We investigate the viability of optically switched network for ML accelerator clusters and compare it to a leaf-spine network with 256/1024 GPUs. Results show almost ideal throughput, sub-μs latency and zero packet-loss for ¡0.6 traffic-load.</p
This paper develops a performance model of an optically interconnected parallel computer system oper...
We present an overview of the application of machine learning for traffic engineering and network op...
This paper proposes a machine-learning (ML)-aided cognitive approach for effective bandwidth reconfi...
We investigate the viability of optically switched network for ML accelerator clusters and compare i...
Following trends that emphasize neural networks for machine learning, many studies regarding computi...
Integration of the machine learning (ML) technique in all-optical networks can enhance the effective...
Distributed deep learning (DDL) systems strongly depend on network performance. Current electronic p...
We numerically assess a novel low-latency HPC network based on fast optical switches (HiFOS) under m...
In this paper, we assess the performance of techniques for optical burst switching (OBS) designed f...
Deep learning has risen to prominence in fields from medicine to autonomous vehicles. This rise has ...
Deep learning has been revolutionizing many aspects of our society, powering various fields includin...
Recent success in deep neural networks has generated strong interest in hardware accelerators to imp...
This paper examines the performance of distributed-shared-memory systems based on the Simultaneous O...
Optical networks based on fast optical switches (FOSes) could potentially solve the latency, bandwid...
In this paper, we assess the performance of techniques for optical burst switching (OBS) designed fo...
This paper develops a performance model of an optically interconnected parallel computer system oper...
We present an overview of the application of machine learning for traffic engineering and network op...
This paper proposes a machine-learning (ML)-aided cognitive approach for effective bandwidth reconfi...
We investigate the viability of optically switched network for ML accelerator clusters and compare i...
Following trends that emphasize neural networks for machine learning, many studies regarding computi...
Integration of the machine learning (ML) technique in all-optical networks can enhance the effective...
Distributed deep learning (DDL) systems strongly depend on network performance. Current electronic p...
We numerically assess a novel low-latency HPC network based on fast optical switches (HiFOS) under m...
In this paper, we assess the performance of techniques for optical burst switching (OBS) designed f...
Deep learning has risen to prominence in fields from medicine to autonomous vehicles. This rise has ...
Deep learning has been revolutionizing many aspects of our society, powering various fields includin...
Recent success in deep neural networks has generated strong interest in hardware accelerators to imp...
This paper examines the performance of distributed-shared-memory systems based on the Simultaneous O...
Optical networks based on fast optical switches (FOSes) could potentially solve the latency, bandwid...
In this paper, we assess the performance of techniques for optical burst switching (OBS) designed fo...
This paper develops a performance model of an optically interconnected parallel computer system oper...
We present an overview of the application of machine learning for traffic engineering and network op...
This paper proposes a machine-learning (ML)-aided cognitive approach for effective bandwidth reconfi...
DOI
Add to ORKG