User Tools

Site Tools


Sidebar

<menu col=1,align=center>

<item>Documentation||[[lara:documentation]]|{{:lara:img:dictionary.png?25}}</item>
<item>Downloads||[[lara:downloads]]| {{:lara:img:projects.png?25}}</item>
<item>Tutorials||[[lara:tutorial]]|{{:lara:img:books.png?25}}</item>
<item>Other uses of LARA||[[lara:other]]|{{:lara:img:globe.png?25}}</item>
<item>About Us||[[https://sites.google.com/site/specsfeup/]]| {{:lara:img:specslogo.png?25}}</item>
<item>Projects||[[lara:usage]]| {{:lara:img:math.png?25}}</item>

</menu>

/* They are empty */ /* <item>FAQ||faq|</item> */ /* <item>Dev. Team||team| </item> */ /* <item>About LARA||about|</item> */

lara:usage

Projects using LARA

MATISSE

Summary

MATISSE is a compilation framework which generates customized low-level implementations from high-level descriptions. Currently the framework accepts MATLAB as a high-level description language, and generates implementations in C (with an experimental OpenCL backend). It aims to generate several implementations from the same high-level description, with the use of aspects written in LARA, which provide implementation and target-specific information (e.g., variable types, memory allocation policy…).

Publications

João Bispo, Pedro Pinto, Ricardo Nobre, Tiago Carvalho, João M. P. Cardoso, Pedro C. Diniz, “The MATISSE MATLAB Compiler - A MATrix(MATLAB)-aware compiler InfraStructure for embedded computing SystEms,” in IEEE International Conference on Industrial Informatics (INDIN’2013), Bochum, Germany, 29-31 July 2013.

Acknowledgements

MATISSE is a compilation framework completely redesigned in the context of the FP7 EU-funded project REFLECT. We acknowledge the REFLECT members, especially José Gabriel Coutinho, Zlatko Petrov and Pedro Diniz. We also acknowledge the previous contributions done in the context of the AMADEUS project. The AMADEUS project was partially funded by the Portuguese Science Foundation - Fundação para a Ciência e Tecnologia (FCT) under research grants PTDC/EEA-ELC/71556/2006 and PTDC/EIA/70271/2006.

People involved

João Bispo (MATISSE) - http://pt.linkedin.com/in/bispoj/

Luís Reis (OpenCL backend) - http://paginas.fe.up.pt/~ei09030

European Projects:

HARNESS (Hardware- and Network-Enhanced Software Systems for Cloud Computing), FP7 Project

In the FP7 Harness project we are using LARA to derive optimised programs targeting a heterogeneous cloud platform. In particular, cloud programmers can express different versions of the same C function using different semantics, such as based on dataflow computing. LARA aspects, which capture domain-specific knowledge about the cloud infrastructure, can then be used to automatically optimise the code for different types of resources. This way, programmers do not have to know details about the cloud infrastructure while being able leverage their resources.

In our current project, we have identified four types of aspects for dataflow engines: (1) system aspects capture transformations and optimisation strategies that affect the whole application, such as hardware/software partitioning and run-time reconfiguration; (2) architectural aspects focus on low-level design optimisations that can be applied to designs to improve timing, resource usage or exploit specialised architectural features; (3) exploration aspects deal with strategies that generate multiple designs to find an optimal implementation based on user-specified constraints; (4) development aspects relate to concerns that have an impact on the design process, such as debugging, kernel simulation and improving compilation speed.

PhD Thesis and MSc Dissertations:

Paul Grigoras, Aspect Oriented Design for Dataflow Engines, Master of Engineering in Computing of Imperial College London, June 2013. http://www.doc.ic.ac.uk/teaching/distinguished-projects/2013/p.grigoras.pdf


lara/usage.txt · Last modified: 2019/09/22 15:47 (external edit)