Author: qinbafrank

Let's discuss the logic behind being optimistic about optical interconnects. In mid and late July, we talked about optimistic sectors in the US stock market, which have already exploded (CSP, software, optical interconnects). The optimistic logic for CSP/software was discussed thoroughly in a long article in mid-July (Related reading: AI adoption enters the era of engineering, new opportunities and challenges for ultra-large CSP, new evaluation criteria for AI commercialization process). Today, let's talk about why we are optimistic about optical interconnects? What are the catalysts?

Looking back, the optical interconnect sector has been the earliest and longest-adjusted sector in this wave of semiconductor adjustments. From a personal perspective, in mid-May, Bernstein's report stated that CPO deployment would not be so quick, and this year, we still observe optical modules and connections. After that, the optical sector began to adjust, with only connections performing well in the US stock market.

Recently, I see two catalysts:

1) “NVIDIA's Senior Vice President Gilad Shainer announced at a recent technology forum that CPO has begun mass production. Switches developed in collaboration with its supply chain have been delivered to close customers and are being deployed in NVIDIA's own facilities. A large number of switches adopting CPO technology are expected to be introduced to global AI factories this year;

2) Lumentum CEO Michael Hurlston stated that the shortage of AI indium phosphide lasers will be more severe than the memory crisis.

How to interpret this logic:

1) The delivery of CPO means that the market should anticipate that with mass production and delivery, CPO should gradually ramp up, and business realization will begin to materialize;

2) The shortage of indium phosphide and the market's memory of shortages still exist;

3) The early and prolonged adjustment of optics means that since semiconductors have concentrated on storage recently, there has been very little discussion about optical interconnects, which indicates an expectation gap;

4) The financial reports of ultra-large CSP preliminarily prove capital expenditure ROIC, and the construction of data centers will be rapid. The demand for wide-area cluster interconnects and large-scale data center interconnects is indeed further exploding.

In summary:

The demand for AI data center optical interconnects is real and accelerating, and the entire industry chain has entered the "supply bottleneck" stage from "demand verification"; the official mass production of NVIDIA CPO is a historic signal, coexisting in the short term (at least until 2027) with pluggable optical modules, with the final bottleneck locked on the capacity of indium phosphide (InP) laser chips.

Technology roadmap timeline:

1) Scale-up (inside rack): Primarily copper cables/pluggable.

2) Scale-out (cross-rack/cross-data center): CPO first penetrates (power consumption from 30W→9W/port, significantly increased bandwidth).

3) Coexistence window: Pluggable dominates the next 18-24 months; CPO and pluggable are non-zero sum, at least until 2027.

"Demand has been validated across the entire chain, and the bottleneck lies in InP lasers; whoever can self-produce/expand the fastest and has the safest supply chain will reap the greatest flexibility."

The next two weeks will see a concentration of financial reports from companies in the optical interconnect field, and if these reports are further validated, it could be a catalytic window.

免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。