Explicit Platform Descriptions for Heterogeneous Many-Core Architectures

Sandrieser, Martin; Benkner, Siegfried; Pllana, Sabri

doi:10.1109/ipdps.2011.280

Cited by 9 publications

(9 citation statements)

References 15 publications

(24 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In these cases, an explicitly defined Platform Model (PM) is not used and the transformation becomes restrict to a specific platform [3], [13]. Some initiatives adopt the concept of explicitly defined PM, [2], [4], [25], [26]. Such initiatives aim at the design and use of Platform Models, although they do not provide specific artifacts for modeling RTOS-based embedded systems services.…”

Section: Discussion and Related Workmentioning

confidence: 99%

MT-AMP: A Model Transformation for Embedded Software

Agner¹,

Soares²,

Stadzisz³

et al. 2016

JSW

View full text Add to dashboard Cite

Abstract:MDA is an approach to software development based on the design and transformation of models. In the MDA approach, models are the core artifacts throughout the software development lifecycle, and thus a key role is played by model transformations. Nevertheless, most MDA initiatives are dedicated, i.e., the platform features are implicitly employed in the transformation rules. As the aspects associated with the platform are not separated from the transformation rules, for each selected platform there must be a corresponding model transformation. This makes the model transformation development easier, although the model transformation becomes limited to a specific platform, once the platform features are strongly associated with the transformation rules. An open challenge consists of how to create transformation rules independently of the platform features. This paper presents a model transformation, called MT-AMP, particularly applied to embedded software development based on PI-MT Method. In addition, it allows the use of the model transformation to different RTOS-based platforms by means of Platform Models explicitly defined. As a result, the MT-AMP provides independence between the model transformation rules and the platform features.

show abstract

Section: Discussion and Related Workmentioning

confidence: 99%

MT-AMP: A Model Transformation for Embedded Software

Agner¹,

Soares²,

Stadzisz³

et al. 2016

JSW

View full text Add to dashboard Cite

show abstract

“…Details for each PU (such as, id, number of cores, cache size, core frequency, and global memory) are extracted from the platform description file (.pdl), which needs to be provided as input to our compiler. The PDL is a platform description language for the explicit description of heterogeneous parallel computing systems [19].…”

Section: A Designmentioning

confidence: 99%

HSTREAM: A Directive-Based Language Extension for Heterogeneous Stream Computing

Memeti

Pllana

2018

2018 IEEE International Conference on Computational Science and Engineering (CSE)

Self Cite

View full text Add to dashboard Cite

Big data streaming applications require utilization of heterogeneous parallel computing systems, which may comprise multiple multi-core CPUs and many-core accelerating devices such as NVIDIA GPUs and Intel Xeon Phis. Programming such systems require advanced knowledge of several hardware architectures and device-specific programming models, including OpenMP and CUDA. In this paper, we present HSTREAM, a compiler directive-based language extension to support programming stream computing applications for heterogeneous parallel computing systems. HSTREAM source-to-source compiler aims to increase the programming productivity by enabling programmers to annotate the parallel regions for heterogeneous execution and generate target specific code. The HSTREAM runtime automatically distributes the workload across CPUs and accelerating devices. We demonstrate the usefulness of HSTREAM language extension with various applications from the STREAM benchmark. Experimental evaluation results show that HSTREAM can keep the same programming simplicity as OpenMP, and the generated code can deliver performance beyond what CPUs-only and GPUs-only executions can deliver.

show abstract

“…the programming model/language used for the component implementation and the target architecture. The actual platform properties are defined separately in another XML document [6]. Such platform meta-data can be used at multiple levels of the PEPPHER framework.…”

Section: Components Interfaces Implementationsmentioning

confidence: 99%

The PEPPHER Composition Tool: Performance-Aware Dynamic Composition of Applications for GPU-Based Systems

Dastgeer

Keßler

2012

2012 SC Companion: High Performance Computing, Networking Storage and Analysis

View full text Add to dashboard Cite

Abstract-The PEPPHER component model defines an environment for annotation of native C/C++ based components for homogeneous and heterogeneous multicore and manycore systems, including GPU and multi-GPU based systems. For the same computational functionality, captured as a component, different sequential and explicitly parallel implementation variants using various types of execution units might be provided, together with metadata such as explicitly exposed tunable parameters. The goal is to compose an application from its components and variants such that, depending on the run-time context, the most suitable implementation variant will be chosen automatically for each invocation.We describe and evaluate the PEPPHER composition tool, which explores the application's components and their implementation variants, generates the necessary low-level code that interacts with the runtime system, and coordinates the native compilation and linking of the various code units to compose the overall application code. With several applications, we demonstrate how the composition tool provides a high-level programming front-end while effectively utilizing the task-based PEPPHER runtime system (StarPU) underneath.

show abstract

Explicit Platform Descriptions for Heterogeneous Many-Core Architectures

Cited by 9 publications

References 15 publications

MT-AMP: A Model Transformation for Embedded Software

MT-AMP: A Model Transformation for Embedded Software

HSTREAM: A Directive-Based Language Extension for Heterogeneous Stream Computing

The PEPPHER Composition Tool: Performance-Aware Dynamic Composition of Applications for GPU-Based Systems

Contact Info

Product

Resources

About