Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Simulating Dynamic Cloud Marketspaces: Modeling Spot Instance Behavior and Scheduling with CloudSim Plus

Christoph Goldgruber, Benedikt Pittl, Erich Schikuta

TL;DR

This paper extends CloudSim Plus to realistically model spot (preemptible) instances, capturing interruption, hibernation, and dynamic reprovisioning within VM lifecycles. It adapts the HLEM-VMP allocation algorithm to dynamic spot markets, aiming to reduce interruptions and balance host utilization. Through synthetic experiments and Google Cluster Trace-based evaluation, the approach demonstrates fewer spot interruptions and reduced maximum interruption durations, highlighting cost-performance trade-offs in volatile cloud markets. The work delivers a reusable simulation framework and analytical insights for interruption-aware resource management in cloud computing.

Abstract

The increasing reliance on dynamic pricing models, such as spot instances, in public cloud environments presents new challenges for workload scheduling and reliability. While these models offer cost advantages, they introduce volatility and uncertainty that are not fully addressed by current allocation algorithms or simulation tools. This work contributes to the modeling and evaluation of such environments by extending the CloudSim Plus simulation framework to support realistic spot instance lifecycle management, including interruption, termination, hibernation, and reallocation. The enhanced simulator is validated using synthetic scenarios and large-scale simulations based on the Google Cluster Trace dataset. Building on this foundation, the HLEM-VMP allocation algorithm, originally proposed in earlier research, was adapted to operate under dynamic spot market conditions. Its performance was evaluated against baseline allocation strategies to assess its efficiency and resilience in volatile workload environments. The comparison demonstrated a reduction in the number of spot instance interruptions as well as a decrease in the maximum interruption duration. Overall, this work provides both a simulation framework for simulating dynamic cloud behavior and analytical insights into virtual machine allocation performance and market risk, contributing to more robust and cost-effective resource management in cloud computing.

Simulating Dynamic Cloud Marketspaces: Modeling Spot Instance Behavior and Scheduling with CloudSim Plus

TL;DR

This paper extends CloudSim Plus to realistically model spot (preemptible) instances, capturing interruption, hibernation, and dynamic reprovisioning within VM lifecycles. It adapts the HLEM-VMP allocation algorithm to dynamic spot markets, aiming to reduce interruptions and balance host utilization. Through synthetic experiments and Google Cluster Trace-based evaluation, the approach demonstrates fewer spot interruptions and reduced maximum interruption durations, highlighting cost-performance trade-offs in volatile cloud markets. The work delivers a reusable simulation framework and analytical insights for interruption-aware resource management in cloud computing.

Abstract

The increasing reliance on dynamic pricing models, such as spot instances, in public cloud environments presents new challenges for workload scheduling and reliability. While these models offer cost advantages, they introduce volatility and uncertainty that are not fully addressed by current allocation algorithms or simulation tools. This work contributes to the modeling and evaluation of such environments by extending the CloudSim Plus simulation framework to support realistic spot instance lifecycle management, including interruption, termination, hibernation, and reallocation. The enhanced simulator is validated using synthetic scenarios and large-scale simulations based on the Google Cluster Trace dataset. Building on this foundation, the HLEM-VMP allocation algorithm, originally proposed in earlier research, was adapted to operate under dynamic spot market conditions. Its performance was evaluated against baseline allocation strategies to assess its efficiency and resilience in volatile workload environments. The comparison demonstrated a reduction in the number of spot instance interruptions as well as a decrease in the maximum interruption duration. Overall, this work provides both a simulation framework for simulating dynamic cloud behavior and analytical insights into virtual machine allocation performance and market risk, contributing to more robust and cost-effective resource management in cloud computing.

Paper Structure

This paper contains 78 sections, 11 equations, 16 figures, 3 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. State of the Art
  3. Enabling Technologies
  4. Common Purchase Models
  5. On-Demand Instances.
  6. Reserved Instances.
  7. Spot / Preemptible Instances.
  8. Dedicated Hosts.
  9. Cloud Simulation
  10. VM Allocation Algorithms
  11. Research Questions
  12. Requirement Analysis
  13. Background
  14. Cloud Datacenters
  15. Virtual Machine Placement
  16. ...and 63 more sections

Figures (16)

  • Figure 1: Execution flow of a discrete event simulation law2013simmodeling.
  • Figure 2: Spot instance lifecycle based on EC2, showing events, status messages, and user actions pham2018performance.
  • Figure 3: Component diagram of the spot instance modeling extension. Yellow: original modules; orange: new components.
  • Figure 4: Spot instance lifecycle state transitions.
  • Figure 5: Example Output -- RestartingInterruptedSpot (DynamicVmTableBuilder)
  • ...and 11 more figures