DUNE: Improving Accuracy for Sketch-INT Network Measurement Systems

TL;DR

DUNE is a novel system that enhances network traffic measurement accuracy by efficiently transmitting and reconstructing sketches using intelligent bucket selection and freshness tracking, demonstrated through theoretical proofs and real-world experiments.

Contribution

DUNE introduces a new scatter sketchlet design and lightweight data structures for smarter bucket selection, improving measurement accuracy in sketch-INT systems.

Findings

Significantly improves measurement accuracy with minimal overhead.

Proven effectiveness through theoretical analysis and real-world testing.

Successfully implemented on commodity programmable switches.

Abstract

In-band Network Telemetry (INT) and sketching algorithms are two promising directions for measuring network traffics in real time. To combine sketch with INT and preserve their advantages, a representative approach is to use INT to send a switch sketch in small pieces (called sketchlets) to end-host for reconstructing an identical sketch. However, in this paper, we reveal that when naively selecting buckets to sketchlets, the end-host reconstructed sketch is inaccurate. To overcome this problem, we present DUNE, an innovative sketch-INT network measurement system. DUNE incorporates two key innovations: First, we design a novel scatter sketchlet that is more efficient in transferring measurement data by allowing a switch to select individual buckets to add to sketchlets; Second, we propose lightweight data structures for tracing "freshness" of the sketch buckets, and present algorithms for smartly selecting buckets that contain valuable measurement data to send to end-host. We theoretically prove the effectiveness of our proposed methods, and implement a prototype on commodity programmable switch. The results of extensive experiments driven by real-world traffics on DUNE suggest that our proposed system can substantially improve the measurement accuracy at a trivial cost.