Nanyan Jiang

Enabling autonomic power-aware management of instrumented data centers

Sensor networks support flexible, non-intrusive and fine-grained data collection and processing and can enable online monitoring of data center operating conditions as well as autonomic data center management. This paper describes the architecture and implementation of an autonomic power aware data center management framework, which is based on the integration of embedded sensors with computational models and workload schedulers to improve data center performance in terms of energy consumption and throughput. Specifically, workload schedulers use online information about data center operating conditions obtained from the sensors to generate appropriate management policies. Furthermore, local processing within the sensor network is used to enable timely responses to changes in operating conditions and determine job migration strategies. Experimental results demonstrate that the framework achieves near optimal management, and in-network analysis enables timely response while reducing overheads.

ASGrid: autonomic management of hybrid sensor grid systems and applications

In this paper, we propose an autonomic management framework (ASGrid) to address the requirements of emerging large-scale applications in hybrid grid and sensor network systems. To the best of our knowledge, we are the first who proposed the notion of autonomic sensor grid systems in a holistic manner, aiming at non-trivial large applications. To bridge the gap between the physical world and the digital world and facilitate information analysis and decision making, ASGrid is designed to smooth the integration of sensor networks and grid systems and efficiently use both on demand. Under the blueprint of ASGrid, we present several building blocks that fulfill the following major features: (1) self-configuration through content-based aggregation and associative rendezvous mechanisms; (2) self-optimisation through utility-based sensor selection, model-driven hierarchical sensing task scheduling and auction-based game-theoretic approach for grid scheduling; (3) self-protection through ActiveKey dynamic key management and S3Trust trust management mechanisms. Experimental and simulation results on these aspects are presented.

A Decentralized Content-based Aggregation Service for Pervasive Environments

This paper presents a content-based decentralized information aggregation service for pervasive environments. The service provides a uniform query interface where aggregation queries are specified using content descriptors in the form of keywords, partial keywords, wildcards and ranges. The service guarantees that all data that matches a complex/range query was located and aggregated in an efficient and scalable way. The design of the aggregation service uses a decentralized aggregation trie along with a distributed and decentralized query engine. The deployment and experimental evaluation of the aggregation service are also presented. Evaluations include simulations as well as experiments using deployments on local-area network at Rutgers and wide-area PlanetLab testbed

A programming system for sensor-driven scientific applications

Technical advances are leading to a pervasive computational ecosystem that integrates computing infrastructures with embedded sensors and actuators, and giving rise to a new paradigm for monitoring, understanding, and managing natural and engineered systems – one that is information/data-driven. This research investigates a programming system that can support such end-to-end sensor-based dynamic data-driven applications. Specifically, it enables these applications at two levels. First, it provides programming abstractions for integrating sensor systems with computational models for scientific and engineering processes and with other application components in an end-to-end experiment. Second, it provides programming abstractions and system software support for developing in-network data processing mechanisms. The former supports complex querying of the sensor system, while the latter enables development of in-network data processing mechanisms such as aggregation, adaptive interpolation and assimilation, both via semantically meaningful abstractions. For the latter, we explore the temporal and spatial correlation of sensor measurements in the targeted application domains to tradeoff between the complexity of coordination among sensor clusters and the savings that result from having fewer sensors for in-network processing, while maintaining an acceptable error threshold. Experimental results show that the proposed in-network mechanisms can facilitate the efficient usage of constraint resources and satisfy data requirement in the presence of dynamics and uncertainty. The research presented in this thesis is evaluated using two application scenarios: (1) the management and optimization of an instrumented oil field; and (2) the management and optimization of an instrumented data center. In the first scenario, the programming abstractions and systems software solutions enable end-to-end management processes for detecting and tracking reservoir changes, assimilating and inverting data for determining reservoir properties, and providing feedback to enhance temporal and spatial resolutions and track other specific processes in the subsurface. The overall goal is to ensure near optimal operation of the reservoir in terms of profitability, safety and/or environmental impact. In the second scenario, the autonomic instrumented data center management system addresses power consumption, heat generation and cooling requirements of the data center, which are critical concerns especially as the scales of these computing environments grow. Experimental results show that the provided programming system reduces overheads while achieving near optimal and timely management and control in both application scenarios.

Programming support for sensor-based scientific applications

Technical advances are enabling a pervasive computational ecosystem that integrates computing infrastructures with embedded sensors and actuators, and are giving rise to a new paradigm for monitoring, understanding, and managing natural and engineered systems - one that is information/data-driven. This research investigates programming systems for sensor-driven applications. It addresses abstractions and runtime mechanisms for integrating sensor systems with computational models for scientific processes, as well as for in- network data processing, e.g., aggregation, adaptive interpolation and assimilation. The current status of this research, as well as initial results are presented.

Autonomic Management of Hybrid Sensor Grid Systems and Applications

In this paper, we propose an autonomic management framework (ASGrid) to address the requirements of emerging large-scale applications in hybrid grid and sensor network systems. To the best of our knowledge, we are the first who proposed the autonomic sensor grid system concept in a holistic manner targeted at non-trivial large applications. To bridge the gap between the physical world and the digital world and facilitate information analysis and decision making, ASGrid is designed to smooth the integration of sensor networks and grid systems and efficiently use both on demand. Under the blueprint of ASGrid, we present several building blocks that fulfill the following major features: (1) Self-configuration through content-based aggregation and associative rendezvous mechanisms; (2) Self-optimization through utility-based sensor selection and model-driven hierarchical sensing task scheduling; (3) Self-protection through ActiveKey dynamic key management and S3Trust trust management mechanisms. Experimental and simulation results on these aspects are presented.

Enabling End-to-End Data-Driven Sensor-Based Scientific and Engineering Applications

Technical advances are leading to a pervasive computational infrastructure that integrates computational processes with embedded sensors and actuators, and giving rise to a new paradigm for monitoring, understanding, and managing natural and engineered systems --- one that is information/data-driven. However, developing and deploying these applications remains a challenge, primarily due to the lack of programming and runtime support. This paper addresses these challenges and presents a programming system for end-to-end sensor/actuator-based scientific and engineering applications. Specifically, the programming system provides semantically meaningful abstractions and runtime mechanisms for integrating sensor systems with computational models for scientific processes, and for in-network data processing such as aggregation, adaptive interpolation and assimilations. The overall architecture of the programming system and the design of its key components, as well as its prototype implementation are described. An end-to-end dynamic data-driven oil reservoir application that combines reservoir simulation models with sensors/actuators in an instrumented oilfield is used as a case study to demonstrate the operation of the programming system, as well as to experimentally demonstrate its effectiveness and performance.

In-network data estimation for sensor-driven scientific applications

Sensor networks employed by scientific applications oftenneed to support localized collaboration of sensor nodes to perform in-network data processing. This includes new quantitative synthesis andhypothesis testing in near real time, as data streaming from distributedinstruments, to transform raw data into high level domain-dependent information. This paper investigates in-network data processing mechanismswith dynamic data requirements in resource constrained heterogeneoussensor networks. Particularly, we explore how the temporaland spatial correlation of sensor measurements can be used to trade offbetween the complexity of coordination among sensor clusters and thesavings that result from having fewer sensors involved in in-network processing,while maintaining an acceptable error threshold. Experimentalresults show that the proposed in-network mechanisms can facilitate theefficient usage of resources and satisfy data requirement in the presenceof dynamics and uncertainty.

Opportunistic application flows in pervasive environments

Summary form only given. Supporting applications in sensor-based pervasive environments requires a programming and management paradigm where the behaviors as well as the interactions of applications elements (sensors, actuators, and services) are dynamic, opportunistic, and context, content and capability aware. We present a programming model that enables opportunistic flows in pervasive environments. The model builds on content-based discovery and routing services and defines associative rendezvous as an abstraction for content-based decoupled interactions. Cascading local behaviors (CLB) then build on associative rendezvous to enable opportunistic application flows to emerge as a result of context and content based local behaviors. We also present the design, prototype implementation and experimental evaluation of the Meteor programming framework and content-based middleware.