VNsnap: Taking Snapshots of Virtual Networked Infrastructures in the Cloud

A virtual networked infrastructure (VNI) consists of virtual machines (VMs) connected by a virtual network. Created for individual users on a shared cloud infrastructure, VNIs reflect the concept of "Infrastructure as a Service” (IaaS) as part of the emerging cloud computing paradigm. The ability to take snapshots of an entire VNI—including images of the VMs with their execution, communication, and storage states—yields a unique approach to reliability as a VNI snapshot can be used to restore the operation of the entire virtual infrastructure. We present VNsnap, a system that takes distributed snapshots of VNIs. Unlike many existing distributed snapshot/checkpointing solutions, VNsnap does not require any modifications to the applications, libraries, or (guest) operating systems (OSs) running in the VMs. Furthermore, by performing much of the snapshot operation concurrently with the VNI's normal operation, VNsnap incurs only seconds of downtime. We have implemented VNsnap on top of Xen. Our experiments with real-world parallel and distributed applications demonstrate VNsnap's effectiveness and efficiency.