying on Storage Resource Manager interface (SRM). The migration process, carried on with continuous and gradual steps, has required a large effort to guarantee a smooth transition maintaining the production infrastructure fully operational. In this contribution we show the transition process, the tool of support developed to monitor step by step the status of third-party-copy support with WebDAV protocol by storages of the collaboration tested in both case pull and push, the strategy adopted to configure DIRAC and the solutions put in place for the corner cases. Finally, we will present some statistics of utilization and we will analyse the achieved results.
Full text
Available for:
IZUM, KILJ, NUK, PILJ, PNG, SAZU, UL, UM, UPUK
3.
Preface Sun, Xian-He; Yu, Weikuan
Journal of computer science and technology,
01/2020, Volume:
35, Issue:
1
Journal Article
Se entiende por simulación al proceso por medio del cual se representa, reproduce o imita el comportamiento observable de un proceso o sistema real a lo largo del tiempo y el espacio. La simulación ...distribuida tiene la capacidad de acelerar la ejecución de un único modelo, vincular y reutilizar múltiples modelos para simular modelos más grandes y acelerar la ejecución de etapas de experimentación. En este contexto, la construcción de simulaciones distribuidas ha mejorado en los últimos años gracias al surgimiento de nuevas tecnologías de la información. En este artículo se describen los principios, modos de trabajo y enfoques de administración de tiempo asociados a esta técnica junto con las herramientas de software que, en la actualidad, brindan soporte a su aplicación. Además, se presenta una revisión bibliográfica que evidencia el crecimiento (y la importancia) de su uso como técnica de estudio en diferentes dominios. Palabras-clave: computación distribuida y paralela; gestión del tiempo. Simulation is the process by which the observable behavior of a real process or system is represented, reproduced or imitated in time and space. Distributed simulation can be used for accelerate the execution of models, reuse models in larger models, and accelerate the execution of experiments. Given the emergence of new information technologies, the use of distributed simulation has grown. This paper describes the fundamentals, modes and time management approaches used in distributed simulations along with the software tools that improves its development. Also, a literature review is presented to show how this technique is applied in distinct domains. Keywords: parallel and distribute computing; time management.
The mcplots.cem.ch web site (MCPLOTS) provides a simple online repository of plots made with high-energy-physics event generators, comparing them to a wide variety of experimental data. The ...repository is based on the HEPDATA online database of experimental results and on the RIVET Monte Carlo analysis tool. The repository is continually updated and relies on computing power donated by volunteers, via the LHC@HOME 2.0 platform.
Full text
Available for:
DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, SIK, UILJ, UKNU, UL, UM, UPUK
This article introduces a novel middleware that utilizes cost-effective, low-power computing devices like Raspberry Pi to analyze data from wireless sensor networks (WSNs). It is designed for indoor ...settings like historical buildings and museums, tracking visitors and identifying points of interest. It serves as an evacuation aid by monitoring occupancy and gauging the popularity of specific areas, subjects, or art exhibitions. The middleware employs a basic form of the MapReduce algorithm to gather WSN data and distribute it across available computer nodes. Data collected by RFID sensors on visitor badges is stored on mini-computers placed in exhibition rooms and then transmitted to a remote database after a preset time frame. Utilizing MapReduce for data analysis and a leader election algorithm for fault tolerance, this middleware showcases its viability through metrics, demonstrating applications like swift prototyping and accurate validation of findings. Despite using simpler hardware, its performance matches resource-intensive methods involving audiovisual and AI techniques. This design’s innovation lies in its fault-tolerant, distributed setup using budget-friendly, low-power devices rather than resource-heavy hardware or methods. Successfully tested at a historical building in Greece (M. Hatzidakis’ residence), it is tailored for indoor spaces. This paper compares its algorithmic application layer with other implementations, highlighting its technical strengths and advantages. Particularly relevant in the wake of the COVID-19 pandemic and general monitoring middleware for indoor locations, this middleware holds promise in tracking visitor counts and overall building occupancy.
Full text
Available for:
IZUM, KILJ, NUK, PILJ, PNG, SAZU, UL, UM, UPUK