YOLO: Speeding Up VM and Docker Boot Time by Reducing I/O Operations

Abstract

Although this comes as a surprise, the time to boot a Dockerbased container can last as long as a virtual machine in high consolidated cloud scenarios. Because this time is critical as boot duration defines how an application can react w.r.t. demands' fluctuations (horizontal elasticity), we present in this paper the YOLO mechanism (You Only Load Once). YOLO reduces the number of I/O operations generated during a boot process by relying on a boot image abstraction, a subset of the VM/container image that contains data blocks necessary to complete the boot operation. Whenever a VM or a container is booted, YOLO intercepts all read accesses and serves them directly from the boot image, which has been locally stored on fast access storage devices (e.g., memory, SSD, etc.). In addition to YOLO, we show that another mechanism is required to ensure that files related to VM/container management systems remain in the cache of the host OS. Our results show that the use of the two technics can speed up boot duration 2-13 times for VMs and 2 times for containers. The benefit on containers is limited due to internal choices of the docker design. We underline that our proposal can be easily applied to other types of virtualization (e.g., Xen) and containerization because it does not require intrusive modifications on the virtualization/container management system nor the base image structure.

Toward safe and efficient reconfiguration with Concerto

“…Dependency declarations and parallelism -Minimizing the execution time of a reconfiguration, assuming that reconfiguration actions have already been individually optimized [58,59], comes down to increasing the level of parallelism between these actions. This is directly correlated to the ability to express dependencies between operations of several components, in other words between the lifecycles of components.…”

Predictable Efficiency for Reconfiguration of Service-Oriented Systems with Concerto

“…This work focuses on performance in reconfiguration and its estimation ahead of execution. Increasing performance can be done in many ways such as command optimizations [4] or by enhancing data management [5]. In this paper we focus on two generic aspects which can increase performance: life-cycle management and parallelism.…”