Evaluating the Arm Ecosystem for High Performance Computing

Jackson, Adrian; Turner, Andrew; Weiland, Michele; Johnson, Nick; Perks, Olly; Parsons, Mark

doi:10.1145/3324989.3325722

Cited by 20 publications

(8 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The first step is configuring the topology because, without proper network connectivity, the experiment process will not run as intended. Figures 5(a The evaluation process was carried out using a benchmark with high-performance computing systems [24], [28], [29]. The model's evaluation is an experiment with a total of four Raspberry Pi boards and one computer.…”

Section: Methodsmentioning

confidence: 99%

A new approach of scalable traffic capture model with Pi cluster

Hartomo

Daru

Purnomo

2023

IJECE

View full text Add to dashboard Cite

<span lang="EN-US">The development of the internet of things (IoT), which functions as servers, device monitors, and controllers of several peripherals inside the smart home, eased workload in many sectors. Most devices are accessible through the internet because they communicate with wired or wireless interfaces. However, this feature makes them prone to the risk of being exposed to the public. The exposed devices are an easy target for the third party to launch a flooding attack through the network. This attack overloads the system due to the low processing capability, thereby interrupting any running process and harming the device. Therefore, this study proposed a scalable network capturing model that utilized multiple Raspberry Pi boards in parallel to monitor the network traffics simultaneously. An isolated experiment was used for evaluation by running simultaneous flooding attacks on each device. The result showed that the model consumed 30.44% more memory with 14.66% lower central processing unit (CPU) usage and 3.63% faster execution time. This means that this model is better in terms of performance and effectiveness than the single capture model.</span>

show abstract

Section: Methodsmentioning

confidence: 99%

A new approach of scalable traffic capture model with Pi cluster

Hartomo

Daru

Purnomo

2023

IJECE

View full text Add to dashboard Cite

show abstract

“…S. McIntosh-Smith et al [18,19] were among the first to measure performance on mainstream Arm ® HPC systems. Later, Jackson et al [20] investigated the performance of distributed memory communications (MPI) through a benchmark suite utilizing MPI on ThunderX2. The energy consumption of these Arm ® processors has been investigated by the Mont-Blanc project [21].…”

Section: Related Workmentioning

confidence: 99%

Deploying a Task-based Runtime System on Raspberry Pi Clusters

Gupta,

Brandt,

Wagle

et al. 2020

Preprint

View full text Add to dashboard Cite

Arm ® technology is becoming increasingly important in HPC. Recently, Fugaku, an Arm ® -based system, was awarded the number one place in the Top500 list. Raspberry Pis provide an inexpensive platform to become familiar with this architecture. However, Pis can also be useful on their own. Here we describe our efforts to configure and benchmark the use of a Raspberry Pi cluster with the HPX/Phylanx platform (normally intended for use with HPC applications) and document the lessons we learned. First, we highlight the required changes in the configuration of the Pi to gain performance. Second, we explore how limited memory bandwidth limits the use of all cores in our shared memory benchmarks. Third, we evaluate whether low network bandwidth affects distributed performance. Fourth, we discuss the power consumption and the resulting trade-off in cost of operation and performance.

show abstract

“…The design includes eight DDR4 memory channels to deliver measured STREAM triad memory bandwidth in excess of 240 GB/s per dual-socket node, significantly greater than the comparable Intel processors available at the time. Evaluation of scientific applications on that platform have been documented [4] [5], demonstrating comparable performance for a range of applications when compared to similar Intel processors. This paper extends this work by adding the novel A64FX processor architecture, and expanding the range of applications, as well as the systems, benchmarked.…”

Section: Related Workmentioning

confidence: 99%

Investigating Applications on the A64FX

Jackson¹,

Weiland²,

Brown³

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

The A64FX processor from Fujitsu, being designed for computational simulation and machine learning applications, has the potential for unprecedented performance in HPC systems. In this paper, we evaluate the A64FX by benchmarking against a range of production HPC platforms that cover a number of processor technologies. We investigate the performance of complex scientific applications across multiple nodes, as well as single node and mini-kernel benchmarks. This paper finds that the performance of the A64FX processor across our chosen benchmarks often significantly exceeds other platforms, even without specific application optimisations for the processor instruction set or hardware. However, this is not true for all the benchmarks we have undertaken. Furthermore, the specific configuration of applications can have an impact on the runtime and performance experienced.

show abstract

Evaluating the Arm Ecosystem for High Performance Computing

Cited by 20 publications

References 12 publications

A new approach of scalable traffic capture model with Pi cluster

A new approach of scalable traffic capture model with Pi cluster

Deploying a Task-based Runtime System on Raspberry Pi Clusters

Investigating Applications on the A64FX

Contact Info

Product

Resources

About