Abstract
Domain-specific frameworks – including embedded domain-specific languages and libraries – increase programmer productivity by encapsulating proven manual optimization strategies into software modules or (semi-)automated tools. In such frameworks, optimizations and optimization strategies capitalize on knowledge of the requirements of a particular application domain to achieve high performance and architecture portability. While many strategies have been used to develop domain-specific frameworks, this position paper argues the importance of polyhedral compiler technology and autotuning for important classes of high-performance computing domains. Such an approach has the following advantages over other strategies: (1) composability; (2) software reuse; and, (3) facilitates performance portability.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
Acknowledgments
This work has been supported in part by DOE award DE-SC0008682 and NSF award CCF-1564074.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2017 Springer International Publishing AG
About this paper
Cite this paper
Hall, M., Basu, P. (2017). Polyhedral Compiler Technology in Collaboration with Autotuning Important to Domain-Specific Frameworks for HPC. In: Ding, C., Criswell, J., Wu, P. (eds) Languages and Compilers for Parallel Computing. LCPC 2016. Lecture Notes in Computer Science(), vol 10136. Springer, Cham. https://doi.org/10.1007/978-3-319-52709-3_9
Download citation
DOI: https://doi.org/10.1007/978-3-319-52709-3_9
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-52708-6
Online ISBN: 978-3-319-52709-3
eBook Packages: Computer ScienceComputer Science (R0)