A user-space RDMA control plane, using cached libibverbs calls and fork-shared queue pairs, matches a kernel-based design in setup time while improving data plane throughput by 30-46%.
The new linux’perf’tools
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.NI 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Swift: Rethinking RDMA Control Plane for Elastic Computing
A user-space RDMA control plane, using cached libibverbs calls and fork-shared queue pairs, matches a kernel-based design in setup time while improving data plane throughput by 30-46%.