A Reliable and Cost-Efficient Auto-Scaling System for Web Applications Using Heterogeneous Spot Instances

TL;DR

This paper presents a fault-tolerant auto-scaling system for web applications using heterogeneous spot instances, achieving high availability and cost savings despite the inherent risks of spot market volatility.

Contribution

It introduces a novel fault-tolerant model and auto-scaling policies specifically designed for spot instances in web applications, validated through simulation and real-world experiments.

Findings

Significant cost reduction compared to traditional methods

Maintains satisfactory QoS in response time and availability

Effective utilization of heterogeneous spot instances

Abstract

Cloud providers sell their idle capacity on markets through an auction-like mechanism to increase their return on investment. The instances sold in this way are called spot instances. In spite that spot instances are usually 90% cheaper than on-demand instances, they can be terminated by provider when their bidding prices are lower than market prices. Thus, they are largely used to provision fault-tolerant applications only. In this paper, we explore how to utilize spot instances to provision web applications, which are usually considered availability-critical. The idea is to take advantage of differences in price among various types of spot instances to reach both high availability and significant cost saving. We first propose a fault-tolerant model for web applications provisioned by spot instances. Based on that, we devise novel auto-scaling polices for hourly billed cloud markets. We implemented the proposed model and policies both on a simulation testbed for repeatable validation and Amazon EC2. The experiments on the simulation testbed and the real platform against the benchmarks show that the proposed approach can greatly reduce resource cost and still achieve satisfactory Quality of Service (QoS) in terms of response time and availability.