Abstract

Abstract
The ANTAREX project developed an approach to the performance tuning of High Performance applications based on an Aspect-oriented Domain Specific Language (DSL), with the goal to simplify the enforcement of extra-functional properties in large scale applications. The project aims at demonstrating its tools and techniques on two relevant use cases, one in the domain of computational drug discovery, the other in the domain of online vehicle navigation. In this paper, we present an overview of the project and of its main achievements, as well as of the large scale experiments that have been planned to validate the approach.

Abstract
High-level synthesis (HLS) is of paramount importance to enable software developers to map critical computations to FPGA-based hardware accelerators. However, in order to generate efficient hardware accelerators one needs to apply significant code transformations and adequately use the directive-driven approach, part of most HLS tools. The code restructuring and directives needed are dependent not only of the characteristics of the input code but also of the HLS tools and target FPGAs. These aspects require a deep knowledge about the subjects involved and tend to exclude software developers. This paper presents our recent approach for automatic code restructuring targeting HLS tools. Our approach uses an unfolded graph representation, which can be generated from program execution traces, and graph-based optimizations, such as folding, to generate suitable HLS C code. In this paper, we describe the approach and the new optimizations proposed. We evaluate the approach with a number of representative kernels and the results show its capability to generating efficient hardware implementations only achievable using manual restructuring of the input software code and manual insertion of adequate HLS directives. © 2019, Springer Nature Switzerland AG.

Abstract
Usually, Aspect-Oriented Programming (AOP) languages are an extension of a specific target programming language (e.g., AspectJ for JAVA and AspectC++ for C++). Although providing AOP support with target language extensions may ease the adoption of an approach, it may impose constraints related with constructs and semantics. Furthermore, by tightly coupling the AOP language to the target language the reuse potential of many aspects, especially the ones regarding non-functional requirements, is lost. LARA is a domain-specific language inspired by AOP concepts, having the specification of source-to-source transformations as one of its main goals. LARA has been designed to be, as much as possible, independent of the target language and to provide constructs and semantics that ease the definition of concerns, especially related to non-functional requirements. In this paper, we propose techniques to overcome some of the challenges presented by a multilanguage approach to AOP of cross-cutting concerns focused on non-functional requirements and applied through the use of a weaving process. The techniques mainly focus on providing well-defined library interfaces that can have concrete implementations for each supported target language. The developer uses an agnostic interface and the weaver provides a specific implementation for the target language. We evaluate our approach using 8 concerns with varying levels of language agnosticism that support 4 target languages (C, C++, JAVA and MATLAB) and show that the proposed techniques contribute to more concise LARA aspects, high reuse of aspects, and to significant effort reductions when developing weavers for new imperative, object-oriented programming languages. © 2018 Elsevier Ltd

Abstract
Writing mixed-precision kernels allows to achieve higher throughput together with outputs whose precision remain within given limits. The recent introduction of native half-precision arithmetic capabilities in several GPUs, such as NVIDIA P100 and AMD Vega 10, contributes to make precision-tuning even more relevant as of late. However, it is not trivial to manually find which variables are to be represented as half-precision instead of single- or double-precision. Although the use of half-precision arithmetic can speed up kernel execution considerably, it can also result in providing non-usable kernel outputs, whenever the wrong variables are declared using the half-precision data-type. In this paper we present an automatic approach for precision tuning. Given an OpenCL kernel with a set of inputs declared by a user (i.e., the person responsible for programming and/or tuning the kernel), our approach is capable of deriving the mixed-precision versions of the kernel that are better improve upon the original with respect to a given metric (e.g., time-to-solution, energy-to-solution). We allow the user to declare and/or select a metric to measure and to filter solutions based on the quality of the output. We implement a proof-of-concept of our approach using an aspect-oriented programming language called LARA. It is capable of generating mixed-precision kernels that result in considerably higher performance when compared with the original single-precision floating-point versions, while generating outputs that can be acceptable in some scenarios. © 2018 Copyright held by the owner/author(s).