Fore! PCI Express 4.0 finally lands on Earth

Laggardly spec will help gear get faster and may vanish soon after

Somewhat later than expected, the Peripheral Component Interconnect Special Interest Group, or PCI-SIG, has finally gotten around to releasing the PCIe 4.0 specification, which describes the technical requirements for connecting devices through the PCI Express I/O bus in personal computers and servers.

"The delivery of the PCIe 4.0 specification to the industry is an important addition to our spec library as it delivers high performance 16GT/s data rates with flexible lane width configurations, while continuing to meet the industry’s requirements for low power," said Al Yanes, chairman and president of the industry consortium, in a blog post.

Version 4.0, first announced in 2011, doubles the data transfer rate available in PCIe 3.0, which debuted in 2010. It includes improvements like lane margining (which allows product engineers to assess the electrical margin available in a given system), lower latency, RAS features (reliability, availability, serviceability), and more robust I/O virtualization and platform integration.

The seven year gap between PCIe 4.0 and PCI 3.0 appears to have emboldened other data shuttling schemes. CCIX, which stands for the Cache Coherent Interconnect for Accelerators, counts companies like AMD, ARM, Broadcom, IBM, Micron, Qualcomm, Red Hat, Texas Instruments, and Xilinx as members.

The server-focused consortium in August announced that it had managed to transfer data at a rate of 25 Gbps, three times faster than PCIe 3.0, the current standard, and faster than PCIe 4.0.


That rate is similar to what another interconnection group, OpenCAPI, claims it can achieve.

PCI-SIG may be able to avoid being outpaced by its members through the acceleration of its delivery timeline. PCIe 5.0, announced in June and capable of 32GT/s, is scheduled to arrive in less than two years, Q2 2019.

Fast Cloud

PCIe speed to double by 2019 to 128GB/s


Doubling the PCI Express data transfer rate yet again has potential benefits for a variety of modern workloads like machine learning, gaming, and image processing, where large amounts of data get transferred over I/O channels.

"PCI-SIG has been able to accelerate the 5.0 timeline as we have improved the specification development process to minimize review cycles and thereby reduce delays," explained Yanes in an email to The Register. "We focus on delivering specs in a timeframe that our members require."

One change to the process, Yanes said, involves simplifying and streamlining the review of earlier versions of a specification.

"In addition, we were able to accelerate the timeline as the framework for the PCIe 5.0 specification was already in place thanks to improved silicon design processes, key functional enhancements and a future-proofed architectural design set forth in the PCIe 4.0 specification," he said.

Yanes cites several challenges moving from 3.0 to 4.0, including protocol improvements, lane margining, and 16GT/s support.

In version 5.0, Yanes said, the changes will be mainly limited to a speed upgrade. "The PCIe protocol already supports higher speeds via extended tags and credits, and existing PHYs in the industry already run at 28GHz/56GHz. For PCIe 5.0, we will leverage existing techniques out in the industry, new materials like Megtron 6, and new connectors will need to be considered," he said.

IBM's POWER9 processor, expected to ship this year, is said to be the first processor that will incorporate PCIe 4.0. Intel last month showed off its forthcoming 10nm Falcon Mesa FPGA, which includes a PCIe 4.0 interface.

More kit should follow, but with 5.0 so near, don't expect a long-term relationship. ®

Similar topics

Broader topics

Other stories you might like

  • GPL legal battle: Vizio told by judge it will have to answer breach-of-contract claims
    Fine-print crucially deemed contractual agreement as well as copyright license in smartTV source-code case

    The Software Freedom Conservancy (SFC) has won a significant legal victory in its ongoing effort to force Vizio to publish the source code of its SmartCast TV software, which is said to contain GPLv2 and LGPLv2.1 copyleft-licensed components.

    SFC sued Vizio, claiming it was in breach of contract by failing to obey the terms of the GPLv2 and LGPLv2.1 licenses that require source code to be made public when certain conditions are met, and sought declaratory relief on behalf of Vizio TV owners. SFC wanted its breach-of-contract arguments to be heard by the Orange County Superior Court in California, though Vizio kicked the matter up to the district court level in central California where it hoped to avoid the contract issue and defend its corner using just federal copyright law.

    On Friday, Federal District Judge Josephine Staton sided with SFC and granted its motion to send its lawsuit back to superior court. To do so, Judge Staton had to decide whether or not the federal Copyright Act preempted the SFC's breach-of-contract allegations; in the end, she decided it didn't.

    Continue reading
  • US brings first-of-its-kind criminal charges of Bitcoin-based sanctions-busting
    Citizen allegedly moved $10m-plus in BTC into banned nation

    US prosecutors have accused an American citizen of illegally funneling more than $10 million in Bitcoin into an economically sanctioned country.

    It's said the resulting criminal charges of sanctions busting through the use of cryptocurrency are the first of their kind to be brought in the US.

    Under the United States' International Emergency Economic Powers Act (IEEA), it is illegal for a citizen or institution within the US to transfer funds, directly or indirectly, to a sanctioned country, such as Iran, Cuba, North Korea, or Russia. If there is evidence the IEEA was willfully violated, a criminal case should follow. If an individual or financial exchange was unwittingly involved in evading sanctions, they may be subject to civil action. 

    Continue reading
  • Meta hires network chip guru from Intel: What does this mean for future silicon?
    Why be a customer when you can develop your own custom semiconductors

    Analysis Here's something that should raise eyebrows in the datacenter world: Facebook parent company Meta has hired a veteran networking chip engineer from Intel to lead silicon design efforts in the internet giant's infrastructure hardware engineering group.

    Jon Dama started as director of silicon in May for Meta's infrastructure hardware group, a role that has him "responsible for several design teams innovating the datacenter for scale," according to his LinkedIn profile. In a blurb, Dama indicated that a team is already in place at Meta, and he hopes to "scale the next several doublings of data processing" with them.

    Though we couldn't confirm it, we think it's likely that Dama is reporting to Alexis Bjorlin, Meta's vice president of infrastructure hardware who previously worked with Dama when she was general manager of Intel's Connectivity group before serving a two-year stint at Broadcom.

    Continue reading
  • Lithium production needs investment to keep pace with battery demand
    Report says $42b will need to be poured into industry over next decade

    Growing demand for lithium for batteries means the sector will need $42 billion of investment to meet the anticipated level of orders by the end of the decade, according to a report.

    Lithium is used in batteries that power smartphones and laptops, but there is also rising use in electric vehicles which is putting additional pressure on supplies.

    The report, Benchmark Mineral Intelligence, predicts that demand will reach 2.4 million tons of lithium carbonate equivalent by 2030, roughly four times the 600,000 tons of lithium forecast to be produced this year.

    Continue reading
  • Cars in driver-assist mode hit a third of cyclists, all oncoming cars in tests
    Still think we're ready for that autonomous future?

    Autonomous cars may be further away than believed. Testing of three leading systems found they hit a third of cyclists, and failed to avoid any oncoming cars.

    The tests [PDF] performed by the American Automobile Association (AAA) looked at three vehicles: a 2021 Hyundai Santa Fe with Highway Driving Assist; a 2021 Subaru Forester with EyeSight; and a 2020 Tesla Model 3 with Autopilot.

    According to the AAA, all three systems represent the second of five autonomous driving levels, which require drivers to maintain alertness at all times to seize control from the computer when needed. There are no semi-autonomous cars generally available to the public that are able to operate above level two.

    Continue reading

Biting the hand that feeds IT © 1998–2022