Tag:

Rack

AMD
Gaming Gear

AMD unwraps 2027 AI plans: Verano CPU, Instinct MI500X GPU, next-gen AI rack

by admin



AMD is accelerating its CPU, GPU, and AI rack-scale solutions roadmaps to a yearly cadence, so the company is set to introduce its all-new EPYC ‘Verano’ CPU, Instinct MI500-series accelerators, and next-generation rack-scale AI solution in 2027, the company revealed at its Advancing AI event. 

 ”We are already deep in the development of our 2027 rack-scale solution that will push the envelope even further on performance efficiency and scalability with our next generation Verano CPUs and Instinct MI500X-series GPUs,” said Lisa Su, chief executive of AMD, at the event. 

AMD’s 2026 plans for rack-scale AI solutions already look impressive as the company’s first in-house designed Helios rack-scale system for AI will be based on AMD’s 256-core EPYC ‘Venice’ processor (expected to deliver a 70% generation-to-generation performance improvement); Instinct MI400X-series accelerators projected to double AI inference performance compared to the Instinct MI355X; and Pensando ‘Vulcano’ 800 GbE network cards compliant with the UEC 1.0 specification. But the company is set to introduce something even more impressive the following year. 


You may like

That would be AMD’s second generation rack-scale system powered by its EPYC ‘Verano’ processors, Instinct MI500X-series accelerators, and Pensando ‘Vulcano’ 800 GbE NICs. 

AMD did not reveal any specifications or performance numbers for its second gen rack-scale solution, EPYC ‘Verano’ processors, or Instinct MI500X-series GPUs. However, based on a picture the company provided, the post-Helios rack-scale machine will feature more compute blades, thus boosting performance density. This alone points to higher performance and power consumption, which will come handy as this one will have to rival Nvidia’s NVL576 ‘Kyber’ system based on 144 Rubin Ultra packages (each packing for reticle-sized compute elements). 

Production of EPYC ‘Verano’ CPUs and Instinct MI500X-series accelerators in 2027 align perfectly with TSMC’s roll-out of its A16 process technology in late 2026, its first production node to offer backside power delivery, a technology particularly useful for heavy duty datacenter CPUs and GPUs. We do not know whether AMD’s 2027 processors and accelerators will rely on TSMC’s A16, though it isn’t unreasonable to speculate.

Follow Tom’s Hardware on Google News to get our up-to-date news, analysis, and reviews in your feeds. Make sure to click the Follow button.



Source link

0 comments
0 FacebookTwitterPinterestEmail
AMD's and Pegatron's 128-GPU MI350X system
Product Reviews

Pegatron preps 1,177 PFLOP AI rack with 128 AMD MI350X GPUs

by admin



Pegatron showcased a unique rack-scale solution based on 128 AMD’s next-generation Instinct MI350X accelerators designed for performance-demanding AI inference and training applications at Computex. The system precedes AMD’s in-house designed rack-scale solutions by a generation, so for Pegatron, this system will serve as a training vehicle to build rack-scale AMD Instinct MI450X-based IF64 and IF128 solutions that are about a year away.

The Pegatron AS501-4A1/AS500-4A1 rack-scale system relies on eight 5U compute trays, each packing one AMD EPYC 9005-series processor and four AMD Instinct MI350X AI and HPC accelerators for AI and HPC. The CPU and the accelerators are liquid-cooled to ensure maximum and predictable performance under high loads. The machine comes in a 51OU ORV3 form-factor, making it suitable for cloud datacenters relying on the OCP standards (read: Meta).

(Image credit: Tom’s Hardware)

The machine connects GPUs located in a different chassis using 400 GbE as AMD does not have proprietary switches for Infinity Fabric connections (in any case, the maximum scale-up world size of AMD’s Instinct is eight processors today). This contrasts with Nvidia’s GB200/GB300 NVL72 platform, which has 72 GPUs interconnected with the company’s ultra-fast NVLink connection. To that end, the Instinct MI350X system will barely match the GB200/GB300 NVL72 in terms of scalability.


You may like

The new machine will be used by OCP adopters for their immediate workloads and for learning how to better build AMD Instinct-based systems with loads of GPUs both from hardware and software ports of view. Pegatron’s machine is hard to overestimate for multiple reasons, main of which being setting the stage to challenge Nvidia’s dominance in rack-scale AI solutions.

(Image credit: Tom’s Hardware)

Given what we know about AMD’s Instinct MI350X, Pegatron’s 128-GPU rack-scale system based on these units offers up to a theoretical peak of 1,177 PFLOPS PFLOPs of FP4 compute for inference, assuming a near linear scalability. With each MI350X supporting up to 288GB of HBM3E, the system delivers 36.8TB of high-speed memory, enabling support for massive AI models that exceed the capacity of Nvidia’s current Blackwell-based GPUs.

(Image credit: Tom’s Hardware)

However, its reliance on Ethernet for GPU-to-GPU communication limits the system’s scalability. With a maximum scale-up domain of eight GPUs, the system is probably built for inference workloads or multi-instance training rather than tightly synchronized LLM training, where Nvidia’s NVL72 systems excel. Still, it serves as a high-performance, memory-rich solution today and a precursor toward AMD’s next-generation Instinct MI400-series solutions.

Follow Tom’s Hardware on Google News to get our up-to-date news, analysis, and reviews in your feeds. Make sure to click the Follow button.



Source link

0 comments
0 FacebookTwitterPinterestEmail
Shaurya Malwa
NFT Gaming

XRP Futures Rack Up $1.5M Trading Volumes on CME Debut

by admin



XRP futures contracts began trading on CME Group’s derivatives platform on May 19, recording at least $1.5 million in trading volume during the first session, a modest but notable debut for the major token.

CME data shows 4 standard contracts (each representing 50,000 XRP) traded on day one, totaling around $480,000 in notional volume at an average price of $2.40. The majority of activity came from 106 micro contracts (2,500 XRP each), accounting for over $1 million in additional volume.
The contracts are cash-settled and benchmarked to the CME CF XRP-Dollar Reference Rate, which is published daily at 4:00 P.M. London time. CME’s dual contract structure is designed to attract both institutional players and smaller participants, offering flexibility for various hedging and trading strategies.

“The launch of regulated XRP Futures on @CMEGroup marks a key institutional milestone for XRP,” Ripple CEO Brad Garlinghouse posted on X on Monday. He added that Hidden Road executed the first block trade.

The listing follows the CFTC’s classification of XRP as a commodity, a regulatory green light that cleared the path for CME to offer these products.

Analysts say the debut could also strengthen the case for a spot XRP ETF, with ETF Store president Nate Geraci saying such a product is “only a matter of time.”

While early volumes may appear modest, XRP’s inclusion on CME widens market dynamics for the major token in terms of price discovery, similar to how price-action on BTC and ETH futures is impacted when the U.S. market opens.



Source link

0 comments
0 FacebookTwitterPinterestEmail
Close