Industry Consortium Forms to Drive UCIe Chiplet Interconnect Standard

By Tiffany Trader

March 2, 2022

A new industry consortium aims to establish a die-to-die interconnect standard – Universal Chiplet Interconnect Express (UCIe) – in support of an open chiplet ecosystem. Intel Corporation donated the UCIe 1.0 spec, which was then ratified by the 10 promoter members that span chip companies, semiconductor suppliers and cloud service providers.

The charter roster consists of AMD, Arm, Advanced Semiconductor Engineering, Inc. (ASE), Google Cloud, Intel, Meta (Facebook’s parent company), Microsoft, Qualcomm, Samsung and TSMC. Nvidia is notably absent from the grouping, despite exploring a chiplet path for its next-generation GPU designs.

The UCIe 1.0 specification is a complete standardized die-to-die interconnect, encompassing a physical layer, protocol stack, software model, and compliance testing, the consortium said. It natively maps PCI Express (PCIe) and Compute Express Link (CXL) protocols. The spec will enable end users to mix and match dies and other chiplet components from multiple sources with different packaging options. 

As with other chiplet implementations, there can be compute cores fabbed on the most advanced process node, while the memory and IO controller, for example, can be a prior generation process. This not only optimizes for semiconductor costs, it saves the expense of porting that IP over and validating it again.

An optimized physical layer is defined for both standard (2D) packaging, and advanced (2.5D) packaging as shown in the figure below. Depicted on the right are different packaging options in use by some of the founding companies, namely Intel, TSMC and ASE.

Source: UCIe Consortium

CXL and PCIe address a number of use cases. “With PCIe and CXL, SoC construction, link management, and security solutions that are already deployed can be leveraged to UCIe,” notes a consortium white paper, written by UCIe Chair (and Senior Intel Fellow) Debendra Das Sharma. “The usage models addressed are also comprehensive: data transfer using direct memory access, software discovery, error handling, etc., are addressed with PCIe/CXL.io; the memory use cases are handled through CXL.Mem; and caching requirements for applications such as accelerators are addressed with CXL.cache. UCIe also defines a ‘streaming protocol’ which can be used to map any other protocol.”

In an interview with HPCwire, Das Sharma discussed the motivations for the on-package integration of chiplets. “What we’re seeing is that a lot of our designs are hitting the reticle limit as the demand for processing is insatiable, so it’s easier for us – and by ‘us’ I mean the broader industry – to build smaller chiplets and stitch them together on the package so that they act as a single entity. So this is a scale up kind of solution.”

Enabling the construction of SoCs that exceed maximum reticle size is a key aim of the standard, which also aims to reduce time-to-solution by enabling die reuse, lower portfolio costs, and facilitate bespoke solutions. “You can basically do a mix and match of different dies depending on what the customer needs are, so that you can deliver a purpose-built solution out of a standard set of dies,” said Das Sharma.

One of the most interesting things about this announcement: the coterie of competitors who are coming together to drive the standard. This continues the trend of sophisticated coopetition plays that we saw, for example, with Intel’s March 2021 announcement of expanded partnerships with competitors TSMC and IBM. The more challenging (read: expensive) that design and packaging becomes, the greater the need for collaboration.

Will UCIe lay the way for devices from competing companies to be combined on an SoC? That’s already happening with Intel’s forthcoming Ponte Vecchio GPU, which is using TSMC’s N5 compute dies in addition to a number of other dies fabbed by Intel. The new standard creates the conditions for more strange bedfellows, and, in fact, the motivation is baked in.

The entire purpose of UCIe is to create an open and interoperable chiplet ecosystem,” said Das Sharma. We have a well-defined specification that encompasses electrical (including bump map), die-to-die adapter, protocol mapping, sideband, and configuration register that allows for seamless interoperability between UCIe components. This is similar to existing successful standards like PCIe and CXL. So yes, you can have a package where components from various companies can interoperate with each other.

Source: UCIe Consortium

The UCIe spec supports two broad usage models: package level integration and off-package connectivity. In the former, components that are attached at the board level – such as memory, accelerators, or networking devices – can be integrated at the package level. The second implements off-package connectivity using different types of media (optical, electrical cable, mmWave) and UCIe retimers. The retimers transmit the underlying protocol (PCIe, CXL) out to the rack or pod level, and enable resource pooling or sharing, thus facilitating resource disaggregation.

The consortium claims a 20-fold boost in IO performance at 1/20th the power envelope versus off-package SerDes at launch. The chart below shows key performance indicators.

Source: UCIe Consortium.

The new consortium follows other standards that have been jump-started by Intel, such as CXL, the PCI bus, PCI Express and USB, which have evolved into fully open and independent efforts.

The point is underscored by Jim Pappas, the Intel director and CXL chair acting as a consultant to the UCIe consortium. “Yes, Intel put the pen to the spec, but this is a consortium-owned spec at this point, and all future work on this is not [under the direction of] any one company,” he said. “There are 10 promoters, and the promoters are the ones who make the decisions on what goes into the spec. We each get one vote. Any company can join (within the scope of U.S. laws).”

“The nice thing about starting with a spec is that much of the early decisions have been made,” Pappas added. “You know what it is and where it’s going. The end result won’t necessarily be what was originally donated; it will evolve over time, it’ll get faster, it will have new capabilities, new protocols – the industry will evolve.”

Das Sharma likewise envisions a long life ahead. “My expectation is we have got decades of innovations that we are going to do with this,” he said. “This is really how SoCs are going to be built, from my perspective.”

The founding companies are in the process of finalizing incorporation as an open standards body, and expect to do so by year end. The member companies will then begin work on the next generation of UCIe technology, covering the design of the chiplet form factor, as well as management and enhanced security.


The ten companies provided brief statements of support. Below are the comments from AMD and Intel.

“AMD is proud to continue our long history of supporting industry standards that can enable innovative solutions addressing the evolving needs of our customers. We have been a leader in chiplet technology and welcome a multi-vendor chiplet ecosystem to enable customizable third-party integration. The UCIe standard will be a key factor to drive systems innovation leveraging heterogeneous compute engines and accelerators that will enable the best solutions optimized for performance, cost, and power efficiency,” said Mark Papermaster, executive vice president and chief technology officer, AMD.

“Integrating multiple chiplets in a package to deliver product innovation across market segments is the future of the semiconductor industry and a pillar of Intel’s IDM 2.0 strategy. Critical to this future is an open chiplet ecosystem with key industry partners working together under the UCIe Consortium toward a common goal of transforming the way the industry delivers new products and continues to deliver on the promise of Moore’s Law,” said Sandra Rivera, executive vice president and general manager, data center & AI, at Intel.

For more, see the launch announcement on the consortium website.

Full-size header image (source: UCIe Consortium).

Subscribe to HPCwire's Weekly Update!

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

Hyperion Study Tracks Rise and Impact of Linux Supercomputers

May 17, 2022

That supercomputers produce impactful, lasting value is a basic tenet among the HPC community. To make the point more formally, Hyperion Research has issued a new report, The Economic and Societal Benefits of Linux Super Read more…

ECP Director Doug Kothe Named ORNL Associate Laboratory Director

May 16, 2022

The Department of Energy's Oak Ridge National Laboratory (ORNL) has selected Doug Kothe to be the next Associate Laboratory Director for its Computing and Computational Sciences Directorate (CCSD), HPCwire has learned. K Read more…

Google Cloud’s New TPU v4 ML Hub Packs 9 Exaflops of AI

May 16, 2022

Almost exactly a year ago, Google launched its Tensor Processing Unit (TPU) v4 chips at Google I/O 2021, promising twice the performance compared to the TPU v3. At the time, Google CEO Sundar Pichai said that Google’s datacenters would “soon have dozens of TPU v4 Pods, many of which will be... Read more…

Q&A with Candace Culhane, SC22 General Chair and an HPCwire Person to Watch in 2022

May 14, 2022

HPCwire is pleased to present our interview with SC22 General Chair Candace Culhane, program/project director at Los Alamos National Lab and an HPCwire 2022 Person to Watch. In this exclusive Q&A, Culhane covers her Read more…

Argonne Supercomputer Advances Energy Storage Research

May 13, 2022

The lack of large-scale energy storage bottlenecks many sources of renewable energy, such as sunlight-reliant solar power and unpredictable wind power. Researchers from Lawrence Livermore National Laboratory (LLNL) are w Read more…

AWS Solution Channel

shutterstock 1103121086

Encoding workflow dependencies in AWS Batch

Most users of HPC or Batch systems need to analyze data with multiple operations to get meaningful results. That’s really driven by the nature of scientific research or engineering processes – it’s rare that a single task generates the insight you need. Read more…

Supercomputing an Image of Our Galaxy’s Supermassive Black Hole

May 13, 2022

A supermassive black hole called Sagittarius A* (yes, the asterisk is part of it!) sits at the center of the Milky Way. Now, for the first time, we can see it. The resulting direct image of Sagittarius A*, revealed this Read more…

Google Cloud’s New TPU v4 ML Hub Packs 9 Exaflops of AI

May 16, 2022

Almost exactly a year ago, Google launched its Tensor Processing Unit (TPU) v4 chips at Google I/O 2021, promising twice the performance compared to the TPU v3. At the time, Google CEO Sundar Pichai said that Google’s datacenters would “soon have dozens of TPU v4 Pods, many of which will be... Read more…

Q&A with Candace Culhane, SC22 General Chair and an HPCwire Person to Watch in 2022

May 14, 2022

HPCwire is pleased to present our interview with SC22 General Chair Candace Culhane, program/project director at Los Alamos National Lab and an HPCwire 2022 Per Read more…

Supercomputing an Image of Our Galaxy’s Supermassive Black Hole

May 13, 2022

A supermassive black hole called Sagittarius A* (yes, the asterisk is part of it!) sits at the center of the Milky Way. Now, for the first time, we can see it. Read more…

Royalty-free stock illustration ID: 1919750255

Intel Says UCIe to Outpace PCIe in Speed Race

May 11, 2022

Intel has shared more details on a new interconnect that is the foundation of the company’s long-term plan for x86, Arm and RISC-V architectures to co-exist in a single chip package. The semiconductor company is taking a modular approach to chip design with the option for customers to cram computing blocks such as CPUs, GPUs and AI accelerators inside a single chip package. Read more…

Intel Extends IPU Roadmap Through 2026

May 10, 2022

Intel is extending its roadmap for infrastructure processors through 2026, the company said at its Vision conference being held in Grapevine, Texas. The company's IPUs (infrastructure processing units) are megachips that are designed to improve datacenter efficiency by offloading functions such as networking control, storage management and security that were traditionally... Read more…

Exascale Watch: Aurora Installation Underway, Now Open for Reservations

May 10, 2022

Installation has begun on the Aurora supercomputer, Rick Stevens (associate director of Argonne National Laboratory) revealed today during the Intel Vision event keynote taking place in Dallas, Texas, and online. Joining Intel exec Raja Koduri on stage, Stevens confirmed that the Aurora build is underway – a major development for a system that is projected to deliver more... Read more…

Intel’s Habana Labs Unveils Gaudi2, Greco AI Processors

May 10, 2022

At the hybrid Intel Vision event today, Intel’s Habana Labs team launched two major new products: Gaudi2, the second generation of the Gaudi deep learning training processor; and Greco, the successor to the Goya deep learning inference processor. Intel says that the processors offer significant speedups relative to their predecessors and the... Read more…

IBM Unveils Expanded Quantum Roadmap; Talks Up ‘Quantum-Centric Supercomputer’

May 10, 2022

IBM today issued an extensive and detailed expansion of its Quantum Roadmap that calls for developing a new 1386-qubit processor – Kookaburra – built from modularly scaled chips, and delivering a 4,158-qubit POC system built using three connected Kookaburra processors by 2025. Kookaburra (Australian Kingfisher) is a new architecture... Read more…

Nvidia R&D Chief on How AI is Improving Chip Design

April 18, 2022

Getting a glimpse into Nvidia’s R&D has become a regular feature of the spring GTC conference with Bill Dally, chief scientist and senior vice president of research, providing an overview of Nvidia’s R&D organization and a few details on current priorities. This year, Dally focused mostly on AI tools that Nvidia is both developing and using in-house to improve... Read more…

Royalty-free stock illustration ID: 1919750255

Intel Says UCIe to Outpace PCIe in Speed Race

May 11, 2022

Intel has shared more details on a new interconnect that is the foundation of the company’s long-term plan for x86, Arm and RISC-V architectures to co-exist in a single chip package. The semiconductor company is taking a modular approach to chip design with the option for customers to cram computing blocks such as CPUs, GPUs and AI accelerators inside a single chip package. Read more…

Facebook Parent Meta’s New AI Supercomputer Will Be ‘World’s Fastest’

January 24, 2022

Fresh off its rebrand last October, Meta (née Facebook) is putting muscle behind its vision of a metaversal future with a massive new AI supercomputer called the AI Research SuperCluster (RSC). Meta says that RSC will be used to help build new AI models, develop augmented reality tools, seamlessly analyze multimedia data and more. The supercomputer’s... Read more…

AMD/Xilinx Takes Aim at Nvidia with Improved VCK5000 Inferencing Card

March 8, 2022

AMD/Xilinx has released an improved version of its VCK5000 AI inferencing card along with a series of competitive benchmarks aimed directly at Nvidia’s GPU line. AMD says the new VCK5000 has 3x better performance than earlier versions and delivers 2x TCO over Nvidia T4. AMD also showed favorable benchmarks against several Nvidia GPUs, claiming its VCK5000 achieved... Read more…

In Partnership with IBM, Canada to Get Its First Universal Quantum Computer

February 3, 2022

IBM today announced it will deploy its first quantum computer in Canada, putting Canada on a short list of countries that will have access to an IBM Quantum Sys Read more…

Supercomputer Simulations Show How Paxlovid, Pfizer’s Covid Antiviral, Works

February 3, 2022

Just about a month ago, Pfizer scored its second huge win of the pandemic when the U.S. Food and Drug Administration issued another emergency use authorization Read more…

Nvidia Launches Hopper H100 GPU, New DGXs and Grace Superchips

March 22, 2022

The battle for datacenter dominance keeps getting hotter. Today, Nvidia kicked off its spring GTC event with new silicon, new software and a new supercomputer. Speaking from a virtual environment in the Nvidia Omniverse 3D collaboration and simulation platform, CEO Jensen Huang introduced the new Hopper GPU architecture and the H100 GPU... Read more…

PsiQuantum’s Path to 1 Million Qubits

April 21, 2022

PsiQuantum, founded in 2016 by four researchers with roots at Bristol University, Stanford University, and York University, is one of a few quantum computing startups that’s kept a moderately low PR profile. (That’s if you disregard the roughly $700 million in funding it has attracted.) The main reason is PsiQuantum has eschewed the clamorous public chase for... Read more…

Leading Solution Providers

Contributors

Nvidia Dominates MLPerf Inference, Qualcomm also Shines, Where’s Everybody Else?

April 6, 2022

MLCommons today released its latest MLPerf inferencing results, with another strong showing by Nvidia accelerators inside a diverse array of systems. Roughly fo Read more…

D-Wave to Go Public with SPAC Deal; Expects ~$1.6B Market Valuation

February 8, 2022

Quantum computing pioneer D-Wave today announced plans to go public via a SPAC (special purpose acquisition company) mechanism. D-Wave will merge with DPCM Capital in a transaction expected to produce $340 million in cash and result in a roughly $1.6 billion initial market valuation. The deal is expected to be completed in the second quarter of 2022 and the new company will be traded on the New York Stock... Read more…

Intel Announces Falcon Shores CPU-GPU Combo Architecture for 2024

February 18, 2022

Intel held its 2022 investor meeting yesterday, covering everything from the imminent Sapphire Rapids CPUs to the hotly anticipated (and delayed) Ponte Vecchio GPUs. But somewhat buried in its summary of the meeting was a new namedrop: “Falcon Shores,” described as “a new architecture that will bring x86 and Xe GPU together into a single socket.” The reveal was... Read more…

Industry Consortium Forms to Drive UCIe Chiplet Interconnect Standard

March 2, 2022

A new industry consortium aims to establish a die-to-die interconnect standard – Universal Chiplet Interconnect Express (UCIe) – in support of an open chipl Read more…

Julia Update: Adoption Keeps Climbing; Is It a Python Challenger?

January 13, 2021

The rapid adoption of Julia, the open source, high level programing language with roots at MIT, shows no sign of slowing according to data from Julialang.org. I Read more…

Nvidia Defends Arm Acquisition Deal: a ‘Once-in-a-Generation Opportunity’

January 13, 2022

GPU-maker Nvidia is continuing to try to keep its proposed acquisition of British chip IP vendor Arm Ltd. alive, despite continuing concerns from several governments around the world. In its latest action, Nvidia filed a 29-page response to the U.K. government to point out a list of potential benefits of the proposed $40 billion deal. Read more…

Nvidia Acquires Software-Defined Storage Provider Excelero

March 7, 2022

Nvidia has announced that it has acquired Excelero. The high-performance block storage provider, founded in 2014, will have its technology integrated into Nvidia’s enterprise software stack. Nvidia is not disclosing the value of the deal. Excelero’s core product, Excelero NVMesh, offers software-defined block storage via networked NVMe SSDs. NVMesh operates through... Read more…

Google Launches TPU v4 AI Chips

May 20, 2021

Google CEO Sundar Pichai spoke for only one minute and 42 seconds about the company’s latest TPU v4 Tensor Processing Units during his keynote at the Google I Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire