Virtualization can provide significant benefits in data centers by enabling virtual machine migration to eliminate hotspots. We present Sandpiper, a system that automates the task of monitoring and detecting hotspots, determining a new mapping of physical to virtual resources and initiating the necessary migrations. Sandpiper implements a black-box approach that is fully OS- and application-agnostic and a gray-box approach that exploits OS- and application-level statistics. We implement our techniques in Xen and conduct a detailed evaluation using a mix of CPU, network and memory-intensive applications. Our results show that Sandpiper is able to resolve single server hotspots within 20 seconds and scales well to larger, data center environments. We also show that the gray-box approach can help Sandpiper make more informed decisions, particularly in response to memory pressure.

/pdf/black-box-and-gray-box-strategies-for-virtual-machine-4dlglfch88.pdf

Black-box and gray-box strategies for virtual machine migration

Cloud computing is an emerging commercial infrastructure paradigm that promises to eliminate the need for maintaining expensive computing facilities by companies and institutes alike. Through the use of virtualization and resource time sharing, clouds serve with a single set of physical resources a large user base with different needs. Thus, clouds have the potential to provide to their owners the benefits of an economy of scale and, at the same time, become an alternative for scientists to clusters, grids, and parallel production environments. However, the current commercial clouds have been built to support web and small database workloads, which are very different from typical scientific computing workloads. Moreover, the use of virtualization and resource time sharing may introduce significant performance penalties for the demanding scientific computing workloads. In this work, we analyze the performance of cloud computing services for scientific computing workloads. We quantify the presence in real scientific computing workloads of Many-Task Computing (MTC) users, that is, of users who employ loosely coupled applications comprising many tasks to achieve their scientific goals. Then, we perform an empirical evaluation of the performance of four commercial cloud computing services including Amazon EC2, which is currently the largest commercial cloud. Last, we compare through trace-based simulation the performance characteristics and cost models of clouds and other scientific computing platforms, for general and MTC-based scientific computing workloads. Our results indicate that the current clouds need an order of magnitude in performance improvement to be useful to the scientific community, and show which improvements should be considered first to address this discrepancy between offer and demand.

/pdf/performance-analysis-of-cloud-computing-services-for-many-3e3awq09a7.pdf

Performance Analysis of Cloud Computing Services for Many-Tasks Scientific Computing

Large-scale datacenters (DCs) host tens of thousands of diverse applications each day. However, interference between colocated workloads and the difficulty to match applications to one of the many hardware platforms available can degrade performance, violating the quality of service (QoS) guarantees that many cloud workloads require. While previous work has identified the impact of heterogeneity and interference, existing solutions are computationally intensive, cannot be applied online and do not scale beyond few applications.We present Paragon, an online and scalable DC scheduler that is heterogeneity and interference-aware. Paragon is derived from robust analytical methods and instead of profiling each application in detail, it leverages information the system already has about applications it has previously seen. It uses collaborative filtering techniques to quickly and accurately classify an unknown, incoming workload with respect to heterogeneity and interference in multiple shared resources, by identifying similarities to previously scheduled applications. The classification allows Paragon to greedily schedule applications in a manner that minimizes interference and maximizes server utilization. Paragon scales to tens of thousands of servers with marginal scheduling overheads in terms of time or state.We evaluate Paragon with a wide range of workload scenarios, on both small and large-scale systems, including 1,000 servers on EC2. For a 2,500-workload scenario, Paragon enforces performance guarantees for 91% of applications, while significantly improving utilization. In comparison, heterogeneity-oblivious, interference-oblivious and least-loaded schedulers only provide similar guarantees for 14%, 11% and 3% of workloads. The differences are more striking in oversubscribed scenarios where resource efficiency is more critical.

/pdf/paragon-qos-aware-scheduling-for-heterogeneous-datacenters-1d2sln8y7q.pdf

Paragon: QoS-aware scheduling for heterogeneous datacenters

Virtualization has become commonplace in modern data centers, often referred as "computing clouds". The capability of virtual machine live migration brings benefits such as improved performance, manageability and fault tolerance, while allowing workload movement with a short service downtime. However, service levels of applications are likely to be negatively affected during a live migration. For this reason, a better understanding of its effects on system performance is desirable. In this paper, we evaluate the effects of live migration of virtual machines on the performance of applications running inside Xen VMs. Results show that, in most cases, migration overhead is acceptable but cannot be disregarded, especially in systems where availability and responsiveness are governed by strict Service Level Agreements. Despite that, there is a high potential for live migration applicability in data centers serving modern Internet applications. Our results are based on a workload covering the domain of multi-tier Web 2.0 applications.

Cost of Virtual Machine Live Migration in Clouds: A Performance Evaluation

Virtual Machine (VM) environments (e.g., VMware and Xen) are experiencing a resurgence of interest for diverse uses including server consolidation and shared hosting. An application's performance in a virtual machine environment can differ markedly from its performance in a non-virtualized environment because of interactions with the underlying virtual machine monitor and other virtual machines. However, few tools are currently available to help debug performance problems in virtual machine environments.In this paper, we present Xenoprof, a system-wide statistical profiling toolkit implemented for the Xen virtual machine environment. The toolkit enables coordinated profiling of multiple VMs in a system to obtain the distribution of hardware events such as clock cycles and cache and TLB misses. The toolkit will facilitate a better understanding of performance characteristics of Xen's mechanisms allowing the community to optimize the Xen implementation.We use our toolkit to analyze performance overheads incurred by networking applications running in Xen VMs. We focus on networking applications since virtualizing network I/O devices is relatively expensive. Our experimental results quantify Xen's performance overheads for network I/O device virtualization in uni- and multi-processor systems. With certain Xen configurations, networking workloads in the Xen environment can suffer significant performance degradation. Our results identify the main sources of this overhead which should be the focus of Xen optimization efforts. We also show how our profiling toolkit was used to uncover and resolve performance bugs that we encountered in our experiments which caused unexpected application behavior.

/pdf/diagnosing-performance-overheads-in-the-xen-virtual-machine-zuymytqtxe.pdf

Diagnosing performance overheads in the xen virtual machine environment

Virtual machines (VMs) have recently emerged as the basis for allocating resources in enterprise settings and hosting centers. One benefit of VMs in these environments is the ability to multiplex several operating systems on hardware based on dynamically changing system characteristics. However, such multiplexing must often be done while observing per-VM performance guarantees or service level agreements. Thus, one important requirement in this environment is effective performance isolation among VMs. In this paper, we address performance isolation across virtual machines in Xen [1]. For instance, while Xen can allocate fixed shares of CPU among competing VMs, it does not currently account for work done on behalf of individual VMs in device drivers. Thus, the behavior of one VM can negatively impact resources available to other VMs even if appropriate per-VM resource limits are in place.In this paper, we present the design and evaluation of a set of primitives implemented in Xen to address this issue. First, XenMon accurately measures per-VM resource consumption, including work done on behalf of a particular VM in Xen's driver domains. Next, our SEDF-DC scheduler accounts for aggregate VM resource consumption in allocating CPU. Finally, ShareGuard limits the total amount of resources consumed in privileged and driver domains based on administrator-specified limits. Our performance evaluation indicates that our mechanisms effectively enforce performance isolation for a variety of workloads and configurations.

/pdf/enforcing-performance-isolation-across-virtual-machines-in-1i4127t9im.pdf

Enforcing performance isolation across virtual machines in Xen

Virtual Machine Monitors (VMMs) are gaining popularity in enterprise environments as a software-based solution for building shared hardware infrastructures via virtualization. In this work, using the Xen VMM, we present a light weight monitoring system for measuring the CPU usage of different virtual machines including the CPU overhead in the device driver domain caused by I/O processing on behalf of a particular virtual machine. Our performance study attempts to quantify and analyze this overhead for a set of I/O intensive workloads.

/pdf/measuring-cpu-overhead-for-i-o-processing-in-the-xen-virtual-3r5zxuwrfx.pdf

Measuring CPU overhead for I/O processing in the Xen virtual machine monitor

We present a high level overview of a virtual machine placement system in which an autonomic controller dynamically manages the mapping of virtual machines onto physical hosts in accordance with policies specified by the user. By closely monitoring virtual machine activity and employing advanced policies for dynamic workload placement, such an autonomic solution can achieve substantial cost savings from better utilization of computing resources and less frequent overload situations.

http://www.hpl.hp.com/techreports/2007/HPL-2007-189.pdf

Autonomic Virtual Machine Placement in the Data Center

The goal of this short paper is twofold: 1) it briefly describes a new performance monitoring tool, XenMon, that we built for the Xen-based virtual environment, and 2) it presents a performance case study that demonstrates and explains how different metrics reported by XenMon can be used in gaining insight into an application’s performance and its resource usage/requirements, especially in the case of I/O intensive applications.

http://www.hpl.hp.com/techreports/2005/HPL-2005-187.pdf

XenMon: QoS Monitoring and Performance Profiling Tool

A combined-hardware-and-software secure-platform interface to which operating systems and customized control programs interface within a computer system. The combined-hardware-and-software secure-platform interface employs a hardware platform that provides at least four privilege levels, non-privileged instructions, non-privileged registers, privileged instructions, privileged registers, and firmware interfaces. The combined-hardware-and-software secure-platform interface conceals all privileged instructions, privileged registers, and firmware interfaces and privileged registers from direct access by operating systems and custom control programs, providing to the operating systems and custom control programs the non-privileged instructions and non-privileged registers provided by the hardware platform as well as a set of callable software services. The callable services provide a set of secure-platform management services for operational control of hardware resources that neither exposes privileged instructions, privileged registers, nor firmware interfaces of the hardware nor simulates privileged instructions and privileged registers. The callable services also provide a set of security-management services that employ internally generated secret data, each compartmentalized security-management service managing internal secret data without exposing the internal secret data to computational entities other than the security-management service itself.

Robert D. Gardner

Papers

Enforcing performance isolation across virtual machines in Xen

Measuring CPU overhead for I/O processing in the Xen virtual machine monitor

Autonomic Virtual Machine Placement in the Data Center

XenMon: QoS Monitoring and Performance Profiling Tool

Secure machine platform that interfaces to operating systems and customized control programs