Thermal Tracking and Estimation for Microprocessors

Due to increasing integration density and operating frequency of today's high performance processors, the temperature of a typical chip can easily exceed 100 degrees Celsius. However, the runtime thermal state of a chip is very hard to predict and manage due to the random nature in computing workloads, as well as the process, voltage and ambient temperature variability (together called PVT variability). The uneven nature (both in time and space) of the heat dissipation of the chip could lead to severe reliability issues and error-prone chip behavior (e.g. timing errors). Many dynamic power/thermal management techniques have been proposed to address this issue such as dynamic voltage and frequency scaling (DVFS), clock gating and etc. However, most of such techniques require accurate knowledge of the runtime thermal state of the chip to make efficient and effective control decisions. In this work we address the problem of tracking and managing the temperature of microprocessors which include the following sub-problems: (1) how to design an efficient sensor-based thermal tracking system on a given design that could provide accurate real-time temperature feedback; (2) what statistical techniques could be used to estimate the full-chip thermal profile based on very limited (and possibly noise-corrupted) sensor observations; (3) how do we adapt to changes in the underlying system's behavior, since such changes could impact the accuracy of our thermal estimation. The thermal tracking methodology proposed in this work is enabled by on-chip sensors which are already implemented in many modern processors. We first investigate the underlying relationship between heat distribution and power consumption, then we introduce an accurate thermal model for the chip system. Based on this model, we characterize the temperature correlation that exists among different chip modules and explore statistical approaches (such as those based on Kalman filter) that could utilize such correlation to estimate the accurate chip-level thermal profiles in real time. Such estimation is performed based on limited sensor information because sensors are usually resource constrained and noise-corrupted. We also took a further step to extend the standard Kalman filter approach to account for (1) nonlinear effects such as leakage-temperature interdependency and (2) varying statistical characteristics in the underlying system model. The proposed thermal tracking infrastructure and estimation algorithms could consistently generate accurate thermal estimates even when the system is switching among workloads that have very distinct characteristics. Through experiments, our approaches have demonstrated promising results with much higher accuracy compared to existing approaches. Such results can be used to ensure thermal reliability and improve the effectiveness of dynamic thermal management techniques.

ANALYSIS OF MACROINVERTEBRATE COMMUNITIES IN SEASONAL WETLANDS THROUGH TIME, ACROSS SPACE, AND USING SPECIES TRAITS

Restoration of natural wetlands may be informed by macroinvertebrate community composition. Macroinvertebrate communities of wetlands are influenced by environmental characteristics such as vegetation, soil, hydrology, land use, and isolation. This dissertation explores multiple approaches to the assessment of wetland macroinvertebrate community composition, and demonstrates how these approaches can provide complementary insights into the community ecology of aquatic macroinvertebrates. Specifically, this work focuses on macroinvertebrates of Delmarva Bays, isolated seasonal wetlands found on Maryland’s eastern shore. A comparison of macroinvertebrate community change over a nine years in a restored wetland complex indicated that the macroinvertebrate community of a rehabilitated wetlands more rapidly approximated the community of a reference site than did a newly created wetland. The recovery of a natural macroinvertebrate community in the rehabilitated wetland indicated that wetland rehabilitation should be prioritized over wetland creation and long-term monitoring may be needed to evaluate restoration success. This study also indicated that characteristics of wetland vegetation reflected community composition. The connection between wetland vegetation and macroinvertebrate community composition led to a regional assessment of predaceous diving beetle (Coleoptera: Dytiscidae) community composition in 20 seasonal wetlands, half with and half without sphagnum moss (Sphagnum spp.). Species-level identifications indicated that wetlands with sphagnum support unique and diverse assemblages of beetles. These patterns suggest that sphagnum wetlands provide habitat that supports biodiversity on the Delmarva Peninsula. To compare traits of co-occurring beetles, mandible morphology and temporal and spatial variation were measured between three species of predaceous diving beetles. Based on mandible architecture, all species may consume similarly sized prey, but prey characteristics likely differ in terms of piercing force required for successful capture and consumption. Therefore, different assemblages of aquatic beetles may have different effects on macroinvertebrate community structure. Integrating community-level and species-level data strengthens the association between individual organisms and their ecological role. Effective restoration of imperiled wetlands benefits from this integration, as it informs the management practices that both preserve biodiversity and promote ecosystem services.