Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userDirect N-body application on low-power and energy-efficient parallel architectures
181
0
0.0
(
0
)
Ask ChatGPT about the research
No Arabic abstract
The aim of this work is to quantitatively evaluate the impact of computation on the energy consumption on ARM MPSoC platforms, exploiting CPUs, embedded GPUs and FPGAs. One of them possibly represents the future of High Performance Computing systems: a prototype of an Exascale supercomputer. Performance and energy measurements are made using a state-of-the-art direct $N$-body code from the astrophysical domain. We provide a comparison of the time-to-solution and energy delay product metrics, for different software configurations. We have shown that FPGA technologies can be used for application kernel acceleration and are emerging as a promising alternative to traditional technologies for HPC, which purely focus on peak-performance than on power-efficiency.
rate research
Read More
This work arises on the environment of the ExaNeSt project aiming at design and development of an exascale ready supercomputer with low energy consumption profile but able to support the most demanding scientific and technical applications. The ExaNeSt compute unit consists of densely-packed low-power 64-bit ARM processors, embedded within Xilinx FPGA SoCs. SoC boards are heterogeneous architecture where computing power is supplied both by CPUs and GPUs, and are emerging as a possible low-power and low-cost alternative to clusters based on traditional CPUs. A state-of-the-art direct $N$-body code suitable for astrophysical simulations has been re-engineered in order to exploit SoC heterogeneous platforms based on ARM CPUs and embedded GPUs. Performance tests show that embedded GPUs can be effectively used to accelerate real-life scientific calculations, and that are promising also because of their energy efficiency, which is a crucial design in future exascale platforms.
Cosmological large scale structure $N$-body simulations are computation-light, memory-heavy problems in supercomputing. The considerable amount of memory is usually dominated by an inefficient way of storing more than sufficient phase space information of particles. We present a new parallel, information-optimized, particle-mesh-based $N$-body code CUBE, in which information-efficiency and memory-efficiency are increased by nearly an order of magnitude. This is accomplished by storing particles relative phase space coordinates instead of global values, and in the format of fixed point as light as 1 byte. The remaining information is given by complementary density and velocity fields (negligible in memory space) and proper ordering of particles (no extra memory). Our numerical experiments show that this information-optimized $N$-body algorithm provides accurate results within the error of the particle-mesh algorithm. This significant lowering of the memory-to-computation ratio breaks the bottleneck of scaling up and speeding up large cosmological $N$-body simulations on multi-core and heterogeneous computing systems.
Efficient brain simulation is a scientific grand challenge, a parallel/distributed coding challenge and a source of requirements and suggestions for future computing architectures. Indeed, the human brain includes about 10^15 synapses and 10^11 neurons activated at a mean rate of several Hz. Full brain simulation poses Exascale challenges even if simulated at the highest abstraction level. The WaveScalES experiment in the Human Brain Project (HBP) has the goal of matching experimental measures and simulations of slow waves during deep-sleep and anesthesia and the transition to other brain states. The focus is the development of dedicated large-scale parallel/distributed simulation technologies. The ExaNeSt project designs an ARM-based, low-power HPC architecture scalable to million of cores, developing a dedicated scalable interconnect system, and SWA/AW simulations are included among the driving benchmarks. At the joint between both projects is the INFN proprietary Distributed and Plastic Spiking Neural Networks (DPSNN) simulation engine. DPSNN can be configured to stress either the networking or the computation features available on the execution platforms. The simulation stresses the networking component when the neural net - composed by a relatively low number of neurons, each one projecting thousands of synapses - is distributed over a large number of hardware cores. When growing the number of neurons per core, the computation starts to be the dominating component for short range connections. This paper reports about preliminary performance results obtained on an ARM-based HPC prototype developed in the framework of the ExaNeSt project. Furthermore, a comparison is given of instantaneous power, total energy consumption, execution time and energetic cost per synaptic event of SWA/AW DPSNN simulations when executed on either ARM- or Intel-based server platforms.
We introduce here our new approach to modeling particle cloud evolution off surface of small bodies (asteroids and comets), following the evolution of ejected particles requires dealing with various time and spatial scales, in an efficient, accurate and modular way. In order to improve computational efficiency and accuracy of such calculations, we created an N-body modeling package as an extension to the increasingly popular orbital dynamics N-body integrator Rebound. Our code is currently a stand-alone variant of the Rebound code and is aimed at advancing a comprehensive understanding of individual particle trajectories, external forcing, and interactions, at the scale which is otherwise overlooked by other modeling approaches. The package we developed -- Rebound Ejecta Dynamics (RED) -- is a Python-based implementation with no additional dependencies. It incorporates several major mechanisms that affect the evolution of particles in low-gravity environments and enables a more straightforward simulation of combined effects. We include variable size and velocity distributions, solar radiation pressure, ellipsoidal gravitational potential, binary or triple asteroid systems, and particle-particle interactions. In this paper, we present a sample of the RED package capabilities. These are applied to a small asteroid binary system (characterized following the Didymos/Dimorphos system, which is the target for NASAs Double Asteroid Redirection Test mission)
More and more massive parallel codes running on several hundreds of thousands of cores enter the computational science and engineering domain, allowing high-fidelity computations on up to trillions of unknowns for very detailed analyses of the underlying problems. During such runs, typically gigabytes of data are being produced, hindering both efficient storage and (interactive) data exploration. Here, advanced approaches based on inherently distributed data formats such as HDF5 become necessary in order to avoid long latencies when storing the data and to support fast (random) access when retrieving the data for visual processing. Avoiding file locking and using collective buffering, write bandwidths to a single file close to the theoretical peak on a modern supercomputing cluster were achieved. The structure of the output file supports a very fast interactive visualisation and introduces additional steering functionality.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
