Will High-Speed Copper Interconnects Dim Optical’s Prospects?

By Michael Feldman

August 26, 2008

The prospects for optical cable interconnects never looked so good. With single and double data rate (SDR and DDR) InfiniBand firmly entrenched in HPC clusters and quad data rate (QDR) deployments just around the corner, optical cable makers are looking to displace bulky copper cabling in the next generation of high performance systems. Optical is already the accepted medium for long distance network communications, and now companies like EMCORE, Zarlink, Finisar, Luxtera, Tyco Electronics and Fujitsu are targeting the datacenter with optical solutions for multi-gigabit InfiniBand, Ethernet, and Fibre Channel.

As short-distance connection speeds get into the multi-gigabit per second realm and clusters scale out, copper’s weight, bulkiness and signal integrity problems start to work against it. At distances beyond 10 meters, traditional copper solutions usually can’t maintain good signal integrity at DDR rates (5 Gbps per channel * 4) unless much heavier cabling is used. At QDR rates (10 Gbps per channel * 4), copper peters out at about 7 meters. Even at shorter distances, cable’s weight and bulkiness can become a manageability issue because of limited bend radius and airflow blockage.

Optical has already notched a high-profile HPC win with the IBM Roadrunner supercomputer. That system is linked with EMCORE’s optical gear using DDR (InfiniBand) connections. According to IBM, they used over 55 miles of cable to hook all the compute and storage nodes together. For a massive system such as this, the superior physical manageability of optical fiber versus copper was probably the deciding factor.

But copper is not going down without a fight. Engineering innovation has continued to push performance to match the speeds of the latest network protocols. W. L. Gore & Associates, in particular, has been developing “active” copper cable solutions, targeting high performance interconnects for both HPC and general-purpose enterprise computing. The active component is a silicon device built into each end of the cable assembly that boosts the electronic signal and reduces noise, which enables greater reach or smaller cable diameters at a given distance.

Gore’s first generation active cables use Quellan’s Q:ACTIVE silicon devices to support DDR InfiniBand, 10GbE and 8 Gigabit Fibre Channel. Compared to passive copper, which the company also offers, Gore’s active cables are thinner and lighter, and can reach beyond 10 meters. Its current active product support InfiniBand connectors operating at 5 Gbps per channel, as well as QSFP (4-channel) and SFP+ (2-channel) connectors operating at 10 Gbps per channel.

Currently in development at Gore is an active QSFP offering aimed at the new QDR InfiniBand speeds. The product is being sampled now and will ship later this year. It will go head-to-head against optical assemblies being readied for QDR setups.

“We’re expecting to get at least 15 meters with quad data rate — hopefully 20,” said Russell Hornung, Gore’s director of new product development. The supported length will be a function of the acceptable cable gauge for customers. Since the majority of users are dealing with the problems associated with the weight of the copper, in most cases light gauge cable is going to be preferred. Hornung says they’re looking to achieve 15 meters with standard 28 AWG cable.

According to Gore’s research, more than half of the cabling in a datacenter is less than 10 meters and the majority — maybe 90 percent — is below 20 meters. If Gore’s copper offerings are able to reach those lengths with reasonably thin cable, that doesn’t leave much of a window for optical.

And compared to optical solutions, Gore thinks it has some big advantages in price and power consumption. Based on where the company has positioned its current active solutions, Hornung claims they are 20 to 25 percent less expensive than optical cables. Plus, according to him, they use only one-fifth the power of an optical solution at DDR. For QDR they expect to use only one-quarter to one-half the power, compared to optical. Their DDR assembly uses about 0.5 watt per cable (0.25 watt at each end), and depending on which chipset solutions they finally decide upon for the upcoming QDR products, they expect to use between 0.3 and 1.0 watt per cable.

The main reason active cables have a price and power advantage is that they only have CMOS devices drawing power at each end, while optical assemblies require CMOS plus a variety of optical-electronic components and one or more vertical-cavity surface-emitting lasers (VCSELs). The VCSELs tend to be the weak link in optical cables. Not only do they require a fair amount of power themselves, but they’re also sensitive to overheating. To have these tiny lasers hanging off a heat-generating server is not an ideal cooling scenario, so reliability becomes a concern.

Of course, the CMOS devices in the active copper assemblies are not immune to the heat themselves, but the thermal considerations are a much larger problem for optical than active copper. Hornung says CMOS is orders of magnitude more reliable than VCSELs — or any laser.

The VCSELs also are the biggest source of the cost differential between optical and copper, mainly because manufacturing lasers is not as advanced as it is for CMOS. Only some of the defective VCSELs can be found during fabrication. A significant fraction of bad lasers aren’t detected until they undergo some burn-in — after everything is assembled. By comparison, CMOS fabrication is a much more mature technology. Most troublesome silicon can be weeded out rather early in the manufacturing process.

The other complicating factor is that for multi-channel interconnects like DDR and QDR, most optical solutions use an array of VCSELs — one per channel. (The exception is Luxtera, which can power four transmitters with a single laser.) Multiple lasers complicate the inherent weaknesses of the optical cables, and since these multi-VCSEL products are still fairly new, the industry lacks much experience with device longevity.

Most of the optical vendors claim they can match or beat copper solutions once a specified length is exceeded — perhaps 10 meters or so. Beyond that, the cost of the copper itself starts to become a disadvantage compared to less-expensive optical fiber. But Hornung is not buying the economics. He believes the optical vendors are currently selling at below cost to get a foothold in the market. “To make a play in this space, they’re having to stoop to dangerous pricing levels,” he says. “We’ll see where it ends.”

Subscribe to HPCwire's Weekly Update!

Be the most informed person in the room! Stay ahead of the tech trends with industry updates delivered to you every week!

MLPerf Inference 4.0 Results Showcase GenAI; Nvidia Still Dominates

March 28, 2024

There were no startling surprises in the latest MLPerf Inference benchmark (4.0) results released yesterday. Two new workloads — Llama 2 and Stable Diffusion XL — were added to the benchmark suite as MLPerf continues Read more…

Q&A with Nvidia’s Chief of DGX Systems on the DGX-GB200 Rack-scale System

March 27, 2024

Pictures of Nvidia's new flagship mega-server, the DGX GB200, on the GTC show floor got favorable reactions on social media for the sheer amount of computing power it brings to artificial intelligence.  Nvidia's DGX Read more…

Call for Participation in Workshop on Potential NSF CISE Quantum Initiative

March 26, 2024

Editor’s Note: Next month there will be a workshop to discuss what a quantum initiative led by NSF’s Computer, Information Science and Engineering (CISE) directorate could entail. The details are posted below in a Ca Read more…

Waseda U. Researchers Reports New Quantum Algorithm for Speeding Optimization

March 25, 2024

Optimization problems cover a wide range of applications and are often cited as good candidates for quantum computing. However, the execution time for constrained combinatorial optimization applications on quantum device Read more…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at the network layer threatens to make bigger and brawnier pro Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HBM3E memory as well as the the ability to train 1 trillion pa Read more…

MLPerf Inference 4.0 Results Showcase GenAI; Nvidia Still Dominates

March 28, 2024

There were no startling surprises in the latest MLPerf Inference benchmark (4.0) results released yesterday. Two new workloads — Llama 2 and Stable Diffusion Read more…

Q&A with Nvidia’s Chief of DGX Systems on the DGX-GB200 Rack-scale System

March 27, 2024

Pictures of Nvidia's new flagship mega-server, the DGX GB200, on the GTC show floor got favorable reactions on social media for the sheer amount of computing po Read more…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HB Read more…

Nvidia Looks to Accelerate GenAI Adoption with NIM

March 19, 2024

Today at the GPU Technology Conference, Nvidia launched a new offering aimed at helping customers quickly deploy their generative AI applications in a secure, s Read more…

The Generative AI Future Is Now, Nvidia’s Huang Says

March 19, 2024

We are in the early days of a transformative shift in how business gets done thanks to the advent of generative AI, according to Nvidia CEO and cofounder Jensen Read more…

Nvidia’s New Blackwell GPU Can Train AI Models with Trillions of Parameters

March 18, 2024

Nvidia's latest and fastest GPU, codenamed Blackwell, is here and will underpin the company's AI plans this year. The chip offers performance improvements from Read more…

Nvidia Showcases Quantum Cloud, Expanding Quantum Portfolio at GTC24

March 18, 2024

Nvidia’s barrage of quantum news at GTC24 this week includes new products, signature collaborations, and a new Nvidia Quantum Cloud for quantum developers. Wh Read more…

Alibaba Shuts Down its Quantum Computing Effort

November 30, 2023

In case you missed it, China’s e-commerce giant Alibaba has shut down its quantum computing research effort. It’s not entirely clear what drove the change. Read more…

Nvidia H100: Are 550,000 GPUs Enough for This Year?

August 17, 2023

The GPU Squeeze continues to place a premium on Nvidia H100 GPUs. In a recent Financial Times article, Nvidia reports that it expects to ship 550,000 of its lat Read more…

Shutterstock 1285747942

AMD’s Horsepower-packed MI300X GPU Beats Nvidia’s Upcoming H200

December 7, 2023

AMD and Nvidia are locked in an AI performance battle – much like the gaming GPU performance clash the companies have waged for decades. AMD has claimed it Read more…

DoD Takes a Long View of Quantum Computing

December 19, 2023

Given the large sums tied to expensive weapon systems – think $100-million-plus per F-35 fighter – it’s easy to forget the U.S. Department of Defense is a Read more…

Synopsys Eats Ansys: Does HPC Get Indigestion?

February 8, 2024

Recently, it was announced that Synopsys is buying HPC tool developer Ansys. Started in Pittsburgh, Pa., in 1970 as Swanson Analysis Systems, Inc. (SASI) by John Swanson (and eventually renamed), Ansys serves the CAE (Computer Aided Engineering)/multiphysics engineering simulation market. Read more…

Choosing the Right GPU for LLM Inference and Training

December 11, 2023

Accelerating the training and inference processes of deep learning models is crucial for unleashing their true potential and NVIDIA GPUs have emerged as a game- Read more…

Intel’s Server and PC Chip Development Will Blur After 2025

January 15, 2024

Intel's dealing with much more than chip rivals breathing down its neck; it is simultaneously integrating a bevy of new technologies such as chiplets, artificia Read more…

Baidu Exits Quantum, Closely Following Alibaba’s Earlier Move

January 5, 2024

Reuters reported this week that Baidu, China’s giant e-commerce and services provider, is exiting the quantum computing development arena. Reuters reported � Read more…

Leading Solution Providers

Contributors

Comparing NVIDIA A100 and NVIDIA L40S: Which GPU is Ideal for AI and Graphics-Intensive Workloads?

October 30, 2023

With long lead times for the NVIDIA H100 and A100 GPUs, many organizations are looking at the new NVIDIA L40S GPU, which it’s a new GPU optimized for AI and g Read more…

Shutterstock 1179408610

Google Addresses the Mysteries of Its Hypercomputer 

December 28, 2023

When Google launched its Hypercomputer earlier this month (December 2023), the first reaction was, "Say what?" It turns out that the Hypercomputer is Google's t Read more…

AMD MI3000A

How AMD May Get Across the CUDA Moat

October 5, 2023

When discussing GenAI, the term "GPU" almost always enters the conversation and the topic often moves toward performance and access. Interestingly, the word "GPU" is assumed to mean "Nvidia" products. (As an aside, the popular Nvidia hardware used in GenAI are not technically... Read more…

Shutterstock 1606064203

Meta’s Zuckerberg Puts Its AI Future in the Hands of 600,000 GPUs

January 25, 2024

In under two minutes, Meta's CEO, Mark Zuckerberg, laid out the company's AI plans, which included a plan to build an artificial intelligence system with the eq Read more…

Google Introduces ‘Hypercomputer’ to Its AI Infrastructure

December 11, 2023

Google ran out of monikers to describe its new AI system released on December 7. Supercomputer perhaps wasn't an apt description, so it settled on Hypercomputer Read more…

China Is All In on a RISC-V Future

January 8, 2024

The state of RISC-V in China was discussed in a recent report released by the Jamestown Foundation, a Washington, D.C.-based think tank. The report, entitled "E Read more…

Intel Won’t Have a Xeon Max Chip with New Emerald Rapids CPU

December 14, 2023

As expected, Intel officially announced its 5th generation Xeon server chips codenamed Emerald Rapids at an event in New York City, where the focus was really o Read more…

IBM Quantum Summit: Two New QPUs, Upgraded Qiskit, 10-year Roadmap and More

December 4, 2023

IBM kicks off its annual Quantum Summit today and will announce a broad range of advances including its much-anticipated 1121-qubit Condor QPU, a smaller 133-qu Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire