Live migration of virtual machines (VM) across distinct physical hosts is an important feature of virtualization technology for maintenance, load-balancing and energy reduction, especially so for data centers operators and cluster service providers. Several techniques have been proposed to reduce the downtime of the VM being transferred, often at the expense of the total migration time. In this work, we present a technique to reduce the total time required to migrate a running VM from one host to another while keeping the downtime to a minimum. Based on the observation that modern operating systems use the better part of the physical memory to cache data from secondary storage, our technique tracks the VM's I/O operations to the network-attached storage device and maintains an updated mapping of memory pages that currently reside in identical form on the storage device. During the iterative pre-copy live migration process, instead of transferring those pages from the source to the target host, the memory-to-disk mapping is sent to the target host which then fetches the contents directly from the network-attached storage device. We have implemented our approach into the Xen hypervisor and ran a series of experiments with Linux HVM guests. On average, the presented technique shows a reduction of up over 30% on average of the total transfer time for a series of benchmarks.
Checkpointing, i.e., recording the volatile state of a virtual machine (VM) running as a guest in a virtual machine monitor (VMM) for later restoration, includes storing the memory available to the VM. Typically, a full image of the VM's memory along with processor and device states are recorded. With guest memory sizes of up to several gigabytes, the size of the checkpoint images becomes more and more of a concern. In this work we present a technique for fast and space-efficient checkpointing of virtual machines. In contrast to existing methods, our technique eliminates redundant data and stores only a subset of the VM's memory pages. Our technique transparently tracks I/O operations of the guest to external storage and maintains a list of memory pages whose contents are duplicated on non-volatile storage. At a checkpoint, these pages are excluded from the checkpoint image. We have implemented the proposed technique for paravirtualized as well as fully-virtualized guests in the Xen VMM. Our experiments with a paravirtualized guest (Linux) and two fullyvirtualized guests (Linux, Windows) show a significant reduction in the size of the checkpoint image as well as the time required to complete the checkpoint. Compared to the current Xen implementation, we achieve, on average, an 81% reduction in the stored data and a 74% reduction in the time required to take a checkpoint for the paravirtualized Linux guest. In a fully-virtualized environment runningWindows and Linux guests, we achieve a 64% reduction of the image size along with a 62% reduction in checkpointing time.
26 June 2014 in IEEE Transactions on Parallel and Distributed Systems
To exploit the abundant computational power of the world's fastest supercomputers, an even workload distribution to the typically heterogeneous compute devices is necessary. While relatively accurate performance models exist for conventional CPUs, accurate performance estimation models for modern GPUs do not exist. This paper presents two accurate models for modern GPUs: a sampling-based linear model, and a model based on machine-learning (ML) techniques which improves the accuracy of the linear model and is applicable to modern GPUs with and without caches. We first construct the sampling-based linear model to predict the runtime of an arbitrary OpenCL kernel. Based on an analysis of NVIDIA GPUs' scheduling policies we determine the earliest sampling points that allow an accurate estimation. The linear model cannot capture well the significant effects that memory coalescing or caching as implemented in modern GPUs have on performance. We therefore propose a model based on ML techniques that takes several compiler-generated statistics about the kernel as well as the GPU's hardware performance counters as additional inputs to obtain a more accurate runtime performance estimation for modern GPUs. We demonstrate the effectiveness and broad applicability of the model by applying it to three different NVIDIA GPU architectures and one AMD GPU architecture. On an extensive set of OpenCL benchmarks, on average, the proposed model estimates the runtime performance with less than 7 percent error for a second-generation GTX 280 with no on-chip caches and less than 5 percent for the Fermi-based GTX 580 with hardware caches. On the Kepler-based GTX 680, the linear model has an error of less than 10 percent. On an AMD GPU architecture, Radeon HD 6970, the model estimates with 8 percent of error rates. The proposed technique outperforms existing models by a factor of 5 to 6 in terms of accuracy.