Meta enlists tiny Korean startup to build 'one-chip-like datacenter' CXL architecture introduced by Facebook's parent company and Panmnesia can handle almost 1000 AI GPUs per domain

Date:
Sun, 13 Sep 2026 22:35:00 +0000

Description:
Meta and Panmnesia propose a CXL data center architecture connecting up to
960 accelerators within one coherent computing domain across multiple racks.

FULL STORY ======================================================================Copy link Facebook X Whatsapp Reddit Pinterest Flipboard Threads Email Share this article 0 Join the conversation Follow us Add us as a preferred source on Google Newsletter Subscribe to our newsletter Meta wants nearly 1,000 AI GPUs operating inside one coherent data center domain Panmnesia's CXL design connects CPUs, accelerators, and memory across multiple racks The
architecture increases accelerator coordination from two devices to sixteen per CPU Meta is working with Panmnesia on an artificial intelligence data center design intended to make thousands of processors operate within one coherent environment.

The proposal uses Compute Express Link, or CXL, to connect CPUs ,
accelerators and memory across multiple racks without conventional network links. The architecture could bring as many as 960 AI accelerators into one coherence domain, putting its capacity at almost 1,000 GPUs working as one system. Latest Videos From TechRadar Watch full video here: CXL architecture extends accelerator connectivity The design addresses a problem that becomes increasingly difficult as AI training systems combine hundreds or thousands
of accelerators processing enormous data.

Every accelerator must progress through repeated computational stages,
meaning one delayed component can force other devices to wait before continuing their work. You may like Meta has so many expensive GPUs that it's buying SSDs to kill idle time Meta will reuse terabytes worth of DDR4 memory using CXL tech and avoid paying the RAM tax China wants to spend nearly $300 billion on a national data center grid

The researchers therefore focus on reducing unpredictable communication
delays between racks, where Ethernet or InfiniBand networks normally handle connections beyond individual systems.

Those networks require packet processing and software coordination, which can introduce greater latency variation as workloads spread across additional servers. Are you a pro? Subscribe to our newsletter Sign up to the TechRadar Pro newsletter to get all the top news, opinion, features and guidance your business needs to succeed! Contact me with news and offers from other Future brands Receive email from us on behalf of our trusted partners or sponsors By submitting your information you agree to the Terms & Conditions and Privacy Policy and are aged 16 or over.

CXL instead provides a shared coherence mechanism, allowing processors, accelerators and memory to participate within one connected resource environment.

The proposed architecture adds dedicated hardware intended to keep communication paths and processing behaviour more consistent across the
larger fabric.

Panmnesia's design uses a high-fan-out switch, a link acceleration unit and a fabric controller to manage traffic across the system. What to read next Startup swaps costly AI GPUs for Arm cores and up to 128TB of 'cheap' LPDDR6 RAM instead of expensive HBM to smash through the memory wall How silicon photonics lights the way for data centers Qualcomm targets Nvidia, AMD,
Huawei with Dragonfly AI accelerator rack loaded with 43TB of LPDDR5x, future generations set to smash 7PB/s bandwidth

These components are organized into trays, pods and a fabric, borrowing organizational principles normally associated with arranging functional
blocks inside semiconductor chips.

The company says its fabric controller and link acceleration unit have completed silicon validation, while its switch has already been fabricated.

Its switch has also been fabricated, while pre-release silicon is reportedly being supplied as development continues toward commercial products. Up to 960 accelerators in one coherence domain The review compares the proposed arrangement with NVIDIA 's GB200 NVL72, where one CPU directly coordinates
two accelerators through NVLink-C2C.

Under Panmnesia's architecture, one CPU could coordinate 16 accelerators, representing an eightfold increase over that reference configuration.

According to the published design, around 60 such groups could then form a coherence domain containing approximately 960 accelerators.

Cross-rack access could also fall from microsecond-level timing to several hundred nanoseconds, representing roughly an order-of-magnitude reduction.

The architecture would allow individual failed devices to be replaced without taking an entire server out of service.

That separation could reduce the amount of functioning hardware removed
during failures, although actual operational benefits would depend on implementation.

During the announcement, Myoungsoo Jung, CEO of Panmnesia, said, CXL enables the entire datacenter to operate as a single computing system.

The architecture still faces physical limits because electrical CXL
signalling reaches only about seven meters at 128 GT/s with two retimers.

Panmnesia therefore proposes optical CXL links for longer distances, and says it has already completed hardware proof-of-concept validation for that approach. Follow TechRadar on Google News and add us as a preferred source to get our expert news, reviews, and opinion in your feeds.



======================================================================
Link to news story: https://www.techradar.com/pro/meta-enlists-tiny-korean-startup-to-build-one-ch ip-like-datacenter-cxl-architecture-introduced-by-facebooks-parent-company-and -panmnesia-can-handle-almost-1000-ai-gpus-per-domain


--- Mystic BBS v1.12 A49 (Linux/64)
* Origin: tqwNet Technology News (1337:1/100)