EUVIMEDEUVIMED entdecken
Crossref

Fast In-Memory Transaction Processing Using RDMA and HTM

Haibo Chen, Rong Chen, Xingda Wei, Jiaxin Shi, Yanzhe Chen, Zhaoguo Wang, Binyu Zang, Haibing Guan

ACM Transactions on Computer Systems · 28.02.2017 · 35 (1) · 1-37

DOI: 10.1145/3092701 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ACM Transactions on Computer Systems, online erschienen am 13.07.2017. Verfasst von Haibo Chen und 7 weiteren Autor:innen. Die Arbeit wurde bisher 36-mal zitiert (Crossref) und nennt 70 Literaturangaben. Gefördert von: Top-Notch Youth Talents Program of China, Shanghai Science and Technology Development Fund, Zhangjiang Hi-Tech Program, National Key Research 8 Development Program of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

DrTM is a fast in-memory transaction processing system that exploits advanced hardware features such as remote direct memory access (RDMA) and hardware transactional memory (HTM). To achieve high efficiency, it mostly offloads concurrency control such as tracking read/write accesses and conflict detection into HTM in a local machine and leverages the strong consistency between RDMA and HTM to ensure serializability among concurrent transactions across machines. To mitigate the high probability of HTM aborts for large transactions, we design and implement an optimized transaction chopping algorithm to decompose a set of large transactions into smaller pieces such that HTM is only required to protect each piece. We further build an efficient hash table for DrTM by leveraging HTM and RDMA to simplify the design and notably improve the performance. We describe how DrTM supports common database features like read-only transactions and logging for durability. Evaluation using typical OLTP workloads including TPC-C and SmallBank shows that DrTM has better single-node efficiency and scales well on a six-node cluster; it achieves greater than 1.51, 34 and 5.24, 138 million transactions per second for TPC-C and SmallBank on a single node and the cluster, respectively. Such numbers outperform a state-of-the-art single-node system (i.e., Silo) and a distributed transaction system (i.e., Calvin) by at least 1.9X and 29.6X for TPC-C.

Förderung

  • Top-Notch Youth Talents Program of China, Shanghai Science and Technology Development Fund
  • Zhangjiang Hi-Tech Program
  • National Key Research 8 Development Program of China
  • China National Natural Science Foundation

Zitieren

APA

Chen, H., Chen, R., Wei, X., Shi, J., Chen, Y., Wang, Z., Zang, B., & Guan, H. (2017). Fast In-Memory Transaction Processing Using RDMA and HTM. ACM Transactions on Computer Systems, 35(1), 1-37. https://doi.org/10.1145/3092701

Vancouver

Chen H, Chen R, Wei X, Shi J, Chen Y, Wang Z, et al. Fast In-Memory Transaction Processing Using RDMA and HTM. ACM Transactions on Computer Systems. 2017;35(1):1-37. doi:10.1145/3092701

BibTeX
@article{Chen2017,
  author = {Chen, Haibo and Chen, Rong and Wei, Xingda and Shi, Jiaxin and Chen, Yanzhe and Wang, Zhaoguo and Zang, Binyu and Guan, Haibing},
  title = {Fast In-Memory Transaction Processing Using RDMA and HTM},
  journal = {ACM Transactions on Computer Systems},
  year = {2017},
  volume = {35},
  number = {1},
  pages = {1--37},
  doi = {10.1145/3092701}
}
Kostenlos starten →