HT-Paxos: High Throughput State-Machine Replication Protocol for Large Clustered Data Centers

TL;DR

HT-Paxos is a novel Paxos variant designed for large clustered data centers, significantly offloading the leader to enhance throughput and scalability while maintaining low latency.

Contribution

The paper introduces HT-Paxos, a new Paxos-based protocol that greatly reduces leader workload, improving throughput and scalability in large data center environments.

Findings

HT-Paxos achieves higher throughput than traditional Paxos variants.

It maintains low latency suitable for data center applications.

The protocol scales effectively with large cluster sizes.

Abstract

Paxos is a prominent theory of state machine replication. Recent data intensive Systems those implement state machine replication generally require high throughput. Earlier versions of Paxos as few of them are classical Paxos, fast Paxos and generalized Paxos have a major focus on fault tolerance and latency but lacking in terms of throughput and scalability. A major reason for this is the heavyweight leader. Through offloading the leader, we can further increase throughput of the system. Ring Paxos, Multi Ring Paxos and S-Paxos are few prominent attempts in this direction for clustered data centers. In this paper, we are proposing HT-Paxos, a variant of Paxos that one is the best suitable for any large clustered data center. HT-Paxos further offloads the leader very significantly and hence increases the throughput and scalability of the system. While at the same time, among high throughput state-machine replication protocols, HT-Paxos provides reasonably low latency and response time.