Jeff As you know I have the marks on my back on that issue more than anyone. John From: Maki, Jeffery <jeffery.maki@xxxxxxx>
Sent: Wednesday, July 1, 2026 4:12 PM
To: jdambrosia@xxxxxxxxx; STDS-802-3-400GPL@xxxxxxxxxxxxxxxxx
Subject: RE: [802.3_400GPL] [400GPL] Website updates John, Chris said introduced as in proposed. PSM4 failed to be adopted as a solution in IEEE 802.3 until 802.3bs as already noted. I see the point of this to be that we really need to understand the market and how it is changing. Do people recall when it was thought the Ethernet rate should only go up by factors of 10, and 40G Ethernet was the big controversy. Jeffery Maki (he/him/his) Senior Distinguished Technologist, Optics jeffery.maki@xxxxxxx +1 408-936-8575 Hewlett Packard Enterprise HPE.com Chris I just reviewed the 802.3bm specification. Here are the PMDs defined – - 85. Physical Medium Dependent sublayer and baseband medium, type 40GBASE-CR4 and 100GBASE-CR10
- 86. Physical Medium Dependent (PMD) sublayer and medium, type 40GBASE–SR4 and 100GBASE–SR10
- 87. Physical Medium Dependent (PMD) sublayer and medium, type 40GBASE–LR4 and 40GBASE–ER4
- 88. Physical Medium Dependent (PMD) sublayer and medium, type 100GBASE–LR4 and 100GBASE–ER4
- 89. Physical Medium Dependent (PMD) sublayer and medium, type 40GBASE-FR
- 92. Physical Medium Dependent (PMD) sublayer and baseband medium, type 100GBASE-CR4
- 93. Physical Medium Dependent (PMD) sublayer and baseband medium, type 100GBASE-KR4
- 95. Physical Medium Dependent (PMD) sublayer and medium, type 100GBASE-SR4
None of these appear to be a PSM solution. So maybe somewhere else between bm and bs. But that is for another day. Have a happy holiday. John PSM4 was first introduced in bm, as were other solutions for SMF 100G point-to-point. At the time, our interpretation of Broad Market Potential was that there can only be one solution for SMF. Because there was genuine justification for multiple ones, bm did not adopt any new one. This resulted in 100G PSM4 and 100G CWDM4 standardization in MSAs right after the project. The bs project recognized need for multiple solutions, which established the paradigm we have today of DRn, FRn, and LRn. Break-out application was recognized as a possible use of DRn, although not explicitly specified.
400GBASE-DR4 for PSM4 was defined in 802.3bs- not the 100G project as noted below. 500m was selected at the time due to concerns on closing the budget for 2km. i’m glad to see the notion of breakout included in the 802.3dj standard. It's interesting to reflect how on how far DRn has come, from it's introduction as PSM4 in the 100G SMF project, where it was one of many competing point-to-point proposals to today where it's the defacto datacenter optic, although primarily as break-out.
Chair, IEEE 802.3 400Gbps/lane Signaling Study Group
To unsubscribe from the STDS-802-3-400GPL list, click the following link: https://listserv.ieee.org/cgi-bin/wa?SUBED1=STDS-802-3-400GPL&A=1
To unsubscribe from the STDS-802-3-400GPL list, click the following link: https://listserv.ieee.org/cgi-bin/wa?SUBED1=STDS-802-3-400GPL&A=1
To unsubscribe from the STDS-802-3-400GPL list, click the following link: https://listserv.ieee.org/cgi-bin/wa?SUBED1=STDS-802-3-400GPL&A=1
To unsubscribe from the STDS-802-3-400GPL list, click the following link: https://listserv.ieee.org/cgi-bin/wa?SUBED1=STDS-802-3-400GPL&A=1 |