Residency-Aware Virtual Machine Communication Optimization: Design Choices and Techniques

Abstract

Abstract-Network I/O workloads are dominating in many data centers and cloud computing environments today. One way to improve inter Virtual Machine (VM) communication efficiency is to support co-resident VM communication by using shared memory based approaches and to resort to the traditional TCP/IP for inter-VM communications between VMs that are located on different physical hosts. Although a number of independent efforts are dedicated to improving communication efficiency between co-resident VMs, they differ from one another in terms of how the inter-VM communication optimization is carried out and where in the software stack the shared memory channel is established. In this paper, we provide an in-depth overview of the design choices and techniques for optimizing the performance of the co-resident inter-VM communication, with dual objectives. First, we describe the core design guidelines and key issues for optimizing inter-VM communication by using shared memory based mechanisms. Typical issues include choices of implementation layer in the software stack, seamless agility for VM live migration and VM dynamic deployment support, multilevel transparency. Second, we conduct a comprehensive analysis of representative state-of-the-art research efforts and implementation techniques based on the core design guidelines. We also give an analysis of future requirements in advanced features such as reliability, security and stability. The research reported in this paper not only provides the reference for developing the next generation of inter-VM communication optimization mechanisms, but also offers opportunities for both cloud infrastructure providers and cloud service consumers to improve inter-VM communication efficiency in virtualized platforms.

Hypertracing: Tracing Through Virtualization Layers

“…Moreover, to allow efficient and fast communication, the guest OS and the hypervisor are capable of using an optimized I/O interface, rather than depending on TCP/IP. This mechanism is called paravirtualization [20].…”

Interdomain I/O Optimization in Virtualized Sensor Networks

“…The communication between VMs on a single physical machine becomes very complex and frequent. Currently, since communication performance between different VMs on a single physical machine is affected by various aspects, the overhead can largely degrade communication performance [36,37,38]. Memory virtualization is one of the most common virtualization technologies, and the required user memory space may be much larger than the actual size of the machine’s memory.…”

Residency Aware Inter-VM Communication in Virtualized Cloud: Performance Measurement and Analysis

“…For example [2,5,6,] support only OS kernel transparency, [4] supports both OS kernel and VMM transparency but requires modifications in systems call and user library, [11] supports user-level and VMM level transparency but requires modifications of OS kernels , only [8,9,12] support all three levels of transparency. It is recognized that the capability of detecting residency-aware inter-VM co mmunication and the capability of automatic switching between communicating through network stack and through shared memory channel play a fundamental role in providing all three level transparency and the support for VM live migration and live deployment [10]. However, none has examined whether and how much the overhead of residence-aware VM detection and communication channel switch may impact on the performance of remote inter-VM communication.…”

“…Even though it is well recognized that such residencyaware inter-VM commun ication mechanisms hold the potential for speeding up the co-resident VM co mmunication and thus the overall network I/O performance in Cloud based data centers [4,8,9,10,11,12], there lacks of in-depth measurement study with quantitative and qualitative analysis on challenging issues that are critical for real world deployment in operational Cloud based data centers. Here are some examples:  How much performance gains can a residency-aware shared memory inter-VM co mmunication mechanism provide under different mixtures of local and remote network I/O workloads?…”