An official website of the United States government
We propose CURE, a deep reinforcement learning (DRL)-based NoC design framework that simultaneously reduces network latency, improves energy-efficiency, and tolerates transient errors and permanent faults. CURE has several architectural innovations and a DRL-based hardware controller to manage design complexity and optimize trade-offs. First, in CURE, we propose reversible multi-function adaptive channels (RMCs) to reduce NoC power consumption and network latency. Second, we implement a new fault-secure adaptive error correction hardware in each router to enhance reliability for both transient errors and permanent faults. Third, we propose a router power-gating and bypass design that powers off NoC components to reduce power and extend chip lifespan. Further, for the complex dynamic interactions of these techniques, we propose using DRL to train a proactive control policy to provide improved fault-tolerance, reduced power consumption, and improved performance. Simulation using the PARSEC benchmark shows that CURE reduces end-to-end packet latency by 39%, improves energy efficiency by 92%, and lowers static and dynamic power consumption by 24% and 38%, respectively, over conventional solutions. Using mean-time-to-failure, we show that CURE is 7.7x more reliable than the conventional NoC design.
more »
« less
Training self-learning circuits for power-efficient solutions
As the size and ubiquity of artificial intelligence and computational machine learning models grow, the energy required to train and use them is rapidly becoming economically and environmentally unsustainable. Recent laboratory prototypes of self-learning electronic circuits, such as “physical learning machines,” open the door to analog hardware that directly employs physics to learn desired functions from examples at a low energy cost. In this work, we show that this hardware platform allows for an even further reduction in energy consumption by using good initial conditions and a new learning algorithm. Using analytical calculations, simulations, and experiments, we show that a trade-off emerges when learning dynamics attempt to minimize both the error and the power consumption of the solution—greater power reductions can be achieved at the cost of decreasing solution accuracy. Finally, we demonstrate a practical procedure to weigh the relative importance of error and power minimization, improving the power efficiency given a specific tolerance to error.
more »
« less
- Award ID(s):
- 2005749
- PAR ID:
- 10588132
- Publisher / Repository:
- American Institute of Physics
- Date Published:
- Journal Name:
- APL Machine Learning
- Volume:
- 2
- Issue:
- 1
- ISSN:
- 2770-9019
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Datacenter capacity is growing exponentially to satisfy the increasing demand for many emerging computationally-intensive applications, such as deep learning. This trend has led to concerns over datacenters’ increasing energy consumption and carbon footprint. The most basic prerequisite for optimizing a datacenter’s energy- and carbon-efficiency is accurately monitoring and attributing energy consumption to specific users and applications. Since datacenter servers tend to be multi-tenant, i.e., they host many applications, server- and rack-level power monitoring alone does not provide insight into the energy usage and carbon emissions of their resident applications. At the same time, current application-level energy monitoring and attribution techniques are intrusive: they require privileged access to servers and necessitate coordinated support in hardware and software, neither of which is always possible in cloud environments. To address the problem, we design WattScope, a system for non-intrusively estimating the power consumption of individual applications using external measurements of a server’s aggregate power usage and without requiring direct access to the server’s operating system or applications. Our key insight is that, based on an analysis of production traces, the power characteristics of datacenter workloads, e.g., low variability, low magnitude, and high periodicity, are highly amenable to disaggregation of a server’s total power consumption into application-specific values. WattScope adapts and extends a machine learning-based technique for disaggregating building power and applies it to server- and rack-level power meter measurements that are already available in data centers. We evaluate WattScope’s accuracy on a production workload and show that it yields high accuracy, e.g., often 10% normalized mean absolute error, and is thus a potentially useful tool for datacenters in externally monitoring application-level power usage.more » « less
-
Summary Energy‐efficient scientific applications require insight into how high performance computing system features impact the applications' power and performance. This insight can result from the development of performance and power models. In this article, we use the modeling and prediction tool MuMMI (Multiple Metrics Modeling Infrastructure) and 10 machine learning methods to model and predict performance and power consumption and compare their prediction error rates. We use an algorithm‐based fault‐tolerant linear algebra code and a multilevel checkpointing fault‐tolerant heat distribution code to conduct our modeling and prediction study on the Cray XC40 Theta and IBM BG/Q Mira at Argonne National Laboratory and the Intel Haswell cluster Shepard at Sandia National Laboratories. Our experimental results show that the prediction error rates in performance and power using MuMMI are less than 10% for most cases. By utilizing the models for runtime, node power, CPU power, and memory power, we identify the most significant performance counters for potential application optimizations, and we predict theoretical outcomes of the optimizations. Based on two collected datasets, we analyze and compare the prediction accuracy in performance and power consumption using MuMMI and 10 machine learning methods.more » « less
-
During machining, kinetic energy is imparted to a workpiece to remove material. The integrity of the machined surface, which depends on the energy transfer, affects the quality and performance of the product, therefore needs to be quantified. Prior studies have indicated the potential of using machining power, or the power consumption at the tool-chip interface, as a process signature for predicting machined surface integrity. However, direct measurement of machining power is constrained by the availability of special equipment and the associated cost. To address this gap, this paper presents a machine learning-based method for machining power prediction through multi-sensor fusion and sequence-to-sequence translation from acoustic and vibration signals, which represent portions of the in-situ kinetic energy dissipation, to the machining power signal as a process signature. Specifically, a neural network architecture is developed to separately translate the acoustic and vibration signals to corresponding machining power signals. The two predicted power signals are subsequently fused to arrive at a unified power signal prediction. To check for spurious decision logic, the sensor fusion model is interpreted using integrated gradients to reveal temporal regions of the input data which have the most influence on the machining power prediction accuracy of the fusion model. Systematic cutting experiments performed on a lathe using 1018 steel have shown that the developed sensor fusion method for process signature prediction can successfully map machine acoustics to power consumption with 5.6% error, tool vibration to power consumption with 8.2% error, and acoustics and vibration, jointly, to power with 2.5% error. Model parameter interpretation reveals that the vibration signal is more influential on the machining power prediction result than the acoustic signal, but that overall model accuracy is diminished if only the vibration signal is used.more » « less
-
Advancements in mobile hardware and streaming technologies enable high-quality video streaming for mobile users, but this comes at a cost: a boost in power consumption. Despite detailed studies on power consumption during acquisition, existing studies fall short of considering recent technologies and, hence, of accurately capturing video playback power consumption. This paper presents a novel method to model mobile video playback power consumption. First, we identify the major components contributing to power consumption during video playback on mobile devices. Then, we develop models for each component to estimate their power consumption. Our experimental results show that our combined model estimates power consumption with 91% mean accuracy. Furthermore, our model maintains its high accuracy on an unseen device, achieving 88% mean accuracy despite the hardware and screen heterogeneity.more » « less
