´óÐãÖ±²¥

Archive for the 'Coherent DSP' Category

• February 23, 2026

  

  

  

  The Golden Cable Initiative: Enabling the Cable Partner Ecosystem at Hyperscale Speed

  By Michael Arsenault, Director of Product Marketing for AEC DSPs, ´óÐãÖ±²¥

  Rack connectivity is undergoing a historic transformation. Data center operators are demanding both scale-up and scale-out connectivity that can move more data across longer distances and between more systems, while delivering unprecedented levels of energy efficiency and reliability.

  To help cable providers and their customers meet these challenges, ´óÐãÖ±²¥ has launched the?Golden Cable initiative, designed to accelerate the development of?active electrical cables (AECs). AECs are a rapidly growing class of high-bandwidth, enhanced copper interconnects used to link servers, switches, NICs and other?assets in the same rack or across adjacent racks (about two to nine meters).

  The Golden Cable initiative delivers a validated cable architecture tested across leading platforms and built on industry-leading software, reference designs, technical data, firmware and comprehensive support. Participants can combine these assets with their own technology to develop unique AECs powered by?DSPs, optimized for specific customer requirements and use cases.?

  To further enhance performance and ensure broad compatibility, Golden Cable AECs are rigorously tested in the ´óÐãÖ±²¥ Cloud Interoperability Lab. Here, cables are validated across a wide range of customized configuration scenarios involving leading XPUs, CPUs, NICs, servers, switches, optical modules and other critical infrastructure components. This process enables ´óÐãÖ±²¥ and its partners to validate AEC firmware before cables reach end-customers, significantly accelerating customer qualification and deployment timelines. The result is greater confidence from the first plug-in.

  The Golden Cable initiative is designed to rapidly scale and empower the cable partner ecosystem, enabling ´óÐãÖ±²¥ to meet accelerating market demand at true hyperscale speed. By operating in close alignment with key partners, ´óÐãÖ±²¥ is achieving many of the benefits of near?vertical integration, while maintaining the flexibility and scalability of a partner?driven model.

  Read More
• June 06, 2025

  

  

  

  Welcome to the ¡°OFC Film Festival¡±

  By Kirt Zimmer, Head of Social Media Marketing, ´óÐãÖ±²¥

  The OFC 2025 event in San Francisco was so vast that it would be easy to miss a few stellar demos from your favorite optical networking companies. That¡¯s why we took the time to create videos featuring the latest ´óÐãÖ±²¥ technology.

  Put them all together and you have a wonderful film festival for technophiles. Enjoy!

  PCIe + AEC and Optical : A new Era for Connectivity

  Annie Liao ¡ª Product Management Director, Connectivity Marketing at ´óÐãÖ±²¥ ¡ª showcased how PCIe Gen6 signals can be converted to optical¡ªextending trace length beyond traditional electrical limitations. The result? A 10-meter cable reach spanning across multiple racks. Whether connecting GPUs, accelerators, or storage across racks, this kind of extended PCIe connectivity is critical for building the infrastructure that powers advanced AI workloads.

  Read More
• February 09, 2025

  

  

  

  Ten Statistical Snapshots to Better Understand AI, Data Centers and Energy

  By Michael Kanellos, Head of Influencer Relations, ´óÐãÖ±²¥

  You¡¯re likely assaulted daily with some zany and unverifiable AI factoid. By 2027, 93% of AI systems will be able to pass the bar, but limit their practice to simple slip and fall cases! Next-generation training models will consume more energy than all Panera outlets combined!? etc. etc.

  What can you trust? The stats below. Scouring the internet (and leaning heavily on 16 years of employment in the energy industry) I¡¯ve compiled a list of somewhat credible and relevant stats that provide perspective to the energy challenge.

  1. First, the Concerning News: Data Center Demand Could Nearly Triple in a Few Years

  ¹ has issued its latest data center power report and it¡¯s ominous.

  Data center power consumption rose from a stable 60-76 terawatt hours (TWh) per year in the U.S. through 2018 to 176 TWh in 2023, or from 1.9% of total power consumption to 4.4%. By 2028, AI could push it to 6.7%-12%. (Lighting consumes 15%².)?

  

  Report co-author Eric Masanet adds that the total doesn¡¯t include bitcoin, which increases 2023¡¯s consumption by 70 TWh. Add a similar 30-40% to subsequent years too if you want.

  Read More
• February 03, 2025

  

  

  

  The Custom Era of Chips

  By Raghib Hussain, President, Products and Technologies

  This article was originally published in .
  ?

  Artificial intelligence is about to face some serious growing pains.

  Demand for AI services is exploding globally. Unfortunately, so is the challenge of delivering those services in an economical and sustainable manner. AI power demand is forecast to grow by 44.7% annually, a surge that will double data center power consumption to 857 terawatt hours in 2028¹: as a nation today, that would make data centers the sixth largest consumer of electricity, right behind Japan¡¯s² consumption. It¡¯s an imbalance that threatens the ¡°smaller, cheaper, faster¡± mantra that has driven every major trend in technology for the last 50 years.

  It also doesn¡¯t have to happen. Custom silicon¡ªunique silicon optimized for specific use cases¡ªis already demonstrating how we can continue to increase performance while cutting power even as Moore¡¯s Law fades into history. Custom may account for 25% of AI accelerators (XPUs) by 2028³ and that¡¯s just one category of chips going custom.

  The Data Infrastructure is the Computer

  Jensen Huang¡¯s vision for AI factories is apt. These coming AI data centers will churn at an unrelenting pace 24/7. And, like manufacturing facilities, their ultimate success or failure for service providers will be determined by operational excellence, the two-word phrase that rules manufacturing. Are we consuming more, or less, energy per token than our competitor? Why is mean time to failure rising? What¡¯s the current operational equipment effectiveness (OEE)? In oil and chemicals, the end products sold to customers are indistinguishable commodities. Where they differ is in process design, leveraging distinct combinations of technologies to squeeze out marginal gains.

  The same will occur in AI. Cloud operators already are engaged in differentiating their backbone facilities. Some have adopted optical switching to reduce energy and latency. Others have been more aggressive at developing their own custom CPUs. In 2010, the main difference between a million-square-foot hyperscale data center and a data center inside a regional office was size. Both were built around the same core storage devices, servers and switches. Going forward, diversity will rule, and the operators with the lowest cost, least downtime and ability to roll out new differentiating services and applications will become the favorite of businesses and consumers.

  The best infrastructure, in short, will win.

  The Custom Concept

  And the chief way to differentiate infrastructure will be through custom infrastructure that are enabled by custom semiconductors, i.e., chips containing unique IP or features for achieving leapfrog performance for an application. It¡¯s a spectrum ranging from AI accelerators built around distinct, singular design to a merchant chip containing additional custom IP, cores and firmware to optimize it for a particular software environment. While the focus is now primarily on higher value chips such as AI accelerators, every chip will get customized: Meta, for example, recently unveiled a custom NIC, a relatively unsung chip that connects servers to networks, to reduce the impact of downtime.

  Read More
• January 15, 2025

  

  

  

  Next Up for Custom AI Accelerators: Co-Packaged Optics

  By Michael Kanellos, Head of Influencer Relations, ´óÐãÖ±²¥

  Computer architects have touted the performance and efficiency gains that can be achieved by replacing copper interconnects with optical technology in servers and processors for decades¹.

  With AI, it¡¯s finally happening.

  ´óÐãÖ±²¥ earlier this month announced that it will integrate co-packaged optics (CPO) technology into custom AI accelerators to improve the bandwidth, performance and efficiency of the chips powering AI training clusters and inference servers and opening the door to higher-performing scale-up servers.

  The foundation of the offering is the ´óÐãÖ±²¥ 6.4Tbps 3D SiPho Engine announced in December 2023 and first demonstrated at OFC in March 2024. The 3D SiPho Engine effectively combines hundreds of components¡ªdrivers, transimpedance amplifiers, modulators, etc.¡ªinto a chiplet that itself becomes part of the XPU.

  With CPO, XPUs will connect directly into an optical scale-up network, transmitting data further, faster, and with less energy per bit. LightCounting estimates that shipments of CPO-enabled ports in servers and other equipment will rise from a nominal number of shipments per year today to over 18 million by 2029².

  Additionally, the bandwidth provided by CPO lets system architects think big. Instead of populating data centers with conventional servers containing four or eight XPUs, clouds can shift to systems sporting hundreds or even thousands of CPO-enhanced XPUs spread over multiple racks based around novel architectures¡ªinnovative meshes, torus networks¡ªthat can slash cost, latency and power. If supercomputers became clusters of standard servers in the 2000s, AI is shifting the pendulum back and turning servers into supercomputers again.

  ¡°It enables a huge diversity of parallelism schemes that were not possible with a smaller scale-up network domain,¡± wrote in a December article.

  Read More

Recent Posts

• The Golden Cable Initiative: Enabling the Cable Partner Ecosystem at Hyperscale Speed
• ´óÐãÖ±²¥ Named to America¡¯s Best Midsize Employers 2026 Ranking
• Ripple Effects: Why Water Risk Is the Next Major Business Challenge for the Semiconductor Industry
• Boosting AI with CXL Part III: Faster Time-to-First-Token
• ´óÐãÖ±²¥ Wins Interconnect Product of the Year for Ara 3nm 1.6T PAM4 DSP