RDMA-Based Apache Storm for High-Performance Stream Data Processing - Network and Parallel Computing Access content directly
Conference Papers Year : 2021

RDMA-Based Apache Storm for High-Performance Stream Data Processing

Abstract

Apache Storm is a scalable fault-tolerant distributed real-time stream-processing framework widely used in big data applications. For distributed data-sensitive applications, low-latency, high-throughput communication modules have a critical impact on overall system performance. Apache Storm currently uses Netty as its communication component, an asynchronous server/client framework based on TCP/IP protocol stack. The TCP/IP protocol stack has inherent performance flaws due to frequent memory copying and context switching. The Netty component not only limits the performance of the Storm but also increases the CPU load in the IPoIB (IP over InfiniBand) communication mode. In this paper, we introduce two new implementations for Apache Storm communication components with the help of RDMA technology. The performance evaluation on Mellanox QDR Cards (40 Gbps) shows that our implementations can achieve speedup up to 5$$\times $$× compared with IPoIB and 10$$\times $$× with 1 Gigabit Ethernet. Our implementations also significantly reduce the CPU load and increase the throughput of the system.
Fichier principal
Vignette du fichier
511910_1_En_24_Chapter.pdf (5.84 Mo) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-03768735 , version 1 (04-09-2022)

Licence

Attribution

Identifiers

Cite

Ziyu Zhang, Zitan Liu, Qingcai Jiang, Zheng Wu, Junshi Chen, et al.. RDMA-Based Apache Storm for High-Performance Stream Data Processing. 17th IFIP International Conference on Network and Parallel Computing (NPC), Sep 2020, Zhengzhou, China. pp.276-287, ⟨10.1007/978-3-030-79478-1_24⟩. ⟨hal-03768735⟩
34 View
6 Download

Altmetric

Share

Gmail Facebook X LinkedIn More