Verification of PCIe over Electronic and Photonic I/Fs

High-speed computing (HPC) in data centers and for AI/ML computation relies largely on the high bandwidth and low latency of the interconnects between CPU, GPU, memory and storage. The high-speed serial computer expansion bus standard, peripheral component interconnect express (PCIe) 1.0 has been implemented since 2003 in CPUs and SoCs, including chip-to-chip through electronic interface (I/F) such as SerDes. Started at 2.5 GT/s, PCIe has been applied, from electronic I/F, over optic fiber or optics module. As released in 2022, PCIe 6.0 has become the latest standard to be studied over the photonic I/F. Meanwhile, the compute express link (CXL) was developed in 2019, we have initiated some study, 1) Using Alveo U50 FPGA platform for system verification of PCIe 4.0 supporting CCIX (now part of CXL) over electronic I/Fs; 2) Using Agilex 7 FPGA on high-speed Ethernet 400GbE data-rate through CMOS I/F SerDes; and 3) Based on I/F SerDes to fiber (E2O), and/or optic I/F to loopback to SerDes (O2E), the work was continued on CPO I/F. Current study aims at PCIe 5.0, at 32 GT/s or maximum 128 GB/s, over the Root Complex (host side) and the component side (endpoint side through on-board optic (OBO), to determine the system performance. The setup program for the electronic and photonic interconnect, system verification and results are presented.

Authors

Institutions

Publication Details

Published
2025-03-24
DOI
https://doi.org/10.1109/cstic64481.2025.11018059
Primary Topic
Integrated Circuits and Semiconductor Failure Analysis
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Verification of PCIe over Electronic and Photonic I/Fs

Chun-Zhang Chen, Xuhui Liu, Xiaoli Fang, Quan Pan et al.
Integrated Circuits and Semiconductor Failure Analysis
article

Verification of PCIe over Electronic and Photonic I/Fs

Chun-Zhang Chen, Xuhui Liu, Xiaoli Fang, Quan Pan, Hanming Wu, Liang Wang
article en

Abstract

High-speed computing (HPC) in data centers and for AI/ML computation relies largely on the high bandwidth and low latency of the interconnects between CPU, GPU, memory and storage. The high-speed serial computer expansion bus standard, peripheral component interconnect express (PCIe) 1.0 has been implemented since 2003 in CPUs and SoCs, including chip-to-chip through electronic interface (I/F) such as SerDes. Started at 2.5 GT/s, PCIe has been applied, from electronic I/F, over optic fiber or optics module. As released in 2022, PCIe 6.0 has become the latest standard to be studied over the photonic I/F. Meanwhile, the compute express link (CXL) was developed in 2019, we have initiated some study, 1) Using Alveo U50 FPGA platform for system verification of PCIe 4.0 supporting CCIX (now part of CXL) over electronic I/Fs; 2) Using Agilex 7 FPGA on high-speed Ethernet 400GbE data-rate through CMOS I/F SerDes; and 3) Based on I/F SerDes to fiber (E2O), and/or optic I/F to loopback to SerDes (O2E), the work was continued on CPO I/F. Current study aims at PCIe 5.0, at 32 GT/s or maximum 128 GB/s, over the Root Complex (host side) and the component side (endpoint side through on-board optic (OBO), to determine the system performance. The setup program for the electronic and photonic interconnect, system verification and results are presented.

Peng Cheng Laboratory (CN)
Openalex Percentile: Top 90%
Integrated Circuits and Semiconductor Failure Analysis
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Verification of PCIe over Electronic and Photonic I/Fs — Chun-Zhang Chen, Xuhui Liu, et al. · (2025) | TGRS Research Map | TGRS