Optically Connected Memory for Disaggregated Data Centers

Author(s):Show less -

Gonzalez, Jorge ; Gazman, Alexander ; Hattink, Maarten ; Palma, Mauricio G. ; Bahadori, Meisam ; Rubio-Noriega, Ruth ; Orosa, Lois ; Glick, Madeleine ; Mutlu, Onur ; Bergman, Keren ; Azevedo, Rodolfo ; IEEE

• Document type: Journal article
• Source: 2020 IEEE 32ND INTERNATIONAL SYMPOSIUM ON COMPUTER ARCHITECTURE AND HIGH PERFORMANCE COMPUTING (SBAC-PAD 2020); v. N/A, p. 8-pg., 2020-01-01.

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 energyper-bit consumption and (2) OCM performs up to 5.5x faster than a disaggregated memory with 40G PCIe NIC connectors to computing nodes. (AU)

• FAPESP's process: 13/08293-7 - CCES - Center for Computational Engineering and Sciences
• Grantee: Munir Salomao Skaf
• Support Opportunities: Research Grants - Research, Innovation and Dissemination Centers - RIDC
• FAPESP's process: 14/01642-9 - OMPSoC: Heterogeneous Optical Multiprocessor System on Chip Modeling
• Grantee: Jorge Luis Gonzalez Reano
• Support Opportunities: Scholarships in Brazil - Doctorate