Faster and cheaper: how graphics processing units on spot-market instances minimize turnaround time and budget

2020 ◽  
pp. 1-67
Author(s):  
Nicholas T. Okita ◽  
Tiago A. Coimbra
Keyword(s):  
Cloud Computing ◽  
Graphics Processing Units ◽  
High Performance ◽  
Imaging Techniques ◽  
Spot Market ◽  
Turnaround Time ◽  
On Demand ◽  
Order Of Magnitude ◽  
Graphics Processing ◽  
Price Differences

Cloud computing is enabling users to instantiate and access high-performance computing clusters quickly. However, without proper knowledge of the type of application and the nature of the instances, it can become quite expensive. The objective is to show that adequately choosing the instances provides a fast execution, which, in turn, leads to a low execution price, using the pay-as-you-go model on cloud computing. We used graphics processing units instances on the spot market to execute a seismic-dataset interpolation job and compared their performance to regular on-demand CPU instances. Furthermore, we explored how scaling could also improve the execution times at small price differences. The experiments have shown that, by using an instance with eight accelerators on the spot market, we obtain up to three hundred times speed-up compared to the on-demand CPU options, while being one hundred times cheaper. Finally, our results have shown that seismic-imaging processing can be sped up by order of magnitude with a low budget, resulting in faster and cheaper turn around processing time and enabling possible new imaging techniques.

scholarly journals Gappy Pattern Matching on GPUs for On-Demand Extraction of Hierarchical Translation Grammars

2015 ◽  
Vol 3 ◽  
pp. 87-100
Author(s):  
Hua He ◽  
Jimmy Lin ◽  
Adam Lopez
Keyword(s):  
Hierarchical Models ◽  
Graphics Processing Units ◽  
General Purpose ◽  
Practical Applications ◽  
On Demand ◽  
Mt Evaluation ◽  
Order Of Magnitude ◽  
Evaluation Dataset ◽  
The Cost ◽  
Graphics Processing

Grammars for machine translation can be materialized on demand by finding source phrases in an indexed parallel corpus and extracting their translations. This approach is limited in practical applications by the computational expense of online lookup and extraction. For phrase-based models, recent work has shown that on-demand grammar extraction can be greatly accelerated by parallelization on general purpose graphics processing units (GPUs), but these algorithms do not work for hierarchical models, which require matching patterns that contain gaps. We address this limitation by presenting a novel GPU algorithm for on-demand hierarchical grammar extraction that is at least an order of magnitude faster than a comparable CPU algorithm when processing large batches of sentences. In terms of end-to-end translation, with decoding on the CPU, we increase throughput by roughly two thirds on a standard MT evaluation dataset. The GPU necessary to achieve these improvements increases the cost of a server by about a third. We believe that GPU-based extraction of hierarchical grammars is an attractive proposition, particularly for MT applications that demand high throughput.

Adaptive Precision Block-Jacobi for High Performance Preconditioning in the Ginkgo Linear Algebra Software

2021 ◽  
Vol 47 (2) ◽  
pp. 1-28
Author(s):  
Goran Flegar ◽  
Hartwig Anzt ◽  
Terry Cojean ◽  
Enrique S. Quintana-Ortí
Keyword(s):  
Linear Algebra ◽  
Graphics Processing Units ◽  
High Performance ◽  
Numerical Algorithms ◽  
Mixed Precision ◽  
Before And After ◽  
Memory Accesses ◽  
Specialized Hardware ◽  
The Individual ◽  
Graphics Processing

The use of mixed precision in numerical algorithms is a promising strategy for accelerating scientific applications. In particular, the adoption of specialized hardware and data formats for low-precision arithmetic in high-end GPUs (graphics processing units) has motivated numerous efforts aiming at carefully reducing the working precision in order to speed up the computations. For algorithms whose performance is bound by the memory bandwidth, the idea of compressing its data before (and after) memory accesses has received considerable attention. One idea is to store an approximate operator–like a preconditioner–in lower than working precision hopefully without impacting the algorithm output. We realize the first high-performance implementation of an adaptive precision block-Jacobi preconditioner which selects the precision format used to store the preconditioner data on-the-fly, taking into account the numerical properties of the individual preconditioner blocks. We implement the adaptive block-Jacobi preconditioner as production-ready functionality in the Ginkgo linear algebra library, considering not only the precision formats that are part of the IEEE standard, but also customized formats which optimize the length of the exponent and significand to the characteristics of the preconditioner blocks. Experiments run on a state-of-the-art GPU accelerator show that our implementation offers attractive runtime savings.

scholarly journals DSPSR: Digital Signal Processing Software for Pulsar Astronomy

2011 ◽  
Vol 28 (1) ◽  
pp. 1-14 ◽  
Author(s):  
W. van Straten ◽  
M. Bailes
Keyword(s):  
Signal Processing ◽  
Digital Signal Processing ◽  
Graphics Processing Units ◽  
High Performance ◽  
Digital Signal ◽  
General Purpose ◽  
Design Decisions ◽  
Extensive Range ◽  
Processing Software ◽  
Graphics Processing

Abstractdspsr is a high-performance, open-source, object-oriented, digital signal processing software library and application suite for use in radio pulsar astronomy. Written primarily in C++, the library implements an extensive range of modular algorithms that can optionally exploit both multiple-core processors and general-purpose graphics processing units. After over a decade of research and development, dspsr is now stable and in widespread use in the community. This paper presents a detailed description of its functionality, justification of major design decisions, analysis of phase-coherent dispersion removal algorithms, and demonstration of performance on some contemporary microprocessor architectures.

scholarly journals A lightweight approach to performance portability with targetDP

2016 ◽  
Vol 32 (2) ◽  
pp. 288-301
Author(s):  
Alan Gray ◽  
Kevin Stratford
Keyword(s):  
Particle Physics ◽  
Message Passing ◽  
Graphics Processing Units ◽  
High Performance ◽  
Large Scale ◽  
Message Passing Interface ◽  
Graphics Processing Unit ◽  
Processing Unit ◽  
Performance Portability ◽  
Graphics Processing

Leading high performance computing systems achieve their status through use of highly parallel devices such as NVIDIA graphics processing units or Intel Xeon Phi many-core CPUs. The concept of performance portability across such architectures, as well as traditional CPUs, is vital for the application programmer. In this paper we describe targetDP, a lightweight abstraction layer which allows grid-based applications to target data parallel hardware in a platform agnostic manner. We demonstrate the effectiveness of our pragmatic approach by presenting performance results for a complex fluid application (with which the model was co-designed), plus separate lattice quantum chromodynamics particle physics code. For each application, a single source code base is seen to achieve portable performance, as assessed within the context of the Roofline model. TargetDP can be combined with Message Passing Interface (MPI) to allow use on systems containing multiple nodes: we demonstrate this through provision of scaling results on traditional and graphics processing unit-accelerated large scale supercomputers.

scholarly journals Extending the usage of graphics processing units on the cloud for cost savings on seismic data regularization

2021 ◽  
Vol 38 (2) ◽  
Author(s):  
Nicholas Torres Okita ◽  
Tiago A. Coimbra ◽  
José Ribeiro ◽  
Martin Tygel
Keyword(s):  
Cloud Computing ◽  
Graphics Processing Units ◽  
Cost Savings ◽  
Data Sets ◽  
Computing Paradigm ◽  
Common Reflection Surface ◽  
User Demand ◽  
Computationally Intensive ◽  
Zero Offset ◽  
Graphics Processing

ABSTRACT. The usage of graphics processing units is already known as an alternative to traditional multi-core CPU processing, offering faster performance in the order of dozens of times in parallel tasks. Another new computing paradigm is cloud computing usage as a replacement to traditional in-house clusters, enabling seemingly unlimited computation power, no maintenance costs, and cutting-edge technology, dynamically on user demand. Previously those two tools were used to accelerate the estimation of Common Reflection Surface (CRS) traveltime parameters, both in zero-offset and finite-offset domain, delivering very satisfactory results with large time savings from GPU devices alongside cost savings on the cloud. This work extends those results by using GPUs on the cloud to accelerate the Offset Continuation Trajectory (OCT) traveltime parameter estimation. The results have shown that the time and cost savings from GPU devices’ usage are even larger than those seen in the CRS results, being up to fifty times faster and sixty times cheaper. This analysis reaffirms that it is possible to save both time and money when using GPU devices on the cloud and concludes that the larger the data sets are and the more computationally intensive the traveltime operators are, we can see larger improvements.Keywords: cloud computing, GPU, seismic processing. Estendendo o uso de placas gráficas na nuvem para economias em regularização de dados sísmicosRESUMO. O uso de aceleradores gráficos para processamento já é uma alternativa conhecida ao uso de CPUs multi-cores, oferecendo um desempenho na ordem de dezenas de vezes mais rápido em tarefas paralelas. Outro novo paradigma de computação é o uso da nuvem computacional como substituta para os tradicionais clusters internos, possibilitando o uso de um poder computacional aparentemente infinito sem custo de manutenção e com tecnologia de ponta, dinamicamente sob demanda de usuário. Anteriormente essas duas ferramentas foram utilizadas para acelerar a estimação de parâmetros do tempo de trânsito de Common Reflection Surface (CRS), tanto em zero-offset quanto em offsets finitos, obtendo resultados satisfatórios com amplas economias tanto de tempo quanto de dinheiro na nuvem. Este trabalho estende os resultados obtidos anteriormente, desta vez utilizando GPUs na nuvem para acelerar a estimação de parâmetros do tempo de trânsito em Offset Continuation Trajectory (OCT). Os resultados obtidos mostraram que as economias de tempo e dinheiro foram ainda maiores do que aquelas obtidas no CRS, sendo até cinquenta vezes mais rápido e sessenta vezes mais barato. Esta análise reafirma que é possível economizar tanto tempo quanto dinheiro usando GPUs na nuvem, e conclui que quanto maior for o dado e quanto mais computacionalmente intenso for o operador, maiores serão os ganhos de desempenho observados e economias.Palavras-chave: computação em nuvem, GPU, processamento sísmico. 

scholarly journals Big Data and IT Network Data Visualization

2018 ◽  
Vol 3 (1) ◽  
pp. 9-16 ◽  
Author(s):  
Lidong Wang
Keyword(s):  
Big Data ◽  
Network Analysis ◽  
Graphics Processing Units ◽  
Data Analytics ◽  
High Performance ◽  
Big Data Analytics ◽  
Network Visualization ◽  
Network Data ◽  
Graphics Processing ◽  
Performance Computing

Visualization with graphs is popular in the data analysis of Information Technology (IT) networks or computer networks. An IT network is often modelled as a graph with hosts being nodes and traffic being flows on many edges. General visualization methods are introduced in this paper. Applications and technology progress of visualization in IT network analysis and big data in IT network visualization are presented. The challenges of visualization and Big Data analytics in IT network visualization are also discussed. Big Data analytics with High Performance Computing (HPC) techniques, especially Graphics Processing Units (GPUs) helps accelerate IT network analysis and visualization.

scholarly journals The VOLNA-OP2 tsunami code (version 1.5)

2018 ◽  
Vol 11 (11) ◽  
pp. 4621-4635 ◽  
Author(s):  
Istvan Z. Reguly ◽  
Daniel Giles ◽  
Devaraj Gopinathan ◽  
Laure Quivy ◽  
Joakim H. Beck ◽  
...  
Keyword(s):  
Graphics Processing Units ◽  
High Performance ◽  
Shallow Water Equation ◽  
Xeon Phi ◽  
Intel Xeon Phi ◽  
Central Processing ◽  
Domain Specific ◽  
Computing Platforms ◽  
Graphics Processing ◽  
Intel Xeon

Abstract. In this paper, we present the VOLNA-OP2 tsunami model and implementation; a finite-volume non-linear shallow-water equation (NSWE) solver built on the OP2 domain-specific language (DSL) for unstructured mesh computations. VOLNA-OP2 is unique among tsunami solvers in its support for several high-performance computing platforms: central processing units (CPUs), the Intel Xeon Phi, and graphics processing units (GPUs). This is achieved in a way that the scientific code is kept separate from various parallel implementations, enabling easy maintainability. It has already been used in production for several years; here we discuss how it can be integrated into various workflows, such as a statistical emulator. The scalability of the code is demonstrated on three supercomputers, built with classical Xeon CPUs, the Intel Xeon Phi, and NVIDIA P100 GPUs. VOLNA-OP2 shows an ability to deliver productivity as well as performance and portability to its users across a number of platforms.

scholarly journals Accelerated FDPS: Algorithms to use accelerators with FDPS

2020 ◽  
Vol 72 (1) ◽  
Author(s):  
Masaki Iwasawa ◽  
Daisuke Namekata ◽  
Keigo Nitadori ◽  
Kentaro Nomura ◽  
Long Wang ◽  
...  
Keyword(s):  
Graphics Processing Units ◽  
High Performance ◽  
General Purpose ◽  
Performance Model ◽  
Performance Tuning ◽  
Data Types ◽  
Interaction Function ◽  
Current Implementation ◽  
And Performance ◽  
Graphics Processing

Abstract We describe algorithms implemented in FDPS (Framework for Developing Particle Simulators) to make efficient use of accelerator hardware such as GPGPUs (general-purpose computing on graphics processing units). We have developed FDPS to make it possible for researchers to develop their own high-performance parallel particle-based simulation programs without spending large amounts of time on parallelization and performance tuning. FDPS provides a high-performance implementation of parallel algorithms for particle-based simulations in a “generic” form, so that researchers can define their own particle data structure and interparticle interaction functions. FDPS compiled with user-supplied data types and interaction functions provides all the necessary functions for parallelization, and researchers can thus write their programs as though they are writing simple non-parallel code. It has previously been possible to use accelerators with FDPS by writing an interaction function that uses the accelerator. However, the efficiency was limited by the latency and bandwidth of communication between the CPU and the accelerator, and also by the mismatch between the available degree of parallelism of the interaction function and that of the hardware parallelism. We have modified the interface of the user-provided interaction functions so that accelerators are more efficiently used. We also implemented new techniques which reduce the amount of work on the CPU side and the amount of communication between CPU and accelerators. We have measured the performance of N-body simulations on a system with an NVIDIA Volta GPGPU using FDPS and the achieved performance is around 27% of the theoretical peak limit. We have constructed a detailed performance model, and found that the current implementation can achieve good performance on systems with much smaller memory and communication bandwidth. Thus, our implementation will be applicable to future generations of accelerator system.

Sign in / Sign up

Export Citation Format

Share Document