Smart Cities, Wireless Sensor Networks, and Smart Grid Technology
In an increasingly informed world, generating and processing information encompasses several computing domains, from datacenters to embedded systems. SEELab research in this area includes efficient distributed data collection and aggregation to processing and adapting to this data in smart cities, data centers, and the distributed smart grid.
Green Energy, Smart Grids, and The Internet of Things
Renewable energy and smart grid research encompasses several domains, from data centers to embedded systems. Data centers are some of the largest energy consumers with the highest carbon emissions, but their short response times inhibit widespread use of renewable energy sources. SEELab has developed new algorithms for short-term solar and wind energy prediction, leveraging the added information from prediction to increase green energy efficiency and workload completion, as well as extended this work to networked data centers and green-energy-aware routing (GEAR). Residential energy consumption involves analyzing and reducing load energy behavior in the home. Appliances and other loads can be modeled and predicted using machine learning techniques, enabling energy saving through peak power shaving, smart scheduling, and home automation. The larger perspective is the impact of local energy behavior changes on the smart grid for convergent solutions on energy pricing and generation.
Long-term research requiring high-resolution sensor data need platforms large enough to house solar panels and batteries. Leveraging a well-defined sensor appliance created using Sensor-Rocks, we develop novel context-aware power management algorithms to maximize network lifetime and provide unprecedented capability on miniaturized platforms.
Energy Efficient Routing and Scheduling For Ad-Hoc Wireless Networks
In large-scale ad hoc wireless networks, data delivery is complicated by the lack of network infrastructure and limited energy resources. We propose a novel scheduling and routing strategy for ad hoc wireless networks which achieves up to 60% power savings while delivering data efficiently. We test our ideas on a heterogeneous wireless sensor network deployed in southern California - HPWREN.
SHiMmer is a wireless platform that combines active sensing and localized processing with energy harvesting to provide long-lived structural health monitoring. Unlike other sensor networks that periodically monitor a structure and route information to a base station, our device acquires data and processes it locally before communicating with an external device, such as a remote controlled helicopter.
Data Center Energy Efficiency
Energy consumption has become an important problem for large scale data centers due to high and constant power demand and the cost associated with it. The cost of energy consumption of data centers has reached and even surpassed the cost of the physical data center itself. This necessitates research for dynamically reducing the amount of energy used for computing, cooling and maintaining a data center. The primary motivation for this work is to develop different data center power management schemes that deliver energy efficiency while minimizing the impact on performance. We approach to this high consumption problem in various ways. Our approaches include virtual machine (VM) level energy aware scheduling, efficient renewable energy integration in data centers, backbone network and data center peak power management with batteries and augmenting datacenter networks with optical technology. The first one identifies the requirements and characteristics of different workloads running in VMs and schedules VMs across servers to save energy without degrading their quality of service requirements. The second approach uses prediction to improve the reliability of the renewable energy input and significantly reduces the brown energy cost while also improving the performance of the jobs. The third approach reduces the peak power draw of data centers by redirecting the power flow with the help of batteries to reduce the cost associated with the peak power draw from the utility. And the last one explores the bandwidth- and energy-efficient potencial of integrating optical circuit switching in a datacenter to increase performance of applications and mitigate degradations realted to network oversubscription. This project is a part of the Multi Scale Systems Center (MuSyC) .
Temperature Modeling and Variability-aware Reliability Management for Multiprocessor Systems
Thermal hot spots and temperature variations have brought new challenges in reliability, performance and cooling costs in deep submicron system-on-chips (SoCs). Moreover, static variations are caused by manufacturing process imprecision, causing speed and power consumption different from nominal. Counteracting these issues by increasing the design margin is a costly strategy. A more promising approach lowers the design margin and exposes variability to the software stack, to manage it at runtime. The information from these sensors can be exploited to control aging and performance by voltage/frequency assignment and task scheduling. Important studies regarding static variations have already been presented by the Micrel Lab at University of Bologna, Italy. They propose a low-overhead sub-optimal task allocation that minimizes the energy consumption of variability-affected multicore platforms while minimizing the energy consumptions. We instead focus a novel variation-aware dynamic reliability management policy which is able to meet lifetime constraints while minimizing the impact on performance thanks to a user-experience aware sub-control. In addition, to providing an optimal solution for minimizing and balancing temperature on the die, we proposed both reactive and proactive dynamic job scheduling techniques. A related aspect regards the side effects that the cooling infrastructure of data centers have. In particular server fan subsystems are power-hungry and generate vibrations, degrading the performance of data-intensive workloads and inflating the uptime electric bills of cost-sensitive data centers. We propose a systematic measurement methodology to isolate these mechanical effects, benchmark database workloads, and perform detailed thermal simulations of the various server components, including multi-core processors, memory modules, and variable-speed fans.
Mobile Architectures and Platforms
Mobiles are going to play a pivotal role in the growth of computing industry in the coming decades. In our group, we look at various aspects of mobile system design covering a broad spectrum of topics across the computing stack (from VLSI to software). One of the issues in mobile system design is the absence of good representative benchmarks. In our group, we are actively looking at 'mobile workload design' with initial emphasis on characterizing the compute and the memory-bandwidth usage by commonly used apps on state-of-the-art smartphone/tablet platforms. More importantly, this study will also include the characterization of the usage of heterogeneous components (like GPUs), an aspect overlooked in previous workload studies. Heat dissipation is another key challenge in mobile devices and research to date has only considered thermal management of processors, while neglecting the effects of packaging, skin temperature and the rest of the system. We instead plan to consider the entire mobile system, including the RF subsystem, CODEC, power IC design, battery, other PCB components, device packaging and human skin temperature. In our work we specifically intend to analyze the effects these components have on thermal issues observed inside the processor. Since higher levels of integration is key to optimizing performance and energy efficiency, we are also focusing on the 3D stack heterogeneous multicore processors with stacked memory, as the future central processing units in mobile designs. Another important issue is the battery and the associated DC-DC power conversion circuitry. Keeping under control the power dissipation of DC-DC converters has a double impact, namely, increasing the energy efficiency and decreasing the temperature of the sub-system. Even in sophisticated DC-DC power conversion architectures, the efficiency strongly depends on the current request of power, how often this request changes over time and the current state of charge of the battery. In summary, workload design, system level thermal behaviour, 3d stacking and battery runtime optimizations are some of the interesting problems in mobile system design that we are actively investigating in our group.
A key challenge observed during the clinical trials of preventive mobile healthcare systems was extending the battery lifetime of the system. We analyzed that in such systems with processing, sensing, and communication tasks, it is difficult to dynamically allocate and schedule these tasks on the devices. However, how we allocate and schedule tasks becomes very important if we want to achieve a good tradeoff between battery lifetime and high fidelity of sensed and processed data. Since task binding optimization has been shown to be NP complete, we use an optimal and static ILP formulation of this problem as a baseline comparison to our own dynamic heuristic algorithm which executes at system run time with very low overhead. In a past project, a reconfigurable embedded sensor node, SunSPOT, was shown to significantly extend its set of capabilities through run-time reconfiguration. In this project EKG patterns are tracked in patients and analyzed. The device can then reconfigure itself as necessary to be able to detect different types of heart conditions at very low cost.
Energy Management in Heterogeneous Memory Architectures
Power consumption is a major concern in the design of modern systems. Recent studies have shown that current main memory system based on DRAM has become a significant energy consumer, contributing to as much as 30-40% of total consumption on modern server systems. Our goal is to employ novel memory technologies such as Phase-change RAM (PRAM) and to architect a heterogenous memory system to optimize performance, cost, and power. While PRAM consumes little static power due to its non-volatility, it introduces new challenges associated with limited endurance and high power cost of write accesses. Our experimental results of a heterogeneous system composing both DRAM and PRAM indicate that our solution is able to achieve energy savings up to 37% at negligible overhead over conventional memory architecture.
Event-driven Power Management
Power management (PM) algorithms aim at reducing energy consumption at the system-level by selectively placing components into low-power states. Formerly, two classes of heuristic algorithms have been proposed for power management: timeout and predictive. Later, a category of algorithms based on stochastic control was proposed for power management. These algorithms guarantee optimal results as long as the system that is power managed can be modeled well with exponential distributions. Another advantage is that they can meet performance constraints, something that is not possible with heuristics. We show that there is a large mismatch between measurements and simulation results if the exponential distribution is used to model all user request arrivals. We develop two new approaches that better model system behavior for general user request distributions. These approaches are event driven and give optimal results verified by measurements. The first approach is based on renewal theory. This model assumes that the decision to transition to low power state can be made in only one state. Another method we developed is based on the Time-Indexed Semi-Markov Decision Process model (TISMDP). This model allows for transitions into low power states from any state, but it is also more complex than our other approach. The results obtained by renewal model are guaranteed to match results obtained by TISMDP model, as both approaches give globally optimal solutions. We implemented our power management algorithms on two different classes of devices and the measurement results show power savings ranging from a factor of 1.7 up to 5.0 with insignificant variation in performance.
Energy-efficient software design
Time to market of embedded software has become a crucial issue. As a result, embedded software designers often use libraries that have been preoptimized for a given processor to achieve higher code quality. Unfortunately, current software design methodology often leaves high-level arithmetic optimizations and the use of complex library elements up to the designers' ingenuity. We present a tool flow and a methodology that automates the use of complex processor instructions and pre-optimized software library routines using symbolic algebraic techniques. It leverages our profiler that relates energy consumption to the source code and allows designers to quickly obtain energy consumption breakdown by procedures in their source code.
Energy-efficient wireless communication
Today’s wireless networks are highly heterogeneous with diverse range requirements and QoS. Since the battery lifetime is limited, power management of the communication interfaces without any significant degradation in performance has become essential. We show a set of different approaches that efficiently reduce power consumption under different environments and applications.When multiple wireless network interfaces (WNICs) are available, we propose a policy to decides what WNIC to employ for a given application and how to optimize the its usage leading to a large improvement in power savings. In the case of client-server multimedia applications running on wireless portable devices, we can exploit the server knowledge of the workload. We present a client- and a server-PM that by exchanging power control information can achieve more than 67 % with no performance loss. Wireless communication represents a critical aspect also in the design of specific applications such as distributed speech recognition in portable devices. We consider quality-of-service tradeoffs and overall system latency and present a wireless LAN scheduling algorithm to minimize the energy consimption of a distributed speech recognition front-end.