Self-Stabilizing Virtual Machine Hypervisor Architecture for Resilient Cloud

Abstract

This paper presents the architecture for a selfstabilizing hypervisor able to recover itself in the presence of Byzantine faults regardless of the state it is currently in. Our architecture is applicable to wide variety of underlying hardware and software and does not require augmenting computers with special hardware. The actions representing defense and recovery strategies can be specified by a user. We describe our architecture in OS-independent terms, thus making it applicable to various virtualization infrastructures. We also provide a prototype extending the Linux-based hypervisor KVM with the self-stabilizing functionality. These features allow augmenting KVM with robustness functionality in the coming stages and moving to cloud management system architectures such as OpenStack to support more industrial scenarios.

On Resilience in Cloud Computing

“…Binun et.al focus upon the resilience of the VMM [21]. They present a novel self-stabilising hypervisor for increased robustness against malicious faults.…”

A Self-stabilizing Control Plane for the Edge and Fog Ecosystems

“…Algorithms 1 and 2 provide a high-level description of our solution, and the details appear in algorithms 3, 4, 5 and 6, which implement the proposed solution to the above task specifications by considering the code to be executed by the Cloud, IoT devices, cloudlets, and respectively, the emulators of the replicated state-machine. Algorithm 3 assumes the availability of a self-stabilizing cloud infrastructure, such as [7].…”

Perspectives on Resilience in Cloud Computing: Review and Trends

“…The latter research forms part of the focus of our research, where the intention is to shed light on some promising trends in cloud computing resilience with completely alternative architectures. This is further discussed in the next section Reactive error ranking and appropriate technique [11] Reactive introspection [12] Proactive high diversity for replica storage [13] Proactive memory stored backups Service Composition Graph based, interdepdency [15] Service Composition Agent-based [16] Low Level Diversity data-centre [17] Quality Adjustment brownout [18] Diversity Replicas Diversity Structure height Organisation VM Scheduling for resilience [20] Disaster Recovery Storage [21] Reactive reset upon fault Hypervisor [22] Proactive Diversity Hypervisor [23] QoR Evaluation Proactive service provisioning Diversity Geo-distribution [25] VM to PHY Mapping Design [26] VM to PHY Mapping Backup links [27] Redundancy VNet backup links III. ALTERNATIVE ARCHITECTURES FOR CLOUD COMPUTING RESILIENCE Some work will choose to encourage a conventionally different cloud architecture in order to provide increased resilience (Fig.…”