711 research outputs found
Vectorization with Haswell and CilkPlus
Get PDFProject Specification:
This project concerns the parallel computing and vectorization field for Physics Computing at CERN. The document summarises the results and experience from vectorization activities and an initial evaluation of the CilkPlus technology with two different benchmarks from CERN.
Abstract:
With the release of the Intel Sandy Bridge processor, vectorization ceased to be a “nice to have” feature and became a necessity. This work is focused on optimization, running comparative measurements of available vectorization technologies currently under investigation by the CERN Concurrency Forum. In particular, the project involves an assessment of the limits of autovectorization in two compilers, an evaluation of CilkPlus as implemented in ICC/GCC and an evaluation of AVX/AVX2 benefits with respect to legacy SSE workloads
A Geometrical Modeller for HEP
No full textFor ALICE off-line collaboration Geometrical modelling generally provides the geometrical description of a special structure and a set of services to "navigate" through it. HEP geometrical modellers are designed to handle high complexity detector geometries and they are usually embedded within simulation MC frameworks. The fact that these frameworks greatly depend on their specific geometrical tools makes simulation applications hardly portable to MC’s other than the one they were designed for. The ALICE Off-line Project in collaboration with the ROOT team is proposing a multi-purpose geometrical modeller for HEP that is integrated within a virtual MC infrastructure. This tool has been optimised for performance with the geometry setups of several HEP experiments and provides a single representation for the geometry used by different applications such as simulation, reconstruction or event display. Figure 1: The ROOT geometry modeller is able to represent most HEP experiment
An interface for GEANT4 simulation using ROOT geometry navigation
No full textThe ROOT geometry modeller (TGeo) offers powerful tools for detector geometry description. The package provides several functionalities like: navigation, geometry checking, enhanced visualization, geometry editing GUI and many others, using ROOT I/O. A new interface module g4root was recently developed to take advantage of ROOT geometry navigation optimizations in the context of GEANT4 simulation. The interface can be used either by native GEANT4-based simulation applications or in the more general context of the Virtual Monte Carlo (VMC) framework developed by ALICE offline and ROOT teams. The latter allows running GEANT3, GEANT4 and FLUKA simulations without changing either the geometry description or the user code. The interface was tested and stressed in the context of ALICE simulation framework. A description of the interface, its usage as well as recent results in terms of reliability and performance will be presented. Some benchmarks will be compared for ROOT-TGeo or GEANT4 based navigation
A Geometrical Modeller for HEP
No full textGeometrical modelling generally provides the geometrical description of a special structure and a set of services to "navigate" through its structure. HEP geometrical modellers are designed to handle high complexity detector geometries and they are usually embedded within simulation MC frameworks. The fact that these frameworks greatly depend on their specific geometrical tools makes simulation applications hardly portable to MC's other than the one they were designed for. The ALICE Off-line Project in collaboration with the ROOT team is proposing a multi-purpose geometrical modeller for HEP that is integrated within a virtual MC infrastructure. This tool has been optimised for performance with the geometry setups of several HEP experiments and provides a single representation for the geometry used by different applications such as simulation, reconstruction or event display.Geometrical modelling generally provides the geometrical description of a special structure and a set of services to navigate through its structure. HEP geometrical modellers are designed to handle high complexity detector geometries and they are usually embedded within simulation MC frameworks. The fact that these frameworks greatly depend on their specific geometrical tools makes simulation applications hardly portable to MC's other than the one they were designed for. The ALICE Off-line Project in collaboration with the ROOT team is proposing a multi-purpose geometrical modeller for HEP that is integrated within a virtual MC infrastructure. This tool has been optimised for performance with the geometry setups of several HEP experiments and provides a single representation for the geometry used by different applications such as simulation, reconstruction or event display
Photoresponsive Nanocarriers Based on Lithium Niobate Nanoparticles for Harmonic Imaging and On-Demand Release of Anticancer Chemotherapeutics
No full textNanoparticle-based drug delivery systems have the potential for increasing the efficiency of chemotherapeutics by enhancing the drug accumulation at specific target sites, therebyreducing adverse side effects and mitigating patient acquired resistance. In particular, photo-responsive nanomaterials have attracted much interest due to their ability to release molecularcargos on demand upon light irradiation. In some settings, they can also provide complementary information by optical imaging on the (sub)cellular scale. We herein present a system based on lithium niobate harmonic nanoparticles (LNO HNPs) for the decoupled multi-harmonic cell imaging and near-infrared light-triggered delivery of an erlotinib derivative (ELA) for the treatment of epidermal growth factor receptor (EGFR)-overexpressing carcinomas. The ELA cargo was covalently conjugated to the surface of silica-coated LNO HNPs through a coumarinyl photo-cleavable linker, achieving a surface loading of the active molecule of 27 nmol/mg NPs. The resulting nanoconjugates (LNO-CM-ELA NPs) were successfully imaged upon pulsed laser excitation at 1250 nm in EGFR-overexpressing human prostate cancer cells DU145 by detecting the second harmonic emission at 625 nm, in the tissue transparency window. Tuning the laser at 790 nm resulted in the uncaging of the ELA cargo as a result of the second harmonic emission of the inorganic HNP core at 395 nm. This protocol induced a significant growth inhibition in DU145 cells, which was only observed upon specific irradiation at 790 nm, highlighting the promising capabilities of LNO-CM-ELA NPs for theranostic applications.SCI-SB-S
Geant4 VMC 3.0
No full textVirtual Monte Carlo (VMC) [1] provides an abstract interface into Monte Carlo transport codes. A user VMC based application, independent from the specific Monte Carlo codes, can be then run with any of the supported simulation programs. Developed by the ALICE Offline Project and further included in ROOT [2], the interface and implementations have reached stability during the last decade and have become a foundation for other detector simulation frameworks, the FAIR facility experiments framework being among the first and largest.Geant4 VMC [3], which provides the implementation of the VMC interface for Geant4 [4], is in continuous maintenance and development, driven by the evolution of Geant4 on one side and requirements from users on the other side. Besides the implementation of the VMC interface, Geant4 VMC also provides a set of examples that demonstrate the use of VMC to new users and also serve for testing purposes. Since major release 2.0, it includes the G4Root navigator package, which implements an interface that allows one to run a Geant4 simulation using a ROOT geometry.The release of Geant4 version 10.00 with the integration of multithreading processing has triggered the development of the next major version of Geant4 VMC (version 3.0), which was released in November 2014. A beta version, available for user testing since March, has helped its consolidation and improvement. We will review the new capabilities introduced in this major version, in particular the integration of multithreading into the VMC design, its impact on the Geant4 VMC and G4Root packages, and the introduction of a new package, MTRoot, providing utility functions for ROOT parallel output in independent files with necessary additions for thread-safety. Migration of user applications to multithreading that preserves the ease of use of VMC will be also discussed. We will also report on the introduction of a new CMake [5] based build system, the migration to ROOT major release 6 and the improvement of the testing suites
Performance optimisations for distributed analysis in ALICE
No full textPerformance is a critical issue in a production system accommodating hundreds of analysis users. Compared to a local session, distributed analysis is exposed to services and network latencies, remote data access and heterogeneous computing infrastructure, creating a more complex performance and efficiency optimization matrix. During the last 2 years, ALICE analysis shifted from a fast development phase to the more mature and stable code. At the same time, the framewo rks and tools for deployment, monitoring and management of large productions have evolved considerably too. The ALICE Grid production system is currently used by a fair share of organized and individual user analysis, consuming up to 30% or the available r esources and ranging from fully I/O - bound analysis code to CPU intensive correlations or resonances studies. While the intrinsic analysis performance is unlikely to improve by a large factor during the LHC long shutdown (LS1), the overall efficiency of the system has still to be improved by an important factor to satisfy the analysis needs. We have instrumented all analysis jobs with "sensors" collecting comprehensive monitoring information on the job running conditions and performance in order to identify bottlenecks in the data processing flow. This data are collected by the MonALISa - based ALICE Grid monitoring system and are used to steer and improve the job submission and management policy, to identify operational problems in real time and to perform aut omatic corrective actions. In parallel with an upgrade of our production system we are aiming for low level improvements related to data format, data management and merging of results to allow for a better performing ALICE analysi
Automatic text summarization using a filter-based approach
Get PDF[eng] Recent advances in deep neural networks (DNNs)
have revolutionized the field of natural language processing
(NLP) with promising results in automatic summarization of
short texts. However, automatic text summarization of long texts
remains challenging, especially when multiple sub-topics are
present in the text. In this work, we present QuBART1
: a coupled
DNN architecture which allows automatic summarization of one
or various topics of a text selected by the user. This architecture is
a two stage DNN. The first stage consists in extracting a subset
of the input text based on keywords introduced by the user.
This extraction is based on the distances of each sentence with
respect to the user input in the latent space of a large language
model. The second stage is an abstractive summarization of the
previous extraction obtaining the final output, the summary. This
approach generates state of the art results while giving the user
the ability of controlling the desired output
CERN openlab Open Day
No full textIn a context where LHC plans on blowing up the luminosity limits forcing experiments into stretching their original performance, the HEP software is waking up from its long "sequential" hibernation. Particle transport simulation rises up as the ideal candidate to redesign the code towards a massively parallel approach and to respond to the increasing need for simulated samples.
The R&D challenge is far from being trivial given the expectations: a factor between 3 to 5 gain in throughput with same or better physics performance, by using parallelism on multiple levels and triggering SIMD vectorisation and better cache usage, while keeping the code portable and able to use standard or opportunistic resources such as GPGPU or co-processors.
The presentation will briefly describe the project path to face these challenges, its current status and plans
A vectorization approach for multifaceted solids in VecGeom
Get PDFVecGeom [1] is a multi-purpose geometry library targeting the optimisation of the 3D-solids’ algorithms used extensively in particle transport and tracking applications. The implementations of these algorithms are templated on the input data type and are vectorised based on the VecCore [2] abstraction library in case of multiple inputs in a SIMD vector. This provides additional performance for applications supporting a multi-particle flow, such as the GeantV [3] prototype. VecGeom allows also scalar queries for all the supported solids, an option that started being used in Geant4 [4] since the release 10.2, as optional replacement of the geometry functionality provided by the native Geant4 solids. In single particle mode, VecGeom can still issue SIMD instructions by vectorizing the geometry algorithms featuring loops over internal data structures. This approach has proven to bring very large benefits for the tessellated solids represented in terms of triangular facets. To expose more vectorization in the scalar mode we have extended the approach used for the triangular tessellations to other multifaceted shapes, such as the extruded polygon, the poly-hedra and different trapezoids. We hereby present the strategy used to vectorise the different processing phases for tessellated solids, the performance improvements compared to the previous scalar implementations for other solids using this approach, and how this is reflected in Geant4 simulations using VecGeom as geometry engine
- …
