Ananth I. Sundararaj

Dynamic topology adaptation of virtual networks of virtual machines

Virtual machine grid computing greatly simplifies the use of widespread computing resources by lowering the level of abstraction, benefiting both resource providers and users. For the user, the Virtuoso middleware that we are developing closely emulates the existing process of buying, configuring and using machines. VNET, a component of Virtuoso, is a simple and efficient layer two virtual network tool that makes these virtual machines appear to be connected to the home network of the user, simplifying network management. Overlays like VNET have great potential as the mechanism for adaptation. Here, we describe our second generation VNET implementation, which includes support for arbitrary topologies and routing, application topology inference, and adaptive control of the overlay. We demonstrate that the performance of unmodified applications, in particular bulk synchronous parallel applications running inside the virtual machines and serviced by VNET, can be significantly (up to a factor of two) enhanced by adapting the VNET topology and forwarding rules on the fly based on intelligent application traffic inference methods. The adaptation scheme requires no knowledge or participation from the user or application developer.

Increasing application performance in virtual environments through run-time inference and adaptation

Virtual machine distributed computing greatly simplifies the use of widespread computing resources by lowering the level of abstraction, benefiting both resource providers and users. Towards that end our Virtuoso middleware closely emulates the existing process of buying, configuring and using physical machines. Virtuosos VNET component is a simple and efficient layer two virtual network tool that makes these virtual machines (VMs) appear to be physically connected to the home network of the user while simultaneously supporting arbitrary topologies and routing among them. Virtuosos VTTIF component continually infers the communication behavior of the application running in a collection of VMs. The combination of overlays like VNET and inference frameworks like VTTIF has great potential to increase the performance, with no user or developer involvement, of existing, unmodified applications by adapting their virtual environments to the underlying computing infrastructure to best suit the applications. We show here how to use the continually inferred application topology and traffic to dynamically control three mechanisms of adaptation, VM migration, overlay topology, and forwarding to significantly increase the performance of two classes of applications, bulk synchronous parallel applications and transactional Web e-commerce applications.

Automatic dynamic run-time optical network reservations

Optical networking may dramatically change high performance distributed computing. One reason is that optical networks can support provisioning dynamically configurable lightpaths, a form of circuit switching, through reservations. However, to use it (and all other network reservation mechanisms), the user or developer must modify the application. We present a system, VRESERVE, that automatically and dynamically creates network reservation requests based on the inferred network demands of running distributed and/or parallel applications with no modification to the application or operating system, and no input from the user or developer. Our execution model is a collection of virtual machines interconnected by an overlay network. The overlay network infers application demands, providing a dynamic run-time assessment of the applications topology and traffic load matrix. We then reserve lightpaths corresponding to the topology and use the overlay to forward virtual network traffic over them. We evaluate our system on the OMNInet network.

Free network measurement for adaptive virtualized distributed computing

An execution environment consisting of virtual machines (VMs) interconnected with a virtual overlay network can use the naturally occurring traffic of an existing, unmodified application running in the VMs to measure the underlying physical network. Based on these characterizations, and characterizations of the applications own communication topology, the execution environment can optimize the execution of the application using application-independent means such as VM migration and overlay topology changes. In this paper, we demonstrate the feasibility of such free automatic network measurement by fusing the Wren passive monitoring and analysis system with Virtuosos virtual networking system. We explain how Wren has been extended to support online analysis, and we explain how Virtuosos adaptation algorithms have been enhanced to use Wrens physical network level information to choose VM-to-host mappings, overlay topology, and forwarding rules

An optimization problem in adaptive virtual environments

A virtual execution environment consisting of virtual machines (VMs) interconnected with virtual networks provides opportunities to dynamically optimize, at run-time, the performance of existing, unmodified distributed applications without any user or programmer intervention. Along with resource monitoring and inference and application-independent adaptation mechanisms, efficient adaptation algorithms are key to the success of such an effort. In previous work we have described our measurement and inference framework, explained our adaptation mechanisms, and proposed simple heuristics as adaptation algorithms. Though we were successful in improving performance as compared to the case with no adaptation, none of our algorithms were characterized by theoretically proven bounds. In this paper, we formalize the adaptation problem, show that it is NP-hard and propose research directions for coming up with an efficient solution.

Hardness of Approximation and Greedy Algorithms for the Adaptation Problem in Virtual Environments

Over the past decade, wide-area distributed computing has emerged as a powerful computing paradigm. Virtual machines greatly simplify wide-area distributed computing by lowering the abstraction to benefit both resource users and providers. A virtual execution environment consisting of virtual machines (VMs) interconnected with virtual networks provides opportunities to dynamically optimize, at run-time, the performance of existing, unmodified distributed applications without any user or programmer intervention. We have formalized the adaptation problem in virtual execution environments and shown that it is NP-hard to both, solve and approximate within a factor of m1/2-δfor any δ > 0, where m is the number of edges in the virtual overlay graph. We also designed and evaluated greedy adaptation algorithms and found them to work well in practice.

Time-sharing parallel applications through performance-targeted feedback-controlled real-time scheduling

Most parallel machines, such as clusters, are space-shared in order to isolate batch parallel applications from each other and optimize their performance. However, this leads to low utilization or potentially long waiting times. We propose a self-adaptive approach to time-sharing such machines that provides isolation and allows the execution rate of an application to be tightly controlled by the administrator. Our approach combines a periodic real-time scheduler on each node with a global feedback-based control system that governs the local schedulers. We have developed an online system that implements our approach. The system takes as input a target execution rate for each application, and automatically and continuously adjusts the applications’ real-time schedules to achieve those rates with proportional CPU utilization. Target rates can be dynamically adjusted. Applications are performance-isolated from each other and from other work that is not using our system. We present an extensive evaluation that shows that the system remains stable with low response times, and that our focus on CPU isolation and control does not come at the significant expense of network I/O, disk I/O, or memory isolation.