اشترك بالحزمة الذهبية واحصل على وصول غير محدود شمرا أكاديميا
تسجيل مستخدم جديدREOH: Using Probabilistic Network for Runtime Energy Optimization of Heterogeneous Systems
65
0
0.0
(
0
)
اسأل ChatGPT حول البحث
ﻻ يوجد ملخص باللغة العربية
Significant efforts have been devoted to choosing the best configuration of a computing system to run an application energy efficiently. However, available tuning approaches mainly focus on homogeneous systems and are inextensible for heterogeneous systems which include several components (e.g., CPUs, GPUs) with different architectures. This study proposes a holistic tuning approach called REOH using probabilistic network to predict the most energy-efficient configuration (i.e., which platform and its setting) of a heterogeneous system for running a given application. Based on the computation and communication patterns from Berkeley dwarfs, we conduct experiments to devise the training set including 7074 data samples covering varying application patterns and characteristics. Validating the REOH approach on heterogeneous systems including CPUs and GPUs shows that the energy consumption by the REOH approach is close to the optimal energy consumption by the Brute Force approach while saving 17% of sampling runs compared to the previous (homogeneous) approach using probabilistic network. Based on the REOH approach, we develop an open-source energy-optimizing runtime framework for selecting an energy efficient configuration of a heterogeneous system for a given application at runtime.
قيم البحث
اقرأ أيضاً
Dynamic resource management has become one of the major areas of research in modern computer and communication system design due to lower power consumption and higher performance demands. The number of integrated cores, level of heterogeneity and amo
unt of control knobs increase steadily. As a result, the system complexity is increasing faster than our ability to optimize and dynamically manage the resources. Moreover, offline approaches are sub-optimal due to workload variations and large volume of new applications unknown at design time. This paper first reviews recent online learning techniques for predicting system performance, power, and temperature. Then, we describe the use of predictive models for online control using two modern approaches: imitation learning (IL) and an explicit nonlinear model predictive control (NMPC). Evaluations on a commercial mobile platform with 16 benchmarks show that the IL approach successfully adapts the control policy to unknown applications. The explicit NMPC provides 25% energy savings compared to a state-of-the-art algorithm for multi-variable power management of modern GPU sub-systems.
In this paper, we propose the first optimum process scheduling algorithm for an increasingly prevalent type of heterogeneous multicore (HEMC) system that combines high-performance big cores and energy-efficient small cores with the same instruction-s
et architecture (ISA). Existing algorithms are all heuristics-based, and the well-known IPC-driven approach essentially tries to schedule high scaling factor processes on big cores. Our analysis shows that, for optimum solutions, it is also critical to consider placing long running processes on big cores. Tests of SPEC 2006 cases on various big-small core combinations show that our proposed optimum approach is up to 34% faster than the IPC-driven heuristic approach in terms of total workload completion time. The complexity of our algorithm is O(NlogN) where N is the number of processes. Therefore, the proposed optimum algorithm is practical for use.
Convolutional Neural Networks (CNNs) achieved great cognitive performance at the expense of considerable computation load. To relieve the computation load, many optimization works are developed to reduce the model redundancy by identifying and removi
ng insignificant model components, such as weight sparsity and filter pruning. However, these works only evaluate model components static significance with internal parameter information, ignoring their dynamic interaction with external inputs. With per-input feature activation, the model component significance can dynamically change, and thus the static methods can only achieve sub-optimal results. Therefore, we propose a dynamic CNN optimization framework in this work. Based on the neural network attention mechanism, we propose a comprehensive dynamic optimization framework including (1) testing-phase channel and column feature map pruning, as well as (2) training-phase optimization by targeted dropout. Such a dynamic optimization framework has several benefits: (1) First, it can accurately identify and aggressively remove per-input feature redundancy with considering the model-input interaction; (2) Meanwhile, it can maximally remove the feature map redundancy in various dimensions thanks to the multi-dimension flexibility; (3) The training-testing co-optimization favors the dynamic pruning and helps maintain the model accuracy even with very high feature pruning ratio. Extensive experiments show that our method could bring 37.4% to 54.5% FLOPs reduction with negligible accuracy drop on various of test networks.
High-performance computing systems are more and more often based on accelerators. Computing applications targeting those systems often follow a host-driven approach in which hosts offload almost all compute-intensive sections of the code onto acceler
ators; this approach only marginally exploits the computational resources available on the host CPUs, limiting performance and energy efficiency. The obvious step forward is to run compute-intensive kernels in a concurrent and balanced way on both hosts and accelerators. In this paper we consider exactly this problem for a class of applications based on Lattice Boltzmann Methods, widely used in computational fluid-dynamics. Our goal is to develop just one program, portable and able to run efficiently on several different combinations of hosts and accelerators. To reach this goal, we define common data layouts enabling the code to exploit efficiently the different parallel and vector options of the various accelerators, and matching the possibly different requirements of the compute-bound and memory-bound kernels of the application. We also define models and metrics that predict the best partitioning of workloads among host and accelerator, and the optimally achievable overall performance level. We test the performance of our codes and their scaling properties using as testbeds HPC clusters incorporating different accelerators: Intel Xeon-Phi many-core processors, NVIDIA GPUs and AMD GPUs.
In this paper, probabilistic shaping is numerically and experimentally investigated for increasing the transmission reach of wavelength division multiplexed (WDM) optical communication system employing quadrature amplitude modulation (QAM). An optimi
zed probability mass function (PMF) of the QAM symbols is first found from a modified Blahut-Arimoto algorithm for the optical channel. A turbo coded bit interleaved coded modulation system is then applied, which relies on many-to-one labeling to achieve the desired PMF, thereby achieving shaping gain. Pilot symbols at rate at most 2% are used for synchronization and equalization, making it possible to receive input constellations as large as 1024QAM. The system is evaluated experimentally on a 10 GBaud, 5 channels WDM setup. The maximum system reach is increased w.r.t. standard 1024QAM by 20% at input data rate of 4.65 bits/symbol and up to 75% at 5.46 bits/symbol. It is shown that rate adaptation does not require changing of the modulation format. The performance of the proposed 1024QAM shaped system is validated on all 5 channels of the WDM signal for selected distances and rates. Finally, it was shown via EXIT charts and BER analysis that iterative demapping, while generally beneficial to the system, is not a requirement for achieving the shaping gain.
سجل دخول لتتمكن من نشر تعليقات
التعليقات
جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
