Search engine for discovering works of Art, research articles, and books related to Art and Culture
Javascript must be enabled to continue!
Formalizing Bottlenecks in Task-Based OpenMP Applications
View through CrossRef
Task support was introduced into OpenMP to address irregular parallelism in shared memory architectures. Creating tasks that are extremely fine granular in applications, however, impedes performance. In this paper, a methodology for analyzing the performance of task-based OpenMP programs and its implementation in Periscope is presented. The paper unveils and concentrates on the newly formulated high-level performance properties that formalize typical performance bottlenecks of task-based programs. In addition, the paper reports on the experimental results which were accomplished for the codes of the Barcelona OpenMP Tasks Suite (BOTS) using Periscope in the SuperMUC supercomputing machine at Garching, Germany.
Title: Formalizing Bottlenecks in Task-Based OpenMP Applications
Description:
Task support was introduced into OpenMP to address irregular parallelism in shared memory architectures.
Creating tasks that are extremely fine granular in applications, however, impedes performance.
In this paper, a methodology for analyzing the performance of task-based OpenMP programs and its implementation in Periscope is presented.
The paper unveils and concentrates on the newly formulated high-level performance properties that formalize typical performance bottlenecks of task-based programs.
In addition, the paper reports on the experimental results which were accomplished for the codes of the Barcelona OpenMP Tasks Suite (BOTS) using Periscope in the SuperMUC supercomputing machine at Garching, Germany.
Related Results
High-level compiler analysis for OpenMP
High-level compiler analysis for OpenMP
Nowadays, applications from dissimilar domains, such as high-performance computing and high-integrity systems, require levels of performance that can only be achieved by means of s...
Towards a safe and efficient OpenMP
Towards a safe and efficient OpenMP
(English) The growing complexity of contemporary multi-core and heterogeneous architectures necessitates parallel programming models capable of efficiently leveraging the available...
Automatic Parallelization for Heterogeneous Embedded Systems
Automatic Parallelization for Heterogeneous Embedded Systems
Parallélisation automatique pour systèmes hétérogènes embarqués
L'utilisation d'architectures hétérogènes, combinant des processeurs multicoeurs avec des accélérate...
Scheduler guided OpenMP execution in cloud VMs
Scheduler guided OpenMP execution in cloud VMs
Exécution OpenMP guidée par le ordonnanceur dans les machines virtuelles cloud
OpenMP est un cadre largement utilisé pour paralléliser les applications, permettant ...
Efficient Parallel Linked List Processing
Efficient Parallel Linked List Processing
OpenMP is a very popular and successful parallel programming API, but efficient parallel traversal of a list (of possibly unknown size) of items linked by pointers is a challenging...
High-level Programming of Vulkan-based GPUs Through OpenMP
High-level Programming of Vulkan-based GPUs Through OpenMP
Abstract
Modern applications often involve complex, structured or data-parallel computations on large datasets. Traditionally, GPUs have served as the primary acc...
Maximize Asset Utilization by Effectively Identifying and Removing Bottlenecks
Maximize Asset Utilization by Effectively Identifying and Removing Bottlenecks
Abstract
ADNOC Gas Processing Ruwais NGL Fractionation plant receives and fractionates the NGL produced in upstream gas processing plants. After operation of newly d...
Task Shifting: Need for a more Cautious and Nuanced Approach
Task Shifting: Need for a more Cautious and Nuanced Approach
Task shifting refers to the strategic redistribution and decentralization of health care tasks from one group to another(including lay workers), the latter usually deemed as having...