Ripal Nathuji

Q-clouds: managing performance interference effects for QoS-aware clouds

Cloud computing offers users the ability to access large pools of computational and storage resources on demand. Multiple commercial clouds already allow businesses to replace, or supplement, privately owned IT assets, alleviating them from the burden of managing and maintaining these facilities. However, there are issues that must be addressed before this vision of utility computing can be fully realized. In existing systems, customers are charged based upon the amount of resources used or reserved, but no guarantees are made regarding the application level performance or quality-of-service (QoS) that the given resources will provide. As cloud providers continue to utilize virtualization technologies in their systems, this can become problematic. In particular, the consolidation of multiple customer applications onto multicore servers introduces performance interference between collocated workloads, significantly impacting application QoS. To address this challenge, we advocate that the cloud should transparently provision additional resources as necessary to achieve the performance that customers would have realized if they were running in isolation. Accordingly, we have developed Q-Clouds, a QoS-aware control framework that tunes resource allocations to mitigate performance interference effects. Q-Clouds uses online feedback to build a multi-input multi-output (MIMO) model that captures performance interference interactions, and uses it to perform closed loop resource management. In addition, we utilize this functionality to allow applications to specify multiple levels of QoS as application Q-states. For such applications, Q-Clouds dynamically provisions underutilized resources to enable elevated QoS levels, thereby improving system efficiency. Experimental evaluations of our solution using benchmark applications illustrate the benefits: performance interference is mitigated completely when feasible, and system utilization is improved by up to 35% using Q-states.

Resource management for isolation enhanced cloud services

The cloud infrastructure provider (CIP) in a cloud computing platform must provide security and isolation guarantees to a service provider (SP), who builds the service(s) for such a platform. We identify last level cache (LLC) sharing as one of the impediments to finer grain isolation required by a service, and advocate two resource management approaches to provide performance and security isolation in the shared cloud infrastructure - cache hierarchy aware core assignment and page coloring based cache partitioning. Experimental results demonstrate that these approaches are effective in isolating cache interference impacts a VM may have on another VM. We also incorporate these approaches in the resource management (RM) framework of our example cloud infrastructure, which enables the deployment of VMs with isolation enhanced SLAs.

VPM tokens: virtual machine-aware power budgeting in datacenters

Power consumption and cooling overheads are becoming increasingly significant for enterprise datacenters, affecting overall costs and the ability to extend resource capacities. To help mitigate these issues, active power management technologies are being deployed aggressively, including power budgeting, which enables improved power provisioning and can address critical periods when power delivery or cooling capabilities are temporarily reduced. Given the use of virtualization to encapsulate application components into virtual machines (VMs), however, such power management capabilities must address the interplay between budgeting physical resources and the performance of the virtual machines used to run these applications. This paper proposes a set of management components and abstractions for use by software power budgeting policies. The key idea is to manage power from a VM-centric point of view, where the goal is to be aware of global utility tradeoffs between different virtual machines (and their applications) when maintaining power constraints for the physical hardware on which they run. Our approach to VM-aware power budgeting uses multiple distributed managers integrated into the VirtualPower Management (VPM) framework whose actions are coordinated via a new abstraction, termed VPM tokens. An implementation with the Xen hypervisor illustrates technical benefits of VPM tokens that include up to 43% improvements in global utility, highlighting the ability to dynamically improve cluster performance while still meeting power budgets. We also demonstrate how VirtualPower based budgeting technologies can be leveraged to improve datacenter efficiency in the context of cooling infrastructure management.

Symbiotic Scheduling for Shared Caches in Multi-core Systems Using Memory Footprint Signature

As the trend of more cores sharing common resources on a single die and more systems crammed into enterprise computing space continue, optimizing the economies of scale for a given compute capacity is becoming more critical. One major challenge in performance scalability is the growing L2 cache contention caused by multiple contexts running on a multi-core processor either natively or under a virtual machine environment. Currently, an OS, at best, relies on history based affinity information to dispatch a process or thread onto a particular processor core. Unfortunately, this simple method can easily lead to destructive performance effect due to conflicts in common resources, thereby slowing down all processes. To ameliorate the allocation/management policy of a shared cache on a multi-core, in this paper, we propose Bloom filter signatures, a low-complexity architectural support to allow an OS or a Virtual Machine Monitor to infer cache footprint characteristics and interference of applications, and then perform job scheduling based on symbiosis. Our scheme integrates hardware-level counting Bloom filters in caches to efficiently summarize cache usage behavior on a per-core, per-process or per-VM basis. We then proposed and studied three resource allocation algorithms to determine the optimal process-to-core mapping to minimize interference in the L2. We executed applications using allocation generated by our new process to-core mapping algorithms on an Intel Core 2 Duo machine and showed an averaged 22% (up to 54%) improvement when applications run natively, and an averaged 9.5% improvement (up to 26%)when running inside VMs.

An analysis of power reduction in datacenters using heterogeneous chip multiprocessors

Power and design constraints have forced the semiconductor industry to look at alternate solutions like heterogeneous chip multiprocessors to continue application performance scaling and improve energy efficiency of multicore processors. In this paper, we present an opportunity analysis of heterogeneous chip multiprocessors in the context of datacenter environments where applications often have latency SLAs. Specifically, we define three use cases of heterogeneous processors for datacenter applications and adopt an analytical approach to quantify relative energy savings of using heterogeneous processors over area-equivalent homogeneous configurations. Based upon our findings, we discuss the practical merits of heterogeneous chip multiprocessors in datacenters, including the issues that must be addressed in order to realize the theoretical benefits.

High-performance, power-aware computing - HPPAC

High-performance computing is and has always been performance-oriented. However, a consequence of the push towards maximum performance is increased energy consumption, especially in datacenters and supercomputing centers. Moreover, as peak performance is rarely attained, some of this energy consumption results in little or no performance gain. In addition, large energy consumption costs datacenters and supercomputing centers a significant amount of money and wastes natural resources.