SEER: Performance-Aware Leader Election in Single-Leader Consensus

04/03/2021
by   Ermin Sakic, et al.
0

Modern stateful web services and distributed SDN controllers rely on log replication to omit data loss in case of fail-stop failures. In single-leader execution, the leader replica is responsible for ordering log updates and the initiation of distributed commits, in order to guarantee log consistency. Network congestions, resource-heavy computation, and imbalanced resource allocations may, however, result in inappropriate leader election and increased cluster response times. We present SEER, a logically centralized approach to performance prediction and efficient leader election in leader-based consensus systems. SEER autonomously identifies the replica that minimizes the average cluster response time, using prediction models trained dynamically at runtime. To balance the exploration and exploitation, SEER explores replicas' performance and updates their prediction models only after detecting significant system changes. We evaluate SEER in a traffic management scenario comprising [3..7] Raft replicas, and well-known data-center and WAN topologies. Compared to the Raft's uniform leader election, SEER decreases the mean control plane response time by up to  32 election procedure and a slight increase in leader reconfiguration frequency, the latter being tunable with a guaranteed upper bound. No safety properties of Raft are invalidated by SEER.

READ FULL TEXT

page 1

page 3

page 9

research
01/14/2021

Leader Confirmation Replication for Millisecond Consensus in Geo-distributed Private Chains

Geo-distributed private chain and database have created higher performan...
research
06/16/2021

Velos: One-sided Paxos for RDMA applications

Modern data centers are becoming increasingly equipped with RDMA-capable...
research
05/21/2019

Linearizable State Machine Replication of State-Based CRDTs without Logs

General solutions of state machine replication have to ensure that all r...
research
02/18/2022

ESCAPE to Precaution against Leader Failures

Leader-based consensus protocols must undergo a view-change phase to ele...
research
06/09/2022

Discussion of Quantum Consensus Algorithms

Leader election is a crucial process in many areas such as cloud computi...
research
06/26/2018

Adaptive guaranteed-performance consensus design for high-order multiagent systems

The current paper addresses the distributed guaranteed-performance conse...
research
08/31/2020

Lecture Notes on Leader-based Sequence Paxos – An Understandable Sequence Consensus Algorithm

Agreement among a set of processes and in the presence of partial failur...

Please sign up or login with your details

Forgot password? Click here to reset