Organizational Unit:

Center for Experimental Research in Computer Systems

Permanent Link

https://hdl.handle.net/1853/70632

Full item page

Publication Search Results

Now showing 1 - 10 of 25

A Virtualized Quality of Service Packet Scheduling Accelerator

(Georgia Institute of Technology, 2008-05-21) Chuang, Kangtao Kendall ; Yalamanchili, Sudhakar ; Gavrilovska, Ada ; Schwan, Karsten

This paper introduces the virtualization of a Quality of Service Packet Scheduler. Virtualization in terms of resource sharing among multiple processes of virtual packet schedulers implementing the DWCS algorithm on an FPGA is implemented in the ShareStreams- V architecture. This work builds on the previous work in ShareStreams, which implemented the Dynamic Window-Constrained Scheduler algorithm. This implementation is parametric, permitting tradeoffs between packet decision latency, decision throughput, and the number of virtual packet schedulers supported. ShareStreams-V is able to schedule minimal size packets faster than one decision per 51.2 ns for up to 64 streams, the throughput required for 10Gbps Ethernet.
Creating Efficient Execution Platforms with V(irtualized) Services

(Georgia Institute of Technology, 2008) Seshasayee, Balasubramanian ; Schwan, Karsten

Virtualization is creating new opportunities for innovative uses of middleware technologies. In this paper, we present such opportunities for mobile or pervasive environments, where virtualization efforts face challenges that include the diversity of platforms and devices present in these domains, dynamic device and resource behaviors, and the efficient use and sharing of such resources. Specifically, we leverage middleware to significantly extend virtualization technologies in terms of their efficient support for resource sharing in the presence of diversity, dynamics, and mobility. The outcome is what we term the Virtualized Services (VServices) approach to representing and using devices and resources. VServices extend the virtual device interfaces used in existing virtualization infrastructures to go beyond sharing physical devices among multiple virtual machines, to also sharing logical entities that internally use middleware to provide new services to end user applications. By using and sharing the higher level VServices abstractions instead of physical devices, opportunities are created (i) to optimize the implementations of certain services without requiring changes to VM implementations, (ii) to enhance device functionalities by combining software service implementations with the physical devices being used, again without changing the basic nature of VM-device interactions, and (iii) to emulate devices whose physical realizations may be remote or non-existent. Experimental evaluations conducted on an implementation of these concepts in the Xen virtualization infrastructure exhibit up to 50% improvements in latency as well as improved performance scalability, when compared to current Xen implementations of such devices. This paper also describes practical realization of device enhancements using VServices.
VMedia: Enhanced Multimedia Services in Virtualized Systems

(Georgia Institute of Technology, 2007) Raj, Himanshu ; Seshasayee, Balasubramanian ; Schwan, Karsten

This paper presents the VMedia multimedia virtualization framework, for sharing media devices among multiple virtual machines (VMs). The framework provides logical media devices to virtual machines. These devices are exported via a well defined, higher level, multimedia access interface to the applications and operating system running in a virtual machine. By using semantically meaningful information, rather than low-level raw data, within the VMedia framework, efficient virtualization solutions can be created for physical devices shared by multiple virtual machines. Experimental results demonstrate that the base cost of virtual device access via VMedia is small compared to native physical device access, and in addition, that these costs scale well with an increasing number of guest VMs. Here, VMedia’s MediaGraph abstraction is a key contributor, since it also allows the framework to support dynamic restructuring, in order to adapt device accesses to changing requirements. Finally, VMedia permits platforms to offer new and enhanced logical device functionality at lower costs than those achievable with alternative solutions.
Towards IQ-Appliances: Quality-awareness in Information Virtualization

(Georgia Institute of Technology, 2007) Niranjan Mysore, Radhika ; Gavrilovska, Ada ; Schwan, Karsten ; Tembey, Priyanka

Our research addresses two important problems that arise in modern large-scale distributed systems: (1) the necessity to virtualize their data flows by applying actions such as filtering, format translation, coalescing or splitting, etc., and (2) the desire to separate such actions from enterprise applications' business logic, to make it easier for future service-oriented codes to interoperate in diverse and dynamic environments. This paper considers the runtimes of the `information appliances' used for these purposes, particularly with respect to their ability to provide diverse levels of Quality of Service (QoS) in lieu of dynamic application behaviors and the consequent changes in the resource needs of their data flows. Our specific contribution is the enrichment of these runtimes with methods for QoS-awareness, thereby giving them the ability to deliver desired levels of QoS even under sudden requirement changes -- IQ-appliances. For experimental evaluation, we enrich a prototype implementation of an IQ-appliance, based on the Intel IXP network processor with the additional functionality needed to guarantee QoS constraints for diverse data streams. Measurements demonstrate the feasibility and utility of the approach.
M-Channels and M-Brokers: New Abstractions for Co-ordinated Management in Virtualized Systems

(Georgia Institute of Technology, 2007) Kumar, Sanjay ; Nathuji, Ripal ; Schwan, Karsten ; Talwar, Vanish ; Ranganathan, Partha

Management and automation are important issues in enterprise environments, often consuming the largest fraction of the overall IT budget. A key challenge here is co-ordination across multiple management solutions deployed in different management domains, including across hardware and software, across different levels of abstraction, and across different hosts. This paper makes three contributions to addressing this problem. First, we propose a novel management co-ordination architecture for virtualized environments. Our architecture includes two powerful abstractions – m-channels, which provide mechanisms for communication between the hardware, virtual machines, and applications and m-brokers, which allow high-level policy co-ordination across different management agents. Second, we discuss Dom-M, an instantiation of our architecture in the context of the Xen hypervisor, and identify tradeoffs between different implementations of such management domains. Finally, to demonstrate the effectiveness of our approach, we implement and evaluate different cross-level solutions for power management using our abstractions, and also discuss qualitatively other applications including storage backup, inventory management, and trust management.
Energy-aware Mobile Service Overlays: Cooperative Dynamic Power Management in Distributed Mobile Systems

(Georgia Institute of Technology, 2007) Seshasayee, Balasubramanian ; Nathuji, Ripal ; Schwan, Karsten

With their increasingly powerful computational resources and high-speed wireless communications, future mobile systems will have the ability to run sophisticated applications on collections of cooperative end devices. Mobility, however, requires dynamic management of these platforms' distributed resources, and such management can also be used to meet application quality requirements and prolong application lifetimes, the latter by best using available energy resources. This paper presents energy-aware Mobile Service Overlays (MSOs), a set of mechanisms and associated policies for running mobile applications across multiple, cooperating machines while actively performing power management to extend system usability lifetimes. MSO policies manage energy consumption by (i) allocating application components to available nodes based upon their current energy capacities and resource availabilities, (ii) monitoring for, and responding to changes in energy and resource characteristics, and (iii) dynamically exploiting energy-performance tradeoffs in overprovisioned situations. Coupled with mobility, such cooperation enables multiple mobile platforms to bring their joint resources to bear on complex application tasks, providing significant benefits to application lifetimes and performance. This paper evaluates MSOs on a MANET computing testbed indicate an extension in system lifetime of upto 10% for an example application.
Netbus: A Transparent Mechanism for Remote Device Access in Virtualized Systems

(Georgia Institute of Technology, 2007) Kumar, Sanjay ; Agarwala, Sandip ; Schwan, Karsten

Efficient and seamless access to local as well as remote devices is a desirable property in multiple settings, including blade-servers, datacenters, enterprises, and even in home-based, personal computing environments. New virtualization technologies developed for PC and server platforms are now making it possible to implement remote device access at a level of abstraction transparent to operating systems and their device drivers. This paper presents a new mechanism for transparent device remoting, resulting in a hypervisor-level abstraction termed Netbus. The Netbus software solution provides both (1) efficient and reliable access to networked devices, and (2) remote access to devices not directly attached to networks, an example being a disk locally present on a bladeserver node. Netbus-based device remoting also supports virtual device migration, device hotswapping and efficient device sharing. A Xen-based prototype implementation of Netbus demonstrates transparent device remoting for block and for USB devices, for both bulk and isochronous USB access methods. Within the same administrative domain, seamless access to these devices is maintained during VM migration and during device hotswapping. Experimental evaluations with microbenchmarks and with representative server applications exhibit comparable performance for Netbus-based remote vs. local devices.
E2EProf: Automated End-to-End Performance Management for Enterprise Systems

(Georgia Institute of Technology, 2007) Agarwala, Sandip ; Alegre, Fernando ; Schwan, Karsten ; Mehalingham, Jegannathan

Distributed systems are becoming increasingly complex, caused by the prevalent use of web services, multi-tier architectures, and grid computing, where dynamic sets of components interact with each other across distributed and heterogeneous computing infrastructures. For these applications to be able to predictably and efficiently deliver services to end users, it is therefore, critical to understand and control their runtime behavior. In a datacenter environment, for instance, understanding the end-to-end dynamic behavior of certain IT subsystems, from the time requests are made to when responses are generated and finally, received, is a key prerequisite for improving application response, to provide required levels of performance, or to meet service level agreements (SLAs). The 'E2EProf' toolkit enables the efficient and non-intrusive capture and analysis of end-to-end program behavior for complex enterprise applications. E2EProf permits an enterprise to recognize and analyze performance problems when they occur -- online, to take corrective actions as soon as possible and wherever necessary along the paths currently taken by user requests -- end-to-end, and to do so without the need to instrument applications -- non-intrusively. Online analysis exploits a novel signal analysis algorithm, termed 'pathmap', which dynamically detects the causal paths taken by client requests through application and backend servers and annotates these paths with end-to-end latencies and with the contributions to these latencies from different path components. Thus, with pathmap, it is possible to dynamically identify the bottlenecks present in selected servers or services and to detect the abnormal or unusual performance behaviors indicative of potential problems or overloads. Pathmap and the E2EProf toolkit successfully detect causal request paths and associated performance bottlenecks in the RUBiS ebay-like multi-tier web application and in one of the datacenter of our industry partner, Delta Air Lines.
O2S2: Enhanced Object-based Virtualized Storage

(Georgia Institute of Technology, 2007) Raj, Himanshu ; Schwan, Karsten

Object based storage devices (OSDs) elevate the level of abstraction presented to clients, thereby permitting them to offer methods for managing, sharing, and securing information that go beyond those offered by block-based stores. The Object-Oriented Storage System (O2S2) architecture presented and evaluated in this paper implements a virtualization service to provide object-based storage in a virtualized environment. This service provides a virtual object-based storage device (vOSD) to virtual machines. The use of vOSDs permits the service provider, i.e., the vOSD storage domain, to offer to guest virtual machines new methods for resource management and consolidation, without requiring the purchase of physical storage devices that faithfully implement OSD functionality. Methods demonstrated in this paper include improved support for access control and for heterogeneity of storage devices. Advantages derived from such methods also include reduced complexity for end clients, i.e., guest VMs. A prototype PVFS-based O2S2 implementation demonstrates that its enhanced services can be provided at low cost, enabled in part by the effcient utilization of otherwise idle domain resources.
Self-Virtualized I/O: High Performance, Scalable I/O Virtualization in Multi-core Systems

(Georgia Institute of Technology, 2006) Raj, Himanshu ; Ganev, Ivan Borissov ; Schwan, Karsten ; Xenidis, Jimi

Virtualizing I/O subsystems and peripheral devices is an integral part of system virtualization. This paper advocates the notion of self-virtualized I/O (S-VIO). Specifically, it proposes a hypervisor-level abstraction that permits guest virtual machines to efficiently exploit the multi-core nature of future machines when interacting with virtualized I/O. The concrete instance of S-VIO developed and evaluated herein (1) provides virtual interfaces to an underlying physical device, the network interface, and (2) manages the way in which the device's physical resources are used by guest operating systems. The performance of this instance differs markedly depending on design choices that include (a) how the S-VIO abstraction is mapped to the underlying host- vs. device-resident resources, (b) the manner and extent to which it interacts with the HV, and (c) its ability to flexibly leverage the multi-core nature of modern computing platforms. A device-centric S-VIO realization yields a self-virtualized network device (SV-NIC) that provides high performance network access to guest virtual machines. Specific performance results show that for high-end network hardware using an IXP2400-based board, a virtual network interface (VIF) from the device-centric S-VIO realization provides 77% more throughput and 53% less latency compared to the VIF from a host-centric S-VIO realization. For 8 VIFs, the aggregate throughput (latency) for device-centric version is 103% more (39% less) compared to the host-centric version. The aggregate throughput and latency of the VIFs scales with guest VMs, ultimately limited by the amount of physical computing resources available on the host platform and device, such as number of cores. The paper also discusses architectural considerations for implementing self-virtualized devices in future multi-core systems.