Optically Connected Memory for Disaggregated Data Centers
Jorge Gonzalez, Alexander Gazman, Maarten Hattink, Mauricio G. Palma,, Meisam Bahadori, Ruth Rubio-Noriega, Lois Orosa, Madeleine Glick, Onur Mutlu,, Keren Bergman, Rodolfo Azevedo

TL;DR
This paper introduces an Optically Connected Memory architecture utilizing integrated photonics to disaggregate memory from compute nodes, achieving high bandwidth and low energy consumption in data centers.
Contribution
It proposes a novel photonic-based memory disaggregation architecture using micro-ring resonators that requires no modifications to existing DRAM modules.
Findings
Interconnects four DDR4 channels with two fibers at 1.07 pJ/bit energy
Performs up to 5.5x faster than traditional disaggregated memory
Uses real photonic device data integrated into system simulation
Abstract
Recent advances in integrated photonics enable the implementation of reconfigurable, high-bandwidth, and low energy-per-bit interconnects in next-generation data centers. We propose and evaluate an Optically Connected Memory (OCM) architecture that disaggregates the main memory from the computation nodes in data centers. OCM is based on micro-ring resonators (MRRs), and it does not require any modification to the DRAM memory modules. We calculate energy consumption from real photonic devices and integrate them into a system simulator to evaluate performance. Our results show that (1) OCM is capable of interconnecting four DDR4 memory channels to a computing node using two fibers with 1.07 pJ energy-per-bit consumption and (2) OCM performs up to 5.5x faster than a disaggregated memory with 40G PCIe NIC connectors to computing nodes.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
