DeltaPath: dataflow-based high-performance incremental routing

TL;DR

DeltaPath introduces a dataflow-based model for high-performance incremental routing that quickly adapts to network changes, supporting various objectives and policies with superior speed and scalability.

Contribution

It presents a novel execution model for incremental routing based on dataflow and graph processing, enabling rapid reactions and flexible policy support.

Findings

Reacts to link failures within 350ms median time

Handles over two million path requests per second

Achieves latency under 1ms per request

Abstract

Routing controllers must react quickly to failures, reconfigurations and workload or policy changes, to ensure service performance and cost-efficient network operation. We propose a general execution model which views routing as an incremental data-parallel computation on a graph-based network model plus a continuous stream of network changes. Our approach supports different routing objectives with only minor re-configuration of its core algorithm, and easily accomodates dynamic user-defined routing policies. Moreover, our prototype demonstrates excellent performance: on Google Jupiter topology it reacts with a median time of 350ms to link failures and serves more than two million path requests per second each with latency under 1ms. This is three orders-of-magnitude faster than the popular ONOS open-source SDN controller.