Zhikui Wang

Adaptive control of virtualized resources in utility computing environments

Data centers are often under-utilized due to over-provisioning as well as time-varying resource demands of typical enterprise applications. One approach to increase resource utilization is to consolidate applications in a shared infrastructure using virtualization. Meeting application-level quality of service (QoS) goals becomes a challenge in a consolidated environment as application resource needs differ. Furthermore, for multi-tier applications, the amount of resources needed to achieve their QoS goals might be different at each tier and may also depend on availability of resources in other tiers. In this paper, we develop an adaptive resource control system that dynamically adjusts the resource shares to individual tiers in order to meet application-level QoS goals while achieving high resource utilization in the data center. Our control system is developed using classical control theory, and we used a black-box system modeling approach to overcome the absence of first principle models for complex enterprise applications and systems. To evaluate our controllers, we built a testbed simulating a virtual data center using Xen virtual machines. We experimented with two multi-tier applications in this virtual data center: a two-tier implementation of RUBiS, an online auction site, and a two-tier Java implementation of TPC-W. Our results indicate that the proposed control system is able to maintain high resource utilization and meets QoS goals in spite of varying resource demands from the applications.

No "power" struggles: coordinated multi-level power management for the data center

Power delivery, electricity consumption, and heat management are becoming key challenges in data center environments. Several past solutions have individually evaluated different techniques to address separate aspects of this problem, in hardware and software, and at local and global levels. Unfortunately, there has been no corresponding work on coordinating all these solutions. In the absence of such coordination, these solutions are likely to interfere with one another, in unpredictable (and potentially dangerous) ways. This paper seeks to address this problem. We make two key contributions. First, we propose and validate a power management solution that coordinates different individual approaches. Using simulations based on 180 server traces from nine different real-world enterprises, we demonstrate the correctness, stability, and efficiency advantages of our solution. Second, using our unified architecture as the base, we perform a detailed quantitative sensitivity analysis and draw conclusions about the impact of different architectures, implementations, workloads, and system design choices.

Automated control of multiple virtualized resources

Virtualized data centers enable sharing of resources among hosted applications. However, it is difficult to satisfy service-level objectives(SLOs) of applications on shared infrastructure, as application workloads and resource consumption patterns change over time. In this paper, we present AutoControl, a resource control system that automatically adapts to dynamic workload changes to achieve application SLOs. AutoControl is a combination of an online model estimator and a novel multi-input, multi-output (MIMO) resource controller. The model estimator captures the complex relationship between application performance and resource allocations, while the MIMO controller allocates the right amount of multiple virtualized resources to achieve application SLOs. Our experimental evaluation with RUBiS and TPC-W benchmarks along with production-trace-driven workloads indicates that AutoControl can detect and mitigate CPU and disk I/O bottlenecks that occur over time and across multiple nodes by allocating each resource accordingly. We also show that AutoControl can be used to provide service differentiation according to the application priorities during resource contention.

Renewable and cooling aware workload management for sustainable data centers

Recently, the demand for data center computing has surged, increasing the total energy footprint of data centers worldwide. Data centers typically comprise three subsystems: IT equipment provides services to customers; power infrastructure supports the IT and cooling equipment; and the cooling infrastructure removes heat generated by these subsystems. This work presents a novel approach to model the energy flows in a data center and optimize its operation. Traditionally, supply-side constraints such as energy or cooling availability were treated independently from IT workload management. This work reduces electricity cost and environmental impact using a holistic approach that integrates renewable supply, dynamic pricing, and cooling supply including chiller and outside air cooling, with IT workload planning to improve the overall sustainability of data center operations. Specifically, we first predict renewable energy as well as IT demand. Then we use these predictions to generate an IT workload management plan that schedules IT workload and allocates IT resources within a data center according to time varying power supply and cooling efficiency. We have implemented and evaluated our approach using traces from real data centers and production systems. The results demonstrate that our approach can reduce both the recurring power costs and the use of non-renewable energy by as much as 60% compared to existing techniques, while still meeting the Service Level Agreements.

Utility-driven workload management using nested control design

Virtualization and consolidation of IT resources have created a need for more effective workload management tools, one that dynamically controls resource allocation to a hosted application to achieve quality of service (QoS) goals. These goals can in turn be driven by the utility of the service, typically based on the applications service level agreement (SLA) as well as the cost of resources allocated. In this paper, we build on our earlier work on dynamic CPU allocation to applications on shared servers, and present a feedback control system consisting of two nested integral control loops for managing the QoS metric of the application along with the utilization of the allocated CPU resource. The control system was implemented on a lab testbed running an Apache Web server and using the 90th percentile of the response times as the QoS metric. Experiments using a synthetic workload based on an industry benchmark validated two important features of the nested control design. First, compared to a single loop for controlling response time only, the nested design is less sensitive to the bimodal behavior of the system resulting in more robust performance. Second, compared to a single loop for controlling CPU utilization only, the new design provides a framework for dealing with the tradeoff between better QoS and lower cost of resources, therefore resulting in better overall utility of the service

Utilization and SLO-Based control for dynamic sizing of resource partitions

This paper deals with a shared server environment where the server is divided into a number of resource partitions and used to host multiple applications at the same time. In a case study where the HP-UX Process Resource Manager is taken as the server partitioning technology, we investigate the technical challenges in performing automated sizing of a resource partition using a feedback control approach, where the CPU entitlement for the partition is dynamically tuned to regulate output metrics such as the CPU utilization or SLO-based application performance metric. We identify the nonlinear and bimodal properties of the models across different operating regions, and discuss their implications for the design of the control loops. To deal with these challenges, we then propose two adaptive controllers for tracking the target utilization and target response time respectively. We evaluate the performance of the closed-loop systems while varying certain operating conditions. We demonstrate that better performance and robustness can be achieved with these controllers compared with other controllers or our prior solution.

Memory overbooking and dynamic control of Xen virtual machines in consolidated environments

The newly emergent cloud computing environments host hundreds to thousands of services on a shared resource pool. The sharing is enhanced by virtualization technologies allowing multiple services to run in different virtual machines (VMs) on a single physical node. Resource overbooking allows more services with time-varying demands to be consolidated reducing operational costs. In the past, researchers have studied dynamic control mechanisms for allocating CPU to virtual machines, when CPU is overbooked with respect to the sum of the peak demands from all the VMs. However, runtime re-allocation of memory among multiple VMs has not been widely studied, except on VMware platforms. In this paper, we present a case study where feedback control is used for dynamic memory allocation to Xen virtual machines in a consolidated environment. We illustrate how memory behaves differently from CPU in terms of its relationship to application-level performance, such as response times. We have built a prototype of a joint resource control system for allocating both CPU and memory resources to co-located VMs in real time. Experimental results show that our solution allows all the hosted applications to achieve the desired performance in spite of their time-varying CPU and memory demands, whereas a solution without memory control incurs significant service level violations.

1000 islands: an integrated approach to resource management for virtualized data centers

Recent advances in hardware and software virtualization offer unprecedented management capabilities for the mapping of virtual resources to physical resources. It is highly desirable to further create a “service hosting abstraction” that allows application owners to focus on service level objectives (SLOs) for their applications. This calls for a resource management solution that achieves the SLOs for many applications in response to changing data center conditions and hides the complexity from both application owners and data center operators. In this paper, we describe an automated capacity and workload management system that integrates multiple resource controllers at three different scopes and time scales. Simulation and experimental results confirm that such an integrated solution ensures efficient and effective use of data center resources while reducing service level violations for high priority applications.

Integrated management of application performance, power and cooling in data centers

Data centers contain IT, power and cooling infrastructures, each of which is typically managed independently. In this paper, we propose a holistic approach that couples the management of IT, power and cooling infrastructures to improve the efficiency of data center operations. Our approach considers application performance management, dynamic workload migration/consolidation, and power and cooling control to “right-provision” computing, power and cooling resources for a given workload. We have implemented a prototype of this for virtualized environments and conducted experiments in a production data center. Our experimental results demonstrate that the integrated solution is practical and can reduce energy consumption of servers by 35% and cooling by 15%, without degrading application performance.

Probabilistic performance modeling of virtualized resource allocation

Virtualization technologies enable organizations to dynamically flex their IT resources based on workload fluctuations and changing business needs. However, only through a formal understanding of the relationship between application performance and virtualized resource allocation can over-provisioning or over-loading of physical IT resources be avoided. In this paper, we examine the probabilistic relationships between virtualized CPU allocation, CPU contention, and application response time, to enable autonomic controllers to satisfy service level objectives (SLOs) while more effectively utilizing IT resources. We show that with only minimal knowledge of application and system behaviors, our methodology can model the probability distribution of response time with a mean absolute error of less than 6% when compared with the measured response time distribution. We then demonstrate the usefulness of a probabilistic approach with case studies. We apply basic laws of probability to our model to investigate whether and how CPU allocation and contention affect application response time, correcting for their effects on CPU utilization. We find mean absolute differences of 8-10% between the modeled response time distributions of certain allocation states, and a similar difference when we add CPU contention. This methodology is general, and should also be applicable to non-CPU virtualized resources and other performance modeling problems.