Towards a global IP anycast service

Abstract

IP anycast, with its innate ability to find nearby resources in a robust and efficient fashion, has long been considered an important means of service discovery. The growth of P2P applications presents appealing new uses for IP anycast. Unfortunately, IP anycast suffers from serious problems: it is very hard to deploy globally, it scales poorly by the number of anycast groups, and it lacks important features like load-balancing. As a result, its use is limited to a few critical infrastructure services such as DNS root servers. The primary contribution of this paper is a new IP anycast architecture, PIAS, that overcomes these problems while largely maintaining the strengths of IP anycast. PIAS makes use of a proxy overlay that advertises IP anycast addresses on behalf of group members and tunnels anycast packets to those members. The paper presents a detailed design of PIAS and evaluates its scalability and efficiency through simulation. We also present preliminary measurement results on anycasted DNS root servers that suggest that IP anycast provides good affinity. Finally, we describe how PIAS supports two important P2P and overlay applications.

Distributed Service Provisioning Using Stateful Anycast Communications

“…According to Ballani et al [5], the main reason for this is the lack of IP routing scalability inherent to native anycast. First, IP anycast routes cannot be aggregated and widespread adoption would lead to an explosive growth of IP routing tables.…”

“…Taking into account both the strengths and weaknesses of IP anycast, we propose an anycast overlay architecture based on PIAS [5] to realize a transparent and scalable service provisioning platform.…”

“…The design objectives for a regional anycast proxy infrastructure supporting session-based network services differ from a global anycast overlay infrastructure such as PIAS [5] or OASIS [8]. In addition to the PIAS objectives, we wish explicit session support in the proxy routers and a dynamic, sessionbased bonding between client proxies and server proxies (and hence, between clients and servers).…”

“…Since the described architecture is based on PIAS, the design goals outlined by Ballani et al [5] are also achieved. More specifically, the overlay architecture drastically improves IP anycast routing scalability because a single IP range can be allocated for all anycast services, thereby aggregating all anycast services into a single routing table entry.…”

“…In a number of ways, the proposed anycast proxy architecture resembles PIAS (Proxy IP Anycast Service) [5] and most PIAS features and benefits remain valid for our architecture. Whereas Ballani et al focus on global routing scalability and motivate part of the PIAS design by relying on BGP route stability, we tailor our anycast proxy architecture to the needs of a service provisioning platform, with explicit support for session-based communications.…”