{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,17]],"date-time":"2026-03-17T08:41:53Z","timestamp":1773736913823,"version":"3.50.1"},"reference-count":31,"publisher":"Springer Science and Business Media LLC","issue":"3-4","license":[{"start":{"date-parts":[[2022,4,22]],"date-time":"2022-04-22T00:00:00Z","timestamp":1650585600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2022,4,22]],"date-time":"2022-04-22T00:00:00Z","timestamp":1650585600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"name":"University of Innsbruck and Medical University of Innsbruck"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Int J Parallel Prog"],"published-print":{"date-parts":[[2022,8]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:p>Providing convenient APIs and notations for data parallelism which remain accessible for programmers while still providing good performance has been a long-term goal of researchers as well as language and library designers. C++20 introduces ranges and views, as well as the composition of operations on them using a concise syntax, but the efficient implementation of these library features is restricted to CPUs. We present the Celerity High-level API, which makes similarly concise mechanisms applicable to GPUs and accelerators, and even distributed memory clusters of GPUs. Crucially, we achieve this very high level of abstraction without a significant negative impact on performance compared to a lower-level implementation, and without introducing any non-standard toolchain components or compilers, by implementing a C++ library infrastructure on top of the Celerity system. This is made possible by two central API design and implementation strategies, which form the core of our contribution. Firstly, gathering as much information as possible at compile-time and using metaprogramming techniques to automatically fuse several distinctly formulated processing steps into a single accelerator kernel invocation. And secondly, leveraging C++20 \u201cConcepts\u201d in order to avoid type erasure, allowing for highly efficient code generation. We have evaluated our approach quantitatively in a comparison to lower-level manual implementations of several benchmarks, demonstrating its low overhead. Additionally, we investigated the individual performance impact of our specific optimizations and design choices, illustrating the advantages afforded by a Concepts-based approach.<\/jats:p>","DOI":"10.1007\/s10766-022-00731-8","type":"journal-article","created":{"date-parts":[[2022,4,22]],"date-time":"2022-04-22T18:02:41Z","timestamp":1650650561000},"page":"341-359","update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":6,"title":["The Celerity High-level API: C++20 for Accelerator Clusters"],"prefix":"10.1007","volume":"50","author":[{"given":"Peter","family":"Thoman","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Florian","family":"Tischler","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Philip","family":"Salzmann","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Thomas","family":"Fahringer","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"297","published-online":{"date-parts":[[2022,4,22]]},"reference":[{"key":"731_CR1","doi-asserted-by":"crossref","unstructured":"Bohr, M.: A 30 year retrospective on Dennard\u2019s mosfet scaling paper. IEEE Solid-State Circuits Soc. Newslett. 12(1), 11\u201313 (2007)","DOI":"10.1109\/N-SSC.2007.4785534"},{"key":"731_CR2","unstructured":"The Top 500 List (2020). http:\/\/www.top500.org"},{"issue":"1","key":"731_CR3","doi-asserted-by":"publisher","first-page":"46","DOI":"10.1109\/99.660313","volume":"5","author":"L Dagum","year":"1998","unstructured":"Dagum, L., Menon, R.: Openmp: an industry standard API for shared-memory programming. Comput. Sci. Eng., IEEE 5(1), 46\u201355 (1998)","journal-title":"Comput. Sci. Eng., IEEE"},{"key":"731_CR4","unstructured":"Group, K.O.W.: The opencl specification, version 2.0. Technical report (2014)"},{"key":"731_CR5","unstructured":"Forum, M.P.I.: MPI: a message-passing interface standard. Technical Report, Knoxville, TN, USA (1994)"},{"key":"731_CR6","unstructured":"Carlson, W.W., Draper, J.M., Culler, D.E.: Introduction to UPC and language specification. Technical Report (1999). https:\/\/www.researchgate.net\/publication\/228734882_Introduction_to_UPC_and_language_specification"},{"key":"731_CR7","doi-asserted-by":"crossref","unstructured":"Blumofe, R.D., Joerg, C.F.: Cilk: an efficient multithreaded runtime system. In: Proc. of the ACM SIGPLAN Symp. on Principles and Practice of Parallel Programming (PPoPP), pp. 207\u2013216 (1995)","DOI":"10.1145\/209937.209958"},{"issue":"3","key":"731_CR8","doi-asserted-by":"publisher","first-page":"291","DOI":"10.1177\/1094342007078442","volume":"21","author":"BL Chamberlain","year":"2007","unstructured":"Chamberlain, B.L., Callahan, D., Zima, H.P.: Parallel programmability and the chapel language. Int. J. High Perform. Comput. Appl. 21(3), 291\u2013312 (2007)","journal-title":"Int. J. High Perform. Comput. Appl."},{"key":"731_CR9","doi-asserted-by":"crossref","unstructured":"Augonnet, C., Thibault, S., Namyst, R., Wacrenier, P.-A.: Starpu: a unified platform for task scheduling on heterogeneous multicore architectures. In: Euro-Par Parallel Processing (2009)","DOI":"10.1007\/978-3-642-03869-3_80"},{"issue":"2","key":"731_CR10","doi-asserted-by":"publisher","first-page":"173","DOI":"10.1142\/S0129626411000151","volume":"21","author":"A Duran","year":"2011","unstructured":"Duran, A., Ayguad\u00e9, E., Badia, R.M., Labarta, J., Martinell, L., Martorell, X., Planas, J.: Ompss: a proposal for programming heterogeneous multi-core architectures. Parallel Process. Lett. 21(2), 173\u2013193 (2011)","journal-title":"Parallel Process. Lett."},{"key":"731_CR11","doi-asserted-by":"crossref","unstructured":"Edwards, H.C., Trott, C.R.: Kokkos: enabling performance portability across manycore architectures. In: Extreme Scaling Workshop (XSW), pp. 18\u201324. IEEE (2013)","DOI":"10.1109\/XSW.2013.7"},{"key":"731_CR12","doi-asserted-by":"publisher","DOI":"10.2172\/1169830","volume-title":"The Raja Portability Layer: Overview and Status","author":"R Hornung","year":"2014","unstructured":"Hornung, R., Keasler, J., et al.: The Raja Portability Layer: Overview and Status. Lawrence Livermore National Laboratory, Livermore (2014)"},{"key":"731_CR13","unstructured":"Medina, D.S., St-Cyr, A., Warburton, T.: Occa: a unified approach to multi-threading languages. arXiv preprint arXiv:1403.0968 (2014)"},{"key":"731_CR14","unstructured":"Group, T.K.: SYCL Specification, Version 1.2.1 Revision 3. https:\/\/www.khronos.org\/registry\/SYCL\/specs\/sycl-1.2.1.pdf. Accessed 05 Feb 2019"},{"key":"731_CR15","doi-asserted-by":"crossref","unstructured":"Thoman, P., Salzmann, P., Cosenza, B., Fahringer, T.: Celerity: high-level c++ for accelerator clusters. In: European Conference on Parallel Processing, pp. 291\u2013303. Springer (2019)","DOI":"10.1007\/978-3-030-29400-7_21"},{"key":"731_CR16","first-page":"17","volume":"3","author":"J Kessenich","year":"2018","unstructured":"Kessenich, J., Ouriel, B., Krisch, R.: Spir-v specification. Khronos Group 3, 17 (2018)","journal-title":"Khronos Group"},{"key":"731_CR17","volume-title":"LINQ in Action","author":"F Marguerie","year":"2008","unstructured":"Marguerie, F., Eicherte, S., Wooley, J.: LINQ in Action. Manning Publications Co, New York City (2008)"},{"key":"731_CR18","doi-asserted-by":"crossref","unstructured":"Reyes, R., Brown, G., Burns, R., Wong, M.: Sycl 2020: More than meets the eye. In: Proceedings of the International Workshop on OpenCL, p. 1. Association for Computing Machinery, New York (2020)","DOI":"10.1145\/3388333.3388649"},{"key":"731_CR19","volume-title":"C++ Template Metaprogramming: Concepts, Tools, and Techniques from Boost and Beyond","author":"D Abrahams","year":"2004","unstructured":"Abrahams, D., Gurtovoy, A.: C++ Template Metaprogramming: Concepts, Tools, and Techniques from Boost and Beyond. Pearson Education, London (2004)"},{"issue":"6","key":"731_CR20","doi-asserted-by":"publisher","first-page":"552","DOI":"10.1063\/1.4822499","volume":"10","author":"SW Haney","year":"1996","unstructured":"Haney, S.W., Dubois, P.F.: Beating the abstraction penalty in C++ using expression templates. Comput. Phys. 10(6), 552\u2013557 (1996)","journal-title":"Comput. Phys."},{"issue":"2","key":"731_CR21","doi-asserted-by":"publisher","first-page":"42","DOI":"10.1137\/110830125","volume":"34","author":"K Iglberger","year":"2012","unstructured":"Iglberger, K., Hager, G., Treibig, J., R\u00fcde, U.: Expression templates revisited: a performance analysis of current methodologies. SIAM J. Sci. Comput. 34(2), 42\u201369 (2012)","journal-title":"SIAM J. Sci. Comput."},{"key":"731_CR22","doi-asserted-by":"crossref","unstructured":"Chen, J., Joo, B.: Automatic offloading C++ expression templates to CUDA enabled GPUS. In: IEEE 26th International Parallel and Distributed Processing Symposium Workshops, pp. 2359\u20132368. IEEE (2012)","DOI":"10.1109\/IPDPSW.2012.293"},{"issue":"12","key":"731_CR23","doi-asserted-by":"publisher","first-page":"3240","DOI":"10.1016\/j.jpdc.2014.07.002","volume":"74","author":"P Esterie","year":"2014","unstructured":"Esterie, P., Falcou, J., Gaunard, M., Laprest\u00e9, J.-T., Lacassagne, L.: The numerical template toolbox: a modern C++ design for scientific computing. J. Parallel Distrib. Comput. 74(12), 3240\u20133253 (2014)","journal-title":"J. Parallel Distrib. Comput."},{"key":"731_CR24","doi-asserted-by":"crossref","unstructured":"Falcou, J., S\u00e9rot, J., Chateau, T., Laprest\u00e9, J.T.: Quaff: efficient C++ design for parallel skeletons. Parallel Comput. 32(7), 604\u2013615 (2006). Algorithmic Skeletons","DOI":"10.1016\/j.parco.2006.06.001"},{"key":"731_CR25","doi-asserted-by":"crossref","unstructured":"Javed, N., Loulergue, F.: Parallel programming and performance predictability with Orl\u00e9ans skeleton library. In: 2011 International Conference on High Performance Computing and Simulation, pp. 257\u2013263 (2011)","DOI":"10.1109\/HPCSim.2011.5999832"},{"key":"731_CR26","doi-asserted-by":"publisher","first-page":"72","DOI":"10.1007\/978-3-642-16478-1_5","volume-title":"Implementation and Application of Functional Languages","author":"K Matsuzaki","year":"2010","unstructured":"Matsuzaki, K., Emoto, K.: Implementing fusion-equipped parallel skeletons by expression templates. In: Moraz\u00e1n, M.T., Scholz, S.-B. (eds.) Implementation and Application of Functional Languages, pp. 72\u201389. Springer, Berlin, Heidelberg (2010)"},{"key":"731_CR27","doi-asserted-by":"publisher","first-page":"79","DOI":"10.1007\/978-3-642-10672-9_8","volume-title":"Programming Languages and Systems","author":"S Sato","year":"2009","unstructured":"Sato, S., Iwasaki, H.: A skeletal parallel framework with fusion optimizer for GPGPU programming. In: Hu, Z. (ed.) Programming Languages and Systems, pp. 79\u201394. Springer, Berlin, Heidelberg (2009)"},{"issue":"5","key":"731_CR28","doi-asserted-by":"publisher","first-page":"5003","DOI":"10.1002\/cpe.5003","volume":"31","author":"A Ernstsson","year":"2019","unstructured":"Ernstsson, A., Kessler, C.: Extending smart containers for data locality-aware skeleton programming. Concurr. Comput.: Pract. Exp. 31(5), 5003 (2019)","journal-title":"Concurr. Comput.: Pract. Exp."},{"issue":"8","key":"731_CR29","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/2370036.2145818","volume":"47","author":"HP Huynh","year":"2012","unstructured":"Huynh, H.P., Hagiescu, A., Wong, W.-F., Goh, R.S.M.: Scalable framework for mapping streaming applications onto multi-GPU systems. SIGPLAN Not. 47(8), 1\u201310 (2012)","journal-title":"SIGPLAN Not."},{"issue":"1","key":"731_CR30","doi-asserted-by":"publisher","first-page":"174","DOI":"10.1109\/TPDS.2018.2855182","volume":"30","author":"B Peccerillo","year":"2018","unstructured":"Peccerillo, B., Bartolini, S.: Phast-a portable high-level modern C++ programming library for GPUS and multi-cores. IEEE Trans. Parallel Distrib. Syst. 30(1), 174\u2013189 (2018)","journal-title":"IEEE Trans. Parallel Distrib. Syst."},{"key":"731_CR31","doi-asserted-by":"crossref","unstructured":"Copik, M., Kaiser, H.: Using sycl as an implementation framework for hpx compute. In: Proceedings of the 5th International Workshop on OpenCL (2017)","DOI":"10.1145\/3078155.3078187"}],"container-title":["International Journal of Parallel Programming"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s10766-022-00731-8.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s10766-022-00731-8\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s10766-022-00731-8.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,7,7]],"date-time":"2022-07-07T22:09:44Z","timestamp":1657231784000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s10766-022-00731-8"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,4,22]]},"references-count":31,"journal-issue":{"issue":"3-4","published-print":{"date-parts":[[2022,8]]}},"alternative-id":["731"],"URL":"https:\/\/doi.org\/10.1007\/s10766-022-00731-8","relation":{},"ISSN":["0885-7458","1573-7640"],"issn-type":[{"value":"0885-7458","type":"print"},{"value":"1573-7640","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,4,22]]},"assertion":[{"value":"5 October 2021","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"16 March 2022","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"22 April 2022","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}}]}}