Reduction is an operation performed on the values of two or more key-value pairs that share the same key. Reduction of sparse data streams finds application in a wide variety of domains such as data and graph analytics, cybersecurity, machine learning, and HPC applications. However, these applications exhibit low locality of reference, rendering traditional architectures and data representations inefficient. This article presents MetaStrider, a significant algorithmic and architectural enhancement to the state-of-the-art, SuperStrider. Furthermore, these enhancements enable a variety of parallel, memory-centric architectures that we propose, resulting in demonstrated performance that scales near-linearly with available memory-level parallelism.
Srikanth, Sriseshan, et al. "MetaStrider: Architectures for Scalable Memory-centric Reduction of Sparse Data Streams." ACM Transactions on Architecture and Code Optimization, vol. 16, no. 4, Oct. 2019. https://doi.org/10.1145/3355396
Srikanth, Sriseshan, Jain, Anirudh, Lennon, Joseph M., Conte, Thomas M., Debenedictis, Erik, & Cook, Jeanine (2019). MetaStrider: Architectures for Scalable Memory-centric Reduction of Sparse Data Streams. ACM Transactions on Architecture and Code Optimization, 16(4). https://doi.org/10.1145/3355396
Srikanth, Sriseshan, Jain, Anirudh, Lennon, Joseph M., et al., "MetaStrider: Architectures for Scalable Memory-centric Reduction of Sparse Data Streams," ACM Transactions on Architecture and Code Optimization 16, no. 4 (2019), https://doi.org/10.1145/3355396
@article{osti_1697985,
author = {Srikanth, Sriseshan and Jain, Anirudh and Lennon, Joseph M. and Conte, Thomas M. and Debenedictis, Erik and Cook, Jeanine},
title = {MetaStrider: Architectures for Scalable Memory-centric Reduction of Sparse Data Streams},
annote = {Reduction is an operation performed on the values of two or more key-value pairs that share the same key. Reduction of sparse data streams finds application in a wide variety of domains such as data and graph analytics, cybersecurity, machine learning, and HPC applications. However, these applications exhibit low locality of reference, rendering traditional architectures and data representations inefficient. This article presents MetaStrider, a significant algorithmic and architectural enhancement to the state-of-the-art, SuperStrider. Furthermore, these enhancements enable a variety of parallel, memory-centric architectures that we propose, resulting in demonstrated performance that scales near-linearly with available memory-level parallelism.},
doi = {10.1145/3355396},
url = {https://www.osti.gov/biblio/1697985},
journal = {ACM Transactions on Architecture and Code Optimization},
issn = {ISSN 1544-3566},
number = {4},
volume = {16},
place = {United States},
publisher = {Association for Computing Machinery},
year = {2019},
month = {10}}
2017 IEEE International Parallel and Distributed Processing Symposium: Workshops (IPDPSW), 2017 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW)https://doi.org/10.1109/IPDPSW.2017.8
MEMSYS '18: The International Symposium on Memory Systems, Proceedings of the International Symposium on Memory Systemshttps://doi.org/10.1145/3240302.3240314