Response Time and Availability Study of RAFT Consensus in Distributed SDN Control Plane

Abstract

Software Defined Networking promises unprecedented flexibility and ease of network operations. While flexibility is an important factor when leveraging advantages of a new technology, critical infrastructure networks also have stringent requirements on network robustness and control plane delays. Robustness in the SDN control plane is realized by deploying multiple distributed controllers, formed into clusters for durability and fast-failover purposes. However, the effect of the controller clustering on the total system response time is not well investigated in current literature. Hence, in this work we provide a detailed analytical study of the distributed consensus algorithm RAFT, implemented in OpenDaylight and ONOS SDN controller platforms. In those controllers, RAFT implements the data-store replication, leader election after controller failures and controller state recovery on successful repairs. To evaluate its performance, we introduce a framework for numerical analysis of various SDN cluster organizations w.r.t. their response time and availability metrics. We use Stochastic Activity Networks for modeling the RAFT operations, failure injection and cluster recovery processes, and using real-world experiments, we collect the rate parameters to provide realistic inputs for a representative cluster recovery model. We also show how a fast rejuvenation mechanism for the treatment of failures induced by software errors can minimize the total response time experienced by the controller clients, while guaranteeing a higher system availability in the long-term.

Impact of Adaptive Consistency on Distributed SDN Applications: An Empirical Study

“…This limitation relates to the requirement of ensuring data consistency in the case of network partitions, an invariant feasible only when a majority of nodes are involved in confirming the transactions [22], [23]. In the best case, the cluster operates at the speed of the leader, and in the worst case, at the speed of the slowest follower [8].…”

“…With the concept of state replication, the SDN controller instances replicate their data store contents to other members that take part in a logical controller cluster, using a state distribution protocol of choice (e.g. RAFT [7], [8] of a controller is suspected, another replica from its cluster is able to take over and continue to serve future application's requests. The selection of the consistency model leveraged by the replication process affects the incurred synchronization overhead in terms of the resulting packet load, the experienced commit response times and the processing order of commits.…”

“…in RAFT), all requests are serialized by a cluster leader, in order to provide for a consistent data store view across all cluster followers. Thus, with SC, a large distributed system consisting of multiple controller replicas, is effectively constrained into a monolithic system where each data store modification incurs a minimum of two message rounds and a linear message complexity (in the case of a stable leader) in order to synchronize the controller views [8]- [10].…”

BFT Protocols for Heterogeneous Resource Allocations in Distributed SDN Control Plane

“…INTRODUCTION AND PROBLEM STATEMENT Software Defined Networking (SDN) centralizes the decision-making in a dedicated controller component. Concepts for achieving crash-fault-tolerance and scalable operation of the controller have been presented in the past [1], [2]. By means of a logical distribution of controller replicas and the state synchronization, the controller instances are able to synchronize the results of their individual computations and come to consistent decisions independent of the instance that handled the client request.…”

“…Namely, the application execution is handled by each controller instance in the cluster. In heterogeneous environments, where particular controller replicas can be assigned a higher resource set compared to the others, this leads to an under-utilization of fast replicas, as the system progresses at best at the speed of the |C|+1 2 +1 fastest replica (|C| is the number of deployed controllers) [2]. Second, these BFT implementations rely on reaching a successful agreement about the sequence number mapping for each arriving client request, prior to its actual execution.…”