CloudSim 7G: An Integrated Toolkit for Modeling and Simulation of Future Generation Cloud Computing Environments

Remo Andreoli, Jie Zhao, Tommaso Cucinotta, Rajkumar Buyya

Abstract

Cloud Computing has established itself as an efficient and cost-effective paradigm for the execution of web-based applications and scientific workloads that need elasticity and on-demand scalability capabilities. However, the evaluation of novel resource provisioning and management techniques is a major challenge due to the complexity of large-scale data centers. Therefore, Cloud simulators are an essential tool for academic and industrial researchers to investigate the effectiveness of novel algorithms and mechanisms in large-scale scenarios. This paper proposes CloudSim 7G, the seventh generation of CloudSim, which features a re-engineered and generalized internal architecture to facilitate the integration of multiple CloudSim extensions within the same simulated environment. As part of the new design, we introduced a set of standardized interfaces to abstract common functionalities and carried out extensive refactoring and refinement of the codebase. The result is a substantial reduction in lines of code with no loss in functionality, significant improvements in run-time performance and memory efficiency (up to 25% less heap memory allocated), as well as increased flexibility, ease-of-use, and extensibility of the framework. These improvements benefit not only CloudSim developers but also researchers and practitioners using the framework for modeling and simulating next-generation Cloud Computing environments.

Introduction

Since its debut in 2008, CloudSim has become the de-facto standard for simulating cloud data centers. However, as cloud computing evolved to include Software-Defined Networking (SDN), containers, and serverless models, the ecosystem became fragmented. Researchers created "single-feature" extensions (modules) that were often incompatible with one another.

The evolution of CloudSim offerings and concepts from its inception in 2008 to the current 2024 landscape.

As shown in the timeline above, the proliferation of concepts—from Virtual Machines to Quantum-as-a-service—created a significant "architectural debt." For example, simulating a container-based application running on a power-aware virtual machine within a specific network topology often required "hacking" disparate modules together, leading to massive code duplication and performance bottlenecks. CloudSim 7G addresses this by providing a unified base layer where these technologies can coexist and interact seamlessly.

Architectural Refactoring and Methods

The development of CloudSim 7G involved a massive refactoring of the existing codebase, removing over 13,000 lines of redundant code. Key technical changes include:

  • Standardized Interfaces: The introduction of HostEntity and GuestEntity interfaces abstracts the behavior of physical machines and virtual components. This allows the simulator to treat VMs and containers through a common logic, enabling native support for nested virtualization (e.g., containers inside VMs).
  • Unified Selection Policies: Placement and migration logic, previously scattered across different packages, were consolidated into a single SelectionPolicy interface. This reduced 26 redundant classes down to just 11.
  • Engine Optimization: The core simulation engine was modernized by replacing inefficient custom linked lists (with $O(n)$ time complexity for event dispatching) with Java’s PriorityQueue, achieving $O(\log n)$ complexity.
  • Virtualization Overhead Modeling: A new parameter was introduced to model the real-world performance "tax" that occurs when data traverses virtual interfaces, a crucial detail for accurate nested virtualization studies.

Performance Evaluation

To validate the performance gains, the authors evaluated CloudSim 7G using real-world CPU utilization data from PlanetLab Workload Traces and complex task dependencies modeled via Scientific Workflows (DAX format).

The benchmarks comparing CloudSim 6G to 7G revealed significant efficiencies:

  • Memory Efficiency: CloudSim 7G utilizes up to 25% less total heap memory in large-scale simulations.
  • Execution Speed: Run-time performance improved by 5% to 12% across various standard power-aware algorithms.
  • Integration Capability: The authors successfully demonstrated a hybrid case study involving networked cloudlets, containers, and virtualization overhead—a scenario that was previously nearly impossible to simulate in a single environment.

Discussion and Key Insights

By removing over 13,000 lines of code, the framework's flexibility actually increased. For instance, the ContainerCloudSim module saw a 64% reduction in lines of code through deduplication. Furthermore, moving to modern Java standards (JDK 21) and replacing synchronized data structures—which were unnecessary for the single-threaded core engine—provided an "out-of-the-box" performance boost for researchers upgrading from older versions.

Conclusion and Future Work

While CloudSim 7G represents a significant leap forward in cloud simulation technology, several areas for future improvement remain. The authors identify the need for more realistic multi-core CPU scheduling models and the introduction of dedicated interfaces for AI/ML accelerators to support modern resource orchestration research. Additionally, moving from MIPS-based performance metrics toward clock frequency-based modeling is suggested to achieve higher simulation fidelity.

Source: Andreoli, R., et al. "CloudSim 7G: An Integrated Toolkit for Modeling and Simulation of Future Generation Cloud Computing Environments." arXiv:2408.13386v4, 2025.

Summary updated

Abstract

Cloud Computing has established itself as an efficient and cost-effective paradigm for the execution of web-based applications, and scientific workloads, that need elasticity and on-demand scalability capabilities. However, the evaluation of novel resource provisioning and management techniques is a major challenge due to the complexity of large-scale data centers. Therefore, Cloud simulators are an essential tool for academic and industrial researchers, to investigate the effectiveness of novel algorithms and mechanisms in large-scale scenarios. This paper proposes CloudSim 7G, the seventh generation of CloudSim, which features a re-engineered and generalized internal architecture to facilitate the integration of multiple CloudSim extensions within the same simulated environment. As part of the new design, we introduced a set of standardized interfaces to abstract common functionalities and carried out extensive refactoring and refinement of the codebase. The result is a substantial reduction in lines of code with no loss in functionality, significant improvements in run-time performance and memory efficiency (up to 25% less heap memory allocated), as well as increased flexibility, ease-of-use, and extensibility of the framework. These improvements benefit not only CloudSim developers but also researchers and practitioners using the framework for modeling and simulating next-generation Cloud Computing environments.
BibTeX
@inproceedings{andreoli2025cloudsim,
  author = {Remo Andreoli and Jie Zhao and Tommaso Cucinotta and Rajkumar Buyya},
  title = {{CloudSim 7G: An Integrated Toolkit for Modeling and Simulation of Future Generation Cloud Computing Environments}},
  booktitle = {Software},
  year = {2025},
  doi = {10.1002/spe.3413},
}

Loading related content...

Figure