LACeS: An Open, Fast, Responsible, and Efficient Longitudinal Anycast Census System

TL;DR

LACeS addresses the need for an accurate, longitudinal, and open census of IP anycast deployments by unifying latency-based GCD analysis with synchronized, multi-protocol probing. By redesigning MAnycast2's pipeline and incorporating iGreedy's latency approach, LACeS achieves high-precision daily censuses for both IPv4 and IPv6, with componentized deployment (Orchestrator, CLI, Workers) and a public data release. The system validates results against operator ground truth, production deployments, and RIPE Atlas measurements, and demonstrates robust longitudinal stability over 17+ months. Its open-source toolchain and daily censuses provide a practical asset for researchers and operators to study internet resilience, routing dynamics, and the geographical distribution of anycast services.

Abstract

IP anycast replicates an address at multiple locations to reduce latency and enhance resilience. Due to anycast's crucial role in the modern Internet, earlier research introduced tools to perform anycast censuses. The first, iGreedy, uses latency measurements from geographically dispersed locations to map anycast deployments. The second, MAnycast2, uses anycast to perform a census of other anycast networks. MAnycast2's advantage is speed and coverage but suffers from problems with accuracy, while iGreedy is highly accurate but slower using author-defined probing rates and costlier. In this paper we address the shortcomings of both systems and present LACeS (Longitudinal Anycast Census System). Taking MAnycast2 as a basis, we completely redesign its measurement pipeline, and add support for distributed probing, additional protocols (DNS over UDP, TCP SYN/ACK, and IPv6) and latency measurements similar to iGreedy. We validate LACeS on an anycast testbed with 32 globally distributed nodes, compare against an external anycast production deployment, extensive latency measurements with RIPE Atlas and cross-check over 60% of detected anycast using operator ground truth that shows LACeS achieves high accuracy. Finally, we provide a longitudinal analysis of anycast, covering 17+ months, showing LACeS achieves high precision. We make continual daily LACeS censuses available to the community and release the source code of the tool under a permissive open source license.

Figures (14)

• Figure 1: Using latency measurements from four VPs (SLC, PHX, BNA, CHS) for GCD-based anycast site (LAS, ATL) detection. These latency measurements imply at least two sites. iGreedy infers the location of each site using the highest populated city in the intersection.
• Figure 2: Using an anycast setup (light blue nodes -- SLC, PHX, BNA, CHS) to detect anycast. Probes to a unicast site (orange lines to OKC) results in responses to a single VP (BNA), while probes to anycast sites (green lines to LAS and ATL) result in responses to multiple VPs (red lines to SLC, CHS).
• Figure 3: Overview of the measurement pipeline.
• Figure 4: Roadmap of analysis conducted, detailing the date at which it was performed, the section in which we discuss the results, and the associated figures & tables.
• Figure 5: FPs found by number of VPs receiving using MAnycast2 with a 13- and 1-minute interval between probes, and LACeS with a 1- and 0-second interval between probes.