@inproceedings{MuellerSchmidtMembarthetal.2022, author = {M{\"u}ller, Andr{\´e} and Schmidt, Bertil and Membarth, Richard and Leißa, Roland and Hack, Sebastian}, title = {AnySeq/GPU: A Novel Approach for Faster Sequence Alignment on GPUs}, pages = {20}, booktitle = {Proceedings of the 36th ACM International Conference on Supercomputing (ICS-2022)}, publisher = {ACM}, address = {New York}, isbn = {978-1-4503-9281-5}, doi = {https://doi.org/10.1145/3524059.3532376}, year = {2022}, language = {en} } @inproceedings{MuellerSchmidtHildebrandtetal.2020, author = {M{\"u}ller, Andr{\´e} and Schmidt, Bertil and Hildebrandt, Andreas and Membarth, Richard and Leißa, Roland and Kruse, Matthis and Hack, Sebastian}, title = {AnySeq: A High Performance Sequence Alignment Library based on Partial Evaluation}, booktitle = {Proceedings: 2020 IEEE 34th International Parallel and Distributed Processing Symposium (IPDPS 2020)}, publisher = {IEEE}, address = {Los Alamitos}, isbn = {978-1-7281-6876-0}, issn = {1530-2075}, doi = {https://doi.org/10.1109/IPDPS47924.2020.00109}, pages = {1030 -- 1040}, year = {2020}, language = {en} } @article{LeissaBoescheHacketal.2018, author = {Leißa, Roland and Boesche, Klaas and Hack, Sebastian and P{\´e}rard-Gayot, Ars{\`e}ne and Membarth, Richard and Slusallek, Philipp and M{\"u}ller, Andr{\´e} and Schmidt, Bertil}, title = {AnyDSL: a partial evaluation framework for programming high-performance libraries}, volume = {2}, pages = {119}, journal = {Proceedings of the ACM on Programming Languages}, number = {OOPSLA}, publisher = {ACM}, address = {New York}, issn = {2475-1421}, doi = {https://doi.org/10.1145/3276489}, year = {2018}, abstract = {This paper advocates programming high-performance code using partial evaluation. We present a clean-slate programming system with a simple, annotation-based, online partial evaluator that operates on a CPS-style intermediate representation. Our system exposes code generation for accelerators (vectorization/parallelization for CPUs and GPUs) via compiler-known higher-order functions that can be subjected to partial evaluation. This way, generic implementations can be instantiated with target-specific code at compile time. In our experimental evaluation we present three extensive case studies from image processing, ray tracing, and genome sequence alignment. We demonstrate that using partial evaluation, we obtain high-performance implementations for CPUs and GPUs from one language and one code base in a generic way. The performance of our codes is mostly within 10\%, often closer to the performance of multi man-year, industry-grade, manually-optimized expert codes that are considered to be among the top contenders in their fields.}, language = {en} }