Tech News
Back to home
Technology

Silicon Photonics Lights The Way To More Efficient Data Centers

Source

SemiEngineering

Published

TL;DR

AI Generated

Silicon photonics is paving the way for more efficient data centers by potentially increasing bandwidth density and reducing power consumption, particularly driven by AI workloads. However, challenges such as process compatibility, thermal issues, and mechanical stress arise due to the use of various materials in photonic interconnects. Integrated electro-optical I/O modules are the desired outcome, but design and process complexities need to be addressed. The article delves into the technical aspects of silicon photonics, including the components involved, such as light sources, modulators, waveguides, and photodetectors, as well as the challenges of integrating optical and electronic components for efficient data transmission.

Read Full Article

Similar Articles

New Jersey datacenter expansion got $77m in tax breaks to create exactly one permanent job — JPMorgan's site already scored $35m and currently employs just 25 workers

New Jersey datacenter expansion got $77m in tax breaks to create exactly one permanent job — JPMorgan's site already scored $35m and currently employs just 25 workers

A JPMorgan datacenter expansion in New Jersey received $77 million in tax breaks to create just one permanent job, marking a high subsidy-to-jobs ratio. The expansion, which is part of a $1 billion investment, currently employs only 25 workers despite the substantial tax incentives. The Industrial Development Agency (IDA) believes the deal will benefit the local economy, projecting a $100 million infusion. While the IDA acknowledges the lack of permanent jobs, they argue that the overall economic impact justifies the subsidy. The situation highlights the need for closer scrutiny of datacenter subsidies and their true economic impact.

Tom's Hardware•
Half of planned US data center builds have been delayed or canceled, growth limited by shortages of power infrastructure and parts from China — the AI build-out flips the breakers

Half of planned US data center builds have been delayed or canceled, growth limited by shortages of power infrastructure and parts from China — the AI build-out flips the breakers

Shortages of power infrastructure components and parts from China have led to delays and cancellations of about half of planned U.S. data center builds. The trade war between the U.S. and China has forced server makers out of China, impacting the availability of electrical equipment crucial for data center construction. Major tech companies like Alphabet, Amazon, Meta, and Microsoft are expected to invest heavily in expanding AI capacity, but constraints in power infrastructure components are hindering progress. The delays in obtaining key electrical components like transformers and switchgear are causing setbacks in deploying AI data centers, with lead times for high-power transformers stretching up to five years. Companies are turning to global markets to address shortages, with Canada, Mexico, and South Korea emerging as major suppliers for high-power transformers.

Tom's Hardware•
Nuclear Power and Design Automation

Nuclear Power and Design Automation

The article discusses the growing interest in nuclear power as an additional source for power generation, particularly due to the increasing demand from large data centers. Notably, investments from companies like Microsoft and individuals like Sam Altman indicate a push to expedite progress in nuclear power beyond the slower pace of traditional utilities. The author, who has expertise in this area, explores the intersection of nuclear power and design automation.

SemiWiki•
Tech titans team up to form optical interconnect alliance to solve the AI buildout's big data bottleneck — Nvidia, AMD, Broadcom & more set sights on building PHY to break through the limitations of copper

Tech titans team up to form optical interconnect alliance to solve the AI buildout's big data bottleneck — Nvidia, AMD, Broadcom & more set sights on building PHY to break through the limitations of copper

Tech giants like Nvidia, AMD, Broadcom, Microsoft, and others have formed the Optical Compute Interconnect Multi-Source Agreement (OCI MSA) group to develop an open connectivity specification for optical interconnections in AI data centers. The goal is to overcome the limitations of copper by enabling higher data transfer speeds and reducing power consumption through optical interconnections. The group aims to create a physical layer (PHY) capable of delivering speeds up to 3.2Tb/s to support large-scale AI workloads. The standardization of optical interconnects would allow for multi-vendor supply chains and alleviate supply chain constraints faced by copper. The alliance seeks to establish a platform-agnostic approach to optical interconnections to drive innovation in AI data center infrastructure.

Tom's Hardware•

We use cookies

We use cookies to ensure you get the best experience on our website. For more information on how we use cookies, please see our cookie policy.