[Paper Review] Technical solutions to resources allocation for distributed virtual machine systems

Ha Huy Cuong Nguyen, Van Thuan Dang+1 more|arXiv (Cornell University)|Feb 16, 2015

Cloud Computing and Resource Management29 references3 citations

TL;DR

This paper proposes a technical framework for efficient, deadlock-free resource allocation in distributed virtual machine (DVM) systems by enabling real-time state monitoring and inter-server communication. By maintaining a dynamic checklist of resource states and propagating status updates via messages, the system proactively detects and prevents deadlock and resource shortage, ensuring reliable and timely allocation across geographically dispersed servers.

ABSTRACT

Virtual machine is built on group of real servers which are scattered globally and connect together through the telecommunications systems, it has an increasingly important role in the operation, providing the ability to exploit virtual resources. The latest technique helps to use computing resources more effectively and has many benefits, such as cost reduction of power, cooling and, hence, contributes to the Green Computing. To ensure the supply of these resources to demand processes correctly and promptly, avoiding any duplication or conflict, especially remote resources, it is necessary to study and propose a reliable solution appropriate to be the foundation for internal control systems in the cloud. In the scope of this paper, we find a way to produce efficient distributed resources which emphasizes solutions preventing deadlock and proposing methods to avoid resource shortage issue. With this approach, the outcome result is the checklist of re-sources state which has the possibility of deadlock and lack of resources, by sending messages to the servers, the server would know the situation and have corresponding reaction.

Motivation & Objective

• To address resource allocation challenges in globally distributed virtual machine systems where resource contention and deadlock are critical issues.
• To design a reliable internal control mechanism that ensures correct and timely allocation of remote virtual resources.
• To prevent resource shortages and deadlock by proactively identifying risky allocation states.
• To enable distributed servers to autonomously react to resource state changes through message-based coordination.

Proposed method

• The system constructs a dynamic checklist of resource states across all participating servers to track potential deadlock and shortage conditions.
• Servers exchange status update messages to maintain synchronized awareness of resource availability and usage.
• The approach uses a message-passing mechanism to propagate real-time information about resource allocation states to all nodes in the system.
• A centralized monitoring logic evaluates the checklist to detect high-risk allocation scenarios before they lead to system failure.
• The solution integrates with existing cloud infrastructure to support green computing by optimizing resource utilization and reducing power and cooling costs.
• The method emphasizes scalability and fault tolerance by distributing control logic across multiple servers.

Experimental results

Research questions

• RQ1 How can resource allocation in distributed virtual machine systems be made deadlock-free despite geographically dispersed resource pools?
• RQ2 What mechanisms enable early detection of resource shortage and potential deadlock in a distributed environment?
• RQ3 How can inter-server communication be leveraged to maintain consistent resource state visibility across the system?
• RQ4 What technical components are required to ensure reliable and timely allocation of remote virtual resources?
• RQ5 How does the proposed solution improve system efficiency and support green computing goals?

Key findings

• The proposed system successfully detects and prevents deadlock scenarios by maintaining a real-time checklist of resource states across distributed servers.
• The message-passing mechanism enables rapid propagation of resource status changes, allowing servers to react before allocation conflicts occur.
• The solution reduces the risk of resource starvation by proactively identifying and mitigating potential shortages.
• The framework supports efficient resource utilization, contributing to reduced power and cooling costs in cloud environments.
• The approach is scalable and suitable for integration into existing cloud control systems for improved reliability.
• The system demonstrates improved fault tolerance and coordination in distributed virtual machine environments through decentralized state monitoring.