Kranthimanoj Nagothu

Prediction of cloud data center networks loads using stochastic and neural models

The increasing demand for cloud computing resources has led to a commensurate increase in the operating power consumption of the systems that comprise the cloud. In this paper, we introduce a novel framework combining load demand prediction and stochastic state transition models. We claim that our model will lead to optimal cloud resource allocation by minimizing energy consumed while maintaining required performance levels. We characterize the ability of neural network and auto-regressive linear prediction algorithms to forecast loads in cloud data center applications. In this paper, the performance of our models against two sets of data at multiple look-ahead times is also presented.

Persistent Net-AMI for Microgrid Infrastructure Using Cognitive Radio on Cloud Data Centers

We address the potential for a truly universal set of integrated wireless communication services, energy management, and control services for a next-generation of National Institute of Standards and Technology microgrid standards. Our approach uses cloud computing data center as the central communication and optimization infrastructure supporting a cognitive radio network of AMI meters which we label netbook advance metering infrastructure (Net-AMI). The Net-AMI is a novel low cost infrastructure of AMI meters that operate akin to netbooks with wireless transceiver that access to cloud data center energy services, cognitive radio services, and wireless communication services. Access occurs via cognitive radios channels. We claim that this solution solves the important problem in smart grid systems of how to develop an extensible, persistent, smart grid information network with a lifespan equivalent to that of most power systems (20-30 years). By persistence, we imply always operable, entirely software upgradeable, and independent of cellular networks. Our system is extensible and can easily handle thousands of variations in power systems, communication protocols, control, and energy optimization protocols. We formulate necessary link analysis and optimum scheduling of downlink and uplink Net-AMI packets in a multiuser cognitive radio environment.

A Beamforming Approach to Smart Grid Systems Based on Cloud Cognitive Radio

In this paper, we desire to use cognitive radio (CR) channels for communication among a wireless network of smart meters. However, self-interference critically limits the performance of CR systems. This is due to the coexistence of many unplanned systems simultaneously accessing the same signaling bands in an uncoordinated manner. To solve this problem, we show a beamforming approach that effectively mitigates the self-interference effects of the smart meter channel. The beamforming approach is based on minimum mean squared error (MMSE) method in smart meter systems. The MMSE beamformer usually requires accurate channel estimates and noise-plus-interference power estimates for effective mitigation of self-interference in CR systems. In this paper, we propose novel channel estimation and noise-plus-interference power estimation methodologies that efficiently exploit the preamble feature of the IEEE802.22 wireless regional area network (WRAN). Our framework is premised upon the utilization of a cloud computing smart grid infrastructure that hosts the IEEE 802.22 WRAN CR standard. The simulation results for a smart grid system with the MMSE beamformer illustrate significant improvements in system capacity and BER.

Multi-domain robotic swarm communication system

As swarm of robots from different domains works together in a system of systems, the need arises for inter-swarm communication. This paper presents a viable solution for robotic swarm communication and navigation for different autonomous applications. Communication is achieved through ZigBee radio modems and an expandable protocol to accommodate different types of data. This proposed communication system also allows dynamic swarm expansion, where a new member can be added to the swarm family. It is a complementary approach for task coordination and navigation. Navigation is an important issue to accomplish the coordination of tasks in a swarm of robots. Different environmental issues, related to navigation, have been discussed and are presented through simulation results and the real-time communication test is presented through the experimental result.

Communications for Underwater Robotics Research Platforms

This paper presents a distributed protocol for communication among autonomous underwater vehicles. It is a complementary approach for coordination between the autonomous underwater vehicles. This paper mainly describes different methods for underwater communication. One of the methods is brute force approach in which messages are broadcasted to all the communication nodes, which in turn will broadcast the acknowledgement. Issues relating to this brute force approach are time delay, number of hops, power consumption, message collision and other practical issues. These issues are discussed and solved by proposing a new method to improve efficiency of this proposed approach and its effectiveness in communication among autonomous underwater vehicles.

Optimal update frequency model for physical machine state change and virtual machine placement in the cloud

Cloud computing is evolving into the default operational framework running modern data centers. Efficient data center operation is concerned with the total amount of energy consumed as well as assuring adequate resources are available to process all of the incoming work requests. Existing research has demonstrated several algorithms that can be used to determine the optimal number of resources required to service these requests. However, a key issue not addressed in these algorithms is determining the frequency of recalculating the number of required resources. Changing the required resources at a rate slower than the optimal update frequency results in lower energy efficiency due to the over allocation of resources. Changing the resources at a rate higher than the optimal frequency results in insufficient time for systems to change state, which results in SLA violations. In this paper, a stochastic optimization model is presented that determines the optimal update frequency for changing the states of the nodes of the cloud as well as determining the proper frequency for recalculating the maximum expected load, which improves the determination of the optimum number of resources required, therefore maximizes energy efficiency and minimizes SLA violations.

On prediction to dynamically assign heterogeneous microprocessors to the minimum joint power state to achieve Ultra Low Power Cloud Computing

Cloud computing centers are designed to be scalable and to process large varieties of software applications. However, the total power required by cloud computing systems is high since excess processors must be available to service both on-demand applications, as well as existing processes. We describe novel concepts that can enable the introduction of Ultra Low Power Cloud Computing systems. Our approach involves using a variety of heterogeneous processors, each with different power and performance capabilities. By predicting the load and jointly allocating tasks to the processors and dynamically turning off reserve processors, we prove that power reductions of up to 60–80% can be achieved.

Applications and prototype for system of systems swarm robotics

In order to develop a robotic system of systems the robotic platforms must be designed and built. For this to happen, the type of application involved should be clear. Swarm robots need to be self contained and powered. They must also be self governing. Here the authors examine various applications and a prototype robot that may be useful in these scenarios.

Control and simulation of robotic swarms in heterogeneous environments

Simulation provides a low cost method of initial testing of control for robotic swarms. The expansion of robotic swarms to heterogeneous environments drives the need to model cooperative operation in those environments. The Autonomous Control Engineering center at The University of Texas at San Antonio is investigating methods of simulation techniques and simulation environments. This paper presents results from adapting simulation tools for diverse environments.

Optimal calculation overhead for energy efficient cloud workload prediction

Amazon recently estimated that the cost of energy for its datacenters reached 42% of the total cost of operation. Our previous research proposed an algorithm to predict how much cloud workload is expected during a future time interval. Accurate knowledge of the future workload allows the datacenter operator to place unneeded physical servers in a low-power state to save energy. If more system capacity is required, servers in a low-power state are transitioned back to an active state. In this paper, we extend our prior research by presenting a new approach to determining the frequency of calculating the prediction of the expected capacity. We present a dynamic prediction quantization method to determine the optimal number of prediction calculation intervals. These new algorithms allow us to predict future load within required Service Level Agreements while minimizing the number of times the prediction calculations must be performed. We finally test this model by simulating the stochastic time horizon and dynamic quantization algorithms and compare the results with three competing methods. We show that our model provides up to a 20% reduction in the number of calculations required while maintaining the given Service Level Agreement.