Exploring the benefits of using co-packaged optics in data center and AI supercomputer networks: a simulation-based analysis [Invited]
Authors:
- 8 January 2024
- journal article
- research article
- Published by Optica Publishing Group in Journal of Optical Communications and Networking
- Vol. 16 (2) , A143-A156
Abstract
We investigate the advantages of using co-packaged optics in next-generation data center and AI supercomputer networks. The increased escape bandwidth offered by co-packaged optics provides multiple possibilities for building 50T switches and beyond, expanding the opportunities in both the data center and supercomputing domains. This provides network architects with the opportunity to expand their design space and develop simplified networks with enhanced network locality properties. Co-packaging at the switch and server points enables networks with double capacity while reducing the switch count by 64% compared to state-of-the-art systems. We evaluate these concepts through discrete-event simulations using all-to-all and all-reduce traffic patterns that simulate collective communications commonly found in network-bound applications. Initially, we investigate the all-to-all overhead involved in distributing the virtual machines of the applications across multiple leaf switches and compare it to the scenario in which all VMs are placed under a single switch. Subsequently, we evaluate the performance of an AI supercomputing cluster by simulating both patterns for different message sizes, while also varying the number of participating nodes. The results suggest that networks with improved locality properties become increasingly important as the network stack operates at higher speeds; for a stack latency of 1.25 µs, placing the applications under multiple switches can result in up to 68% higher completion times than placing them under a single switch. For AI supercomputers, significant improvements are observed in the mean server throughput, reaching more than 90% for configurations involving 256 nodes and message sizes of at least 128 KiB.Keywords
Funding Information
- Advanced Research Projects Agency - Energy (DE-AR0000846)
- U.S. Department of Energy
This publication has 27 references indexed in Scilit:
- Co-packaged optics for HPC and data center networks2021
- Toward lower-diameter large-scale HPC and data center networks with co-packaged opticsJournal of Optical Communications and Networking, 2020
- 1.6 Tbps Silicon Photonics Integrated Circuit and 800 Gbps Photonic Engine for Switch Co-Packaging DemonstrationJournal of Lightwave Technology, 2020
- Is Network the Bottleneck of Distributed Training?2020
- TeraPHY: A Chiplet Technology for Low-Power, High-Bandwidth In-Package Optical I/OIEEE Micro, 2020
- High Speed VCSELs and Co-Packaging for Short Reach Communication within Cloud and High Performance Computing2019
- CloudSim Plus: A cloud computing simulation framework pursuing software engineering principles for improved modularity, extensibility and correctness2017
- Bandwidth-optimal all-to-all exchanges in fat tree networks2013
- Solving the corner-turning problem for large interferometersMonthly Notices of the Royal Astronomical Society, 2010
- Input Versus Output Queueing on a Space-Division Packet SwitchIEEE Transactions on Communications, 1987