Varuna: Enabling Failure-Type Aware RDMA Failover

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Wang, Xiaoyang, Li, Yongkun, Yao, Lulu, Wei, Guoli, Yang, Longcheng, Xu, Yinlong, Kong, Weiqing, Wang, Weiguang, Dong, Peng, Liu, Bingyang
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866911552834109440
author Wang, Xiaoyang
Li, Yongkun
Yao, Lulu
Wei, Guoli
Yang, Longcheng
Xu, Yinlong
Kong, Weiqing
Wang, Weiguang
Dong, Peng
Liu, Bingyang
author_facet Wang, Xiaoyang
Li, Yongkun
Yao, Lulu
Wei, Guoli
Yang, Longcheng
Xu, Yinlong
Kong, Weiqing
Wang, Weiguang
Dong, Peng
Liu, Bingyang
contents RDMA link failures can render connections temporarily unavailable, causing both performance degradation and significant recovery overhead. To tolerate such failures, production datacenters assign each primary link with a standby link and, upon failure, uniformly retransmit all in-flight RDMA request over the backup path. However, we observe that such blanket retransmission is unnecessary. In-flight requests can be split into pre-failure and post-failure categories depending on whether the responder has already executed. Retransmitting post-failure requests is not only redundant (consuming bandwidth), but also incorrect for non-idempotent operations, where duplicate execution can violate application semantics. We present Varuna, a failure-type-aware RDMA recovery mechanism that enables correct retransmission and us-level failover. Varuna piggybacks a lightweight completion log on every RDMA operation; after a link failure, this log deterministically reveals which in-flight requests were executed (post-failure) and which were lost (pre-failure). Varuna then retransmits only the pre-failure subset and fetches/recovers the return values for post-failure requests. Evaluated using synthetic microbenchmarks and end-to-end RDMA TPC-C transactions, Varuna incurs only 0.6-10% steady-state latency overhead in realistic applications, eliminates 65% of recovery retransmission time, preserves transactional consistency, and introduces zero connectivity rebuild overhead and negligible memory overhead during RDMA failover.
format Preprint
id arxiv_https___arxiv_org_abs_2603_28001
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Varuna: Enabling Failure-Type Aware RDMA Failover
Wang, Xiaoyang
Li, Yongkun
Yao, Lulu
Wei, Guoli
Yang, Longcheng
Xu, Yinlong
Kong, Weiqing
Wang, Weiguang
Dong, Peng
Liu, Bingyang
Distributed, Parallel, and Cluster Computing
Networking and Internet Architecture
RDMA link failures can render connections temporarily unavailable, causing both performance degradation and significant recovery overhead. To tolerate such failures, production datacenters assign each primary link with a standby link and, upon failure, uniformly retransmit all in-flight RDMA request over the backup path. However, we observe that such blanket retransmission is unnecessary. In-flight requests can be split into pre-failure and post-failure categories depending on whether the responder has already executed. Retransmitting post-failure requests is not only redundant (consuming bandwidth), but also incorrect for non-idempotent operations, where duplicate execution can violate application semantics. We present Varuna, a failure-type-aware RDMA recovery mechanism that enables correct retransmission and us-level failover. Varuna piggybacks a lightweight completion log on every RDMA operation; after a link failure, this log deterministically reveals which in-flight requests were executed (post-failure) and which were lost (pre-failure). Varuna then retransmits only the pre-failure subset and fetches/recovers the return values for post-failure requests. Evaluated using synthetic microbenchmarks and end-to-end RDMA TPC-C transactions, Varuna incurs only 0.6-10% steady-state latency overhead in realistic applications, eliminates 65% of recovery retransmission time, preserves transactional consistency, and introduces zero connectivity rebuild overhead and negligible memory overhead during RDMA failover.
title Varuna: Enabling Failure-Type Aware RDMA Failover
topic Distributed, Parallel, and Cluster Computing
Networking and Internet Architecture
url https://arxiv.org/abs/2603.28001