Brief Announcement: A Randomized Algorithm for Label Assignment in Dynamic Networks

In large-scale networking environments, such as data centers, a key difficulty is the assignment of labels to network elements. Labels can be assigned statically, e.g. MAC-addresses in traditional Layer 2 networks, or by a central authority as in DHCP in Layer 3 networks. On the other hand, networks requiring a dynamic solution often use a Consensus-based state machine approach. While designing Alias [2], a protocol for automatically assigning hierarchically meaningful addresses in data center networks, we encountered an instance of label assignment with entirely different requirements. In this case, the rules for labels depend on connectivity, and connectivity (and hence, labels) changes over time. Thus, neither static assignment nor a state machine approach is ideal. In the context of large scale data center networks, practical constraints are important. A centralized solution introduces a single point of failure and necessitates either flooding or a separate out-of-band control network to establish communication between the centralized component and all network elements; this is problematic at the scale of a data center. We also require a solution that scales, is robust in the face of miswirings and transient startup conditions, and has low message overhead and convergence time. To this end, we specify the Label Selection Problem (LSP), which is the problem of practical label assignment in data center networks.