Search results for " computing"

Large-scale genome-wide association studies on a GPU cluster using a CUDA-accelerated PGAS programming model

2015

[Abstract] Detecting epistasis, such as 2-SNP interactions, in genome-wide association studies (GWAS) is an important but time consuming operation. Consequently, GPUs have already been used to accelerate these studies, reducing the runtime for moderately-sized datasets to less than 1 hour. However, single-GPU approaches cannot perform large-scale GWAS in reasonable time. In this work we present multiEpistSearch, a tool to detect epistasis that works on GPU clusters. While CUDA is used for parallelization within each GPU, the workload distribution among GPUs is performed with Unified Parallel C++ (UPC++), a novel extension of C++ that follows the Partitioned Global Address Space (PGAS) model…

Large-Scale Clustering of Short Reads for Metagenomics On GPUs

2013

Superparameterised cloud effects in the EMAC general circulation model (v2.50) – influences of model configuration

2020

A new module has been implemented in the fifth generation of the ECMWF/Hamburg (ECHAM5)/Modular Earth Submodel System (MESSy) Atmospheric Chemistry (EMAC) model that simulates cloud-related processes on a much smaller grid. This so-called superparameterisation acts as a replacement for the convection parameterisation and large-scale cloud scheme. The concept of embedding a cloud-resolving model (CRM) inside of each grid box of a general circulation model leads to an explicit representation of cloud dynamics. The new model component is evaluated against observations and the conventional usage of EMAC using a convection parameterisation. In particular, effects of applying different configurat…

Towards a mean body for apparel design

2016

This paper focuses on shape average with applications to the apparel industry. Apparel industry uses a consensus sizing system; its major concern is to fit most of the population into it. Since anthropometric measures do not grow linearly, it is important to find prototypes to accurately represent each size. This is done using random compact mean sets, obtained from a cloud of 3D points given by a scanner and applying to the sample a previous definition of mean set. Additionally, two approaches to define confidence sets are introduced. The methodology is applied to data obtained from a real anthropometric survey. This paper has been partially supported by the following grants: TIN2009-14392…

Evaluating Bluetooth Performance as the Support for Context-Aware Applications

2005

We present an experiment relative to the use of the Bluetooth wireless technology to provide network support for context-aware applications. We describe an approach to provide network interconnection using a combination of wireless and wired network technologies. We also describe the steps taken to create a Bluetooth based context-aware application. We, finally, evaluate, using a small test-bed and simulation the overall performance of this technology when adopted in the area of context-aware and ubiquitous computing.

VarySched: A Framework for Variable Scheduling in Heterogeneous Environments

2016

Despite many efforts to better utilize the potential of GPUs and CPUs, it is far from being fully exploited. Although many tasks can be easily sped up by using accelerators, most of the existing schedulers are not flexible enough to really optimize the resource usage of the complete system. The main reasons are (i) that each processing unit requires a specific program code and that this code is often not provided for every task, and (ii) that schedulers may follow the run-until-completion model and, hence, disallow resource changes during runtime. In this paper, we present VarySched, a configurable task scheduler framework tailored to efficiently utilize all available computing resources in…

Checkpointing Workflows for Fail-Stop Errors

2017

International audience; We consider the problem of orchestrating the exe- cution of workflow applications structured as Directed Acyclic Graphs (DAGs) on parallel computing platforms that are subject to fail-stop failures. The objective is to minimize expected overall execution time, or makespan. A solution to this problem consists of a schedule of the workflow tasks on the available processors and of a decision of which application data to checkpoint to stable storage, so as to mitigate the impact of processor failures. For general DAGs this problem is hopelessly intractable. In fact, given a solution, computing its expected makespan is still a difficult problem. To address this challenge,…

Industrial video sequences for network performance evaluation

2005

The analysis of video traffic characteristics plays a fundamental role in the evaluation of network performance or schedule design. However, the majority of these analyses are based on the study of traffic characteristics generated with different codifiers on typical general purpose films or sequences. These are generally made using asymmetric codifiers which are not usable in live real-time multimedia transmission systems. Moreover, the properties of these sequences are not necessarily equivalent to those obtained in an industrial environment. In this paper industrial monitoring video sequences are presented, describing their various characteristics, from the point of view of the source as…

MAC learning

2016

Cognition as a way to deal with the challenges of future wireless networks has been largely considered by the recent literature, with a main focus on physical layer adaptability and dynamic spectrum access. In this demo, we show how a simple cognition mechanism can be also applied at the MAC layer, by exploiting the emerging paradigm of programmable wireless cards. The idea is using the formal definition of simple MAC protocol components and platform-independent representation of channel events gathered from the wireless node, for emulating the behavior of protocols which are not currently running on the network, learning about their expected performance, and dynamically reconfiguring the w…

Reaction-diffusion on dynamic inhibition areas: A bio-inspired link scheduling algorithm

2014

We present the Dynamic Inhibition Areas Reaction-Diffusion (DIA-RD) algorithm, a distributed medium access control protocol that globally maximizes the spatial reusability (number of simultaneous transmissions per unit area) of wireless sensor networks. This algorithm is able, in consequence, to minimize the number of time slots needed to schedule the set of demanded links, making it very efficient to solve the Shortest Link Schedule problem. DIA-RD combines accurate interference management, provided by the use of dynamic inhibition areas based on the physical interference model; and global intelligent behavior, provided by the bio-inspired technique known as Reaction-Diffusion. This techni…