The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
To avoid ecological collapse, we must manage Earth's ecosystems sustainably. Viewed as a control problem, the two central challenges of ecosystem management are to acquire a model of the system that is sufficient to guide good decision making and then optimize the control policy against that model. This paper describes three efforts aimed at addressing the first of these challenges—machine learning...
While much research has been done using 2D mesh network as a baseline on-chip network topology, recent multi-core chips from vendors leverage a ring topology. In this work, we re-visit the topology comparison in on-chip networks and model the impact of on-chip network on overall performance while holding the entire chip power constant. We vary the amount of power allocated to the on-chip network and...
Total power dissipation in the next-generation processors is projected to reach 200 W/cm2. Such high power densities coupled with complex integration of devices, results in high heat dissipation on-chip. Therefore, thermal management and temperature-aware computing techniques are critical in designing energy-efficient systems. In this tutorial, we present temperature modeling, floorplanning, design-time...
We presented a communications-inspired circuit-level technique that can take advantage of the error tolerance of the overlying applications to reduce power dissipation, and an architecture technique that takes advantage of the spatial and temporal variations in NoC bandwidth requirements to reduce power dissipation. We believe that such an integrated approach, where we explore the opportunities for...
Computing is poised to make major contributions to the global sustainability effort by engaging the fields of energy efficiency, renewable energy generation, and energy delivery. This paper discusses three near term opportunities for computing in the field of building energy efficiency. Integrated design uses building energy modeling to search for ultra-efficient buildings tailored to use requirements...
In this work we propose a high-speed hybrid optical-time-division-multiplexing (OTDM) and wavelength-division-multiplexing (WDM) system that seamlessly generates high bit-rate data (>200Gbit/s) from a low speed (5Gbit/s) quantum-dot mode locked laser pulse train. The high-speed output data can be generated using electro-optical micro-ring modulators that operate as low as (5Gbit/s). By utilizing...
The decreasing cost of disk drives has made it possible to maintain large-scale storage systems and to achieve high performance for I/O demanding applications. One possible optimization that can be applied to these types of systems is to move the I/O load to a subset of the disks; so that the commonly accessed data resides in a set of ‘popular’ disks, whereas the ‘unpopular’ disks can stay in low...
A priority metric for IT equipment is proposed, and a method for extracting it from historical sensor data is devised. The proposed method consists of classification of cooling efficiency from sampled sensor data and calculation of the priority metric from statistics on these cooling-efficiency classes. A proof-of-concept experiment using the priority metric was conducted in a server room with a thermal...
Continuous technology scaling will enable future many-core architectures to interconnect hundreds and even thousands of cores within multi-chip modules. With increasing number of cores, it becomes essential to reduce the hop count while delivering high interconnect bandwidth at low power. In this paper, we propose SPRINT (Scalable Photonic Reconfigurable Interconnect) that can scale to large cores...
This paper proposes and evaluates the helper thread for Time Aggregation Scheduler (TAS) intended to enhance the performance and energy efficiency. TAS is a kernel-level thread scheduler to enhance performances of multi-threaded programs by focusing on the affinity of sibling threads, which have the same parent process and share the context. TAS may change the scheduling order of threads so that we...
A significant portion of all energy capacity is wasted in over-provisioning to meet peak demand. The current state-of-the-art in reducing peak demand requires central authorities to limit device usage directly, and are generally reactive. We apply techniques drawn from established distributed computing principles to propose a novel and proactive solution to decentralize management of demand and to...
Reliability is a growing fundamental challenge in the design of multiprocessor Systems-on-Chip (MPSoCs). This trend is accelerated by the increasingly adverse process variations and wearout mechanisms that result in an increased number of errors. Previously proposed fault-tolerant techniques are ad-hoc and target processors or Networks-on-Chip (NoC) separately. Because each of these two units may...
This paper addresses the joint optimization of performance, energy, and temperature, termed as PET - optimization. This multi-objective PET-optimization is achieved in scheduling DAGs on multi-core systems. Our technique is based on multi-objective evolutionary algorithm (MOEA) for finding Pareto optimal solutions using scheduling and voltage selection. These solutions are not necessarily scalar values...
Municipal water delivery networks face increasing demands due to population growth. We focus on enhancing a water utility's infrastructure to meet its growing demands in a cost effective manner. Specifically, we consider the problem of placing pressure boosting pumps in a water network such that the minimum required delivery pressure is maintained at all the consumption points in the network for a...
Rapid growth of computing during the last few decades has led to a significant increase in power consumption. With increasing operational costs, the data center industry has started to actively search for efficiency improvements to slow down the rising cost of running services. Some of the most prominent ways to cut power consumption are currently perceived to include economizers and free cooling,...
The increasingly parallel landscape of embedded computing platforms is bringing the reliability concern for the on-chip interconnection network (NoC) to the forefront.
Exascale supercomputers will gather hundreds of million cores. The main problem to take care for running applications on such platforms is energy consumption since it is one major limitation if we consider that the currently fastest supercomputer consumes more than 12MW for a maximum performance of 10PFlops. Besides, we also need to overcome important challenges related to fault tolerance and data...
Harnessing wind near existing electrical infrastructure is a low-cost way to tap alternative energy sources. However, it should be done in a manner considerate to converter owners and neighbors. Wireless sensor networks can be used to control wind-to-electricity converters toward this goal and can aid toward re-scheduling electrical loads.
This paper proposes energy-efficient scheduling algorithms for distributed heterogeneous grids. The conservation of energy consumption has a tirade of effects that improve system reliability, increases in the lifespan of the system, ideally with only a linear degradation in performance. With a few restrictions and proper modeling, we convert the Energy Aware Task Allocation (EATA) problem into a Bargaining...
In this paper, we propose a general methodology for energy estimation of bus-based multi-core processors assuming that DVS can be used for both buses and cores. Our formulation can provide tradeoffs between DVS setting for buses and cores. We examine this methodology using various parallel matrix multiplication algorithms that are suitable for shared memory multicore machines with L1 and L2 caches.
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.