DaMoN '21
Playing Fetch with CAT: Composing Cache Partitioning and Prefetching for Task-Based Query Processing
Proceedings of the 17th International Workshop on Data Management on New Hardware, June 2021.
Abstract
Software prefetching and hardware-based cache allocation techniques (CAT) have been successfully applied in main-memory database engines to fetch data into cache before it is needed and to partition a shared last-level cache (LLC) to prevent concurrent tasks from evicting each others' data. We investigate the interaction of these techniques and demonstrate that while a single prefetching strategy is sufficient, the combination of both techniques is only effective if the cache partitioning strategy adapts the partitioning based on the types of tasks currently sharing an LLC. We present a simple, yet effective, scheme that uses prefetching and adapts cache partition allocations dynamically.
BibTeX
@inproceedings{ZENG:2021:CAT,
author = {Qitian Zeng and Kyle C. Hale and Boris Glavic},
title = {Playing Fetch with {CAT}: Composing Cache Partitioning and Prefetching for Task-Based Query Processing},
booktitle = {Proceedings of the 17th International Workshop on Data Management on New Hardware},
series = {DaMoN'21},
year = {2021},
month = {June},
articleno = {15},
numpages = {5},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
location = {Virtual Event, China},
isbn = {9781450385565},
doi = {10.1145/3465998.3466016},
url = {https://doi.org/10.1145/3465998.3466016}
}