Central processing unit

Broadcom Revolutionizes AI Networking Landscape with High-Performance 400G RoCE/RDMA Ethernet NICs

Retrieved on: 
星期一, 五月 20, 2024
Network interface controller, TCO, AVGO, PCI Express, Ecosystem, XPU, Central processing unit, Graphics processing unit, ROCE, Architecture, NIC, PALO, Video game

“Broadcom is prioritizing open standards and fostering collaboration with industry leaders to deliver the most extensive selection of high-performance connectivity solutions for AI infrastructure.

Key Points: 
  • “Broadcom is prioritizing open standards and fostering collaboration with industry leaders to deliver the most extensive selection of high-performance connectivity solutions for AI infrastructure.
  • The Broadcom family of Ethernet adapters meets the growing demand for higher data transfer speeds required in future AI networks.
  • The 400G NICs offered by Broadcom, built on its success in delivering Ethernet at scale, offers open connectivity at an attractive TCO for power-hungry AI applications.”
    This leading-edge next-generation portfolio is backed by Broadcom’s long-term, proven execution and multi-decade heritage of high-performance Ethernet solutions, from 1G to 400G.
  • The 400G PCIe Gen 5.0 Ethernet family (BCM57608) is broadly available from multiple server vendors, as well as from Broadcom.

INTC Class Action Alert: Robbins LLP Reminds Stockholders of the Lead Plaintiff Deadline in the Class Action Against Intel Corporation

Retrieved on: 
星期三, 五月 15, 2024
Intel, INTC, News, SAN, Person, Central processing unit, Growth, Graphics processing unit, LLP, Foundry, Semiconductor

The complaint alleges that on October 11, 2022, CEO Pat Gelsinger announced the Company would shift to an “internal foundry model” (the “Internal Foundry” or “Foundry” model).

Key Points: 
  • The complaint alleges that on October 11, 2022, CEO Pat Gelsinger announced the Company would shift to an “internal foundry model” (the “Internal Foundry” or “Foundry” model).
  • What Now: You may be eligible to participate in the class action against Intel Corporation.
  • Shareholders who want to serve as lead plaintiff for the class must file their motions with the court by July 2, 2024.
  • A lead plaintiff is a representative party who acts on behalf of other class members in directing the litigation.

BOXX AI Workstations Bring AI to the Deskside and the Data Center

Retrieved on: 
星期二, 五月 14, 2024
DNA, AMD, Linux, PCI Express, Engineering, AI, Central processing unit, Graphics processing unit, NVIDIA, Architecture, Knowledge, GPU, M.2, Medical device

AUSTIN, TEXAS, May 14, 2024 (GLOBE NEWSWIRE) -- BOXX Technologies, the leading innovator of high-performance computer workstations, rendering systems, and servers, today announced BOXX AI , a line of deskside and data center workstations created specifically to meet the demands of AI modeling and training with industry-leading power, precision, and performance.

Key Points: 
  • AUSTIN, TEXAS, May 14, 2024 (GLOBE NEWSWIRE) -- BOXX Technologies, the leading innovator of high-performance computer workstations, rendering systems, and servers, today announced BOXX AI , a line of deskside and data center workstations created specifically to meet the demands of AI modeling and training with industry-leading power, precision, and performance.
  • “Accelerating workloads is in our DNA, so now we are applying our knowledge and expertise to AI, creating innovative, high-performance BOXX systems purpose-built for AI design and development.”
    With up to four NVIDIA RTX™ 6000 Ada Gen GPUs (5.8 petaflops of combined AI performance) and 192 GB of GPU memory, NVIDIA-certified BOXX AI deskside workstations optimize deskside AI training and inference for smaller and early-stage AI models.
  • RAXX AI rack-mountable systems assist with the seamless extension of AI development into the data center and cloud, augmenting scaled resources for further model development and production.
  • As owned onsite platforms for experimentation and testing, BOXX AI and RAXX AI workstations let users protect their proprietary AI training and inferencing data at the source.

Numenta Unveils NuPIC 2.0, Elevates CPUs to Superior Choice for Running AI Models

Retrieved on: 
星期二, 五月 21, 2024
Apps, Applications, Technology, Software, Networks, Artificial Intelligence, Internet, Cloud, Partnership, Intel, Privacy, Inc., Intelligence, Product management, Deep learning, Archive, Partner, AI, Central processing unit, Graphics processing unit, RAG, LLM, GPT, Generative, Synthetic media, Video game

Numenta Inc., the world leader in deploying large AI models on CPUs, announced version 2.0 of its flagship product, the Numenta Platform for Intelligent Computing (NuPIC™).

Key Points: 
  • Numenta Inc., the world leader in deploying large AI models on CPUs, announced version 2.0 of its flagship product, the Numenta Platform for Intelligent Computing (NuPIC™).
  • NuPIC empowers companies to deploy large language models (LLMs) on CPUs, offering an efficient, scalable, and secure solution.
  • “Whether running hundreds of models on a single server or tuning models to their performance needs, customers can leverage generative AI techniques such as RAG on NuPIC to build gen AI applications.
  • Best of all, they can build and deploy models with complete privacy on CPUs, all within NuPIC with very little lift.”
    By leveraging the combination of NuPIC’s optimized models and CPUs’ inherent flexibility, NuPIC makes it possible to run large-scale generative AI models, such as Llama-2, on CPUs.

New AMD EPYC 4004 Series Processors Available as North America Retail Launch Exclusive from Newegg and Newegg Business

Retrieved on: 
星期二, 五月 21, 2024
Mobile, Wireless, Technology, Electronic Games, Entertainment, Electronic Commerce, Software, Hardware, Retail, Consumer Electronics, Online Retail, Partnership, Computer, AMD, Ecosystem, Sale, Ageing, Stock, Central processing unit, Personal computer, Commerce, Q&A, Multimedia, Electronics, Video game

Newegg Commerce, Inc. (NASDAQ: NEGG ), a leading global e-commerce company specializing in computers and consumer electronics, today announced that the new AMD EPYC™ 4004 Series Processors are available for preorder as a North America retail launch exclusive through both Newegg and Newegg Business.

Key Points: 
  • Newegg Commerce, Inc. (NASDAQ: NEGG ), a leading global e-commerce company specializing in computers and consumer electronics, today announced that the new AMD EPYC™ 4004 Series Processors are available for preorder as a North America retail launch exclusive through both Newegg and Newegg Business.
  • View the full release here: https://www.businesswire.com/news/home/20240521611405/en/
    AMD EPYC 4004 Series (Image: AMD)
    Newegg and Newegg Business are offering five SKUs of the AMD EPYC 4004 processors for preorder.
  • In addition to the North America retail launch exclusive, Newegg Business will host a free webinar on June 26, 2024, to discuss the new AMD EPYC 4004 processors features, performance, partnership solutions and offering a Q&A with AMD.
  • “AMD is an important partner for Newegg and Newegg Business so offering the AMD EPYC 4004 processors as a North America retail launch exclusive is a priority for our business,” said Ivan McClain, marketing manager for Newegg Business.

KAYTUS Introduces KR1280V2 - a 1U Server with 32 E1.S SSDs, Boosting I/O Performance for LLM Training

Retrieved on: 
星期二, 五月 21, 2024
Apps, Applications, Technology, Software, Networks, Internet, Hardware, IOT (Internet of Things), Data Management, Artificial Intelligence, Intel, TDP, AMD, IOPS, Java, PCI Express, BIOS, Solid-state drive, Data management, Central processing unit, Graphics processing unit, SFF, ARM, ISC High Performance, LLM, TB, Workload, Computer data storage

KAYTUS, a leading IT infrastructure provider, introduced at ISC High Performance 2024 its high-density computing server KR1280V2.

Key Points: 
  • KAYTUS, a leading IT infrastructure provider, introduced at ISC High Performance 2024 its high-density computing server KR1280V2.
  • It supports 32 E1.S SSDs, achieving an impressive flash memory density of 512TB in a 1U form factor.
  • Compared to the last generation, KR1280V2 adopts low latency storage and network design, doubling the high concurrency I/O performance for LLM training scenarios, and improving model training efficiency.
  • Deploying KR1280V2 servers for large-scale LLMs training ensures a twofold increase in high-concurrency I/O performance compared to the previous generation.

Robbins LLP Urges Intel Corporation (INTC) Stockholders with Large Losses to Contact the Firm for Information About Their Rights

Retrieved on: 
星期一, 五月 20, 2024
Class Action Lawsuit, Professional Services, Legal, Intel, INTC, News, Person, Central processing unit, Growth, Graphics processing unit, LLP, Foundry, Semiconductor

The Company’s product portfolio is comprised of central processing units (CPUs), chipsets, processors, graphics processing units (GPUs), and other semiconductor products.

Key Points: 
  • The Company’s product portfolio is comprised of central processing units (CPUs), chipsets, processors, graphics processing units (GPUs), and other semiconductor products.
  • For more information, submit a form , email attorney Aaron Dumas, Jr., or give us a call at (800) 350-6003.
  • The complaint alleges that on October 11, 2022, CEO Pat Gelsinger announced the Company would shift to an “internal foundry model” (the “Internal Foundry” or “Foundry” model).
  • Under the Internal Foundry model, Intel would recognize revenues generated from both external foundry customers and Intel Products, as well as technology development and product manufacturing costs historically allocated to Intel Products.

Astera Labs Extends Interoperability Leadership Driving Seamless PCIe 6.x Deployment

Retrieved on: 
星期二, 五月 28, 2024
Apps, Applications, Technology, Semiconductor, Other Technology, Software, Networks, Internet, Hardware, Data Management, Artificial Intelligence, Cloud, Quanta, Partnership, ALAB, Silicon, ODM, Foxconn, Ecosystem, Exercise, PCI Express, Collection, Senior, AI, Central processing unit, Graphics processing unit, Labs, GM, Enterprise, Quanta Computer, HPC

Astera Labs (Nasdaq: ALAB), a global leader in semiconductor-based connectivity solutions for AI and cloud infrastructure, today announced expanded PCIe® 6.x testing capabilities in its Cloud-Scale Interop Lab to enable seamless interoperability between Aries 6 PCIe/CXL Smart DSP Retimers and a broad range of PCIe 6.x hosts and endpoints.

Key Points: 
  • Astera Labs (Nasdaq: ALAB), a global leader in semiconductor-based connectivity solutions for AI and cloud infrastructure, today announced expanded PCIe® 6.x testing capabilities in its Cloud-Scale Interop Lab to enable seamless interoperability between Aries 6 PCIe/CXL Smart DSP Retimers and a broad range of PCIe 6.x hosts and endpoints.
  • Thad Omura, Chief Business Officer, Astera Labs, said, “As AI systems continue to advance at a rapid pace, data center operators need to deploy increasingly complex systems on an accelerated timeline.
  • It will enable closer collaboration and accelerate the deployment of seamless plug-and-play PCIe 6.x connectivity for AI.
  • We look forward to strengthening our partnership to ensure reliable interoperability and meet hyperscalers' needs for our future PCIe 6.x systems.”

Rafay Announces New Platform-as-a-Service Capabilities for GPU-based Workloads, Empowering Customers to Quickly Leverage the Power of AI

Retrieved on: 
星期三, 五月 15, 2024
Apps, Applications, Technology, Software, Networks, Artificial Intelligence, Internet, Cloud, Partnership, OpenAI, Databricks, Organization, Visual Studio Code, Platform, IDE, NTT, AI, Central processing unit, Agility, Graphics processing unit, Snowflake, System, LLM, MLOps, Platform as a service, NTT DATA, GPU, Video game

This makes compute resources for AI instantly consumable by developers and data scientists with the enterprise-grade guardrails Rafay customers leverage today.

Key Points: 
  • This makes compute resources for AI instantly consumable by developers and data scientists with the enterprise-grade guardrails Rafay customers leverage today.
  • Aspects such as cost are even more critical to control in the new age of AI.
  • In addition to applying its existing capabilities to GPU-based workloads, Rafay has extended its enterprise PaaS with features and capabilities that specifically support GPU workloads and infrastructure.
  • The new GPU-based capabilities in Rafay’s PaaS, along with the AI Suite are now generally available for customers.

KAYTUS Introduces the New K24V2 Multi-Node Server, Featuring Support for 8 High-Performance Processors in a 2U Form Factor

Retrieved on: 
星期二, 五月 14, 2024
Data Management, Apps, Applications, Technology, Software, Networks, Artificial Intelligence, Hardware, Intel, TDP, DAG, Omni-Path, Safety, SCG, VR, Central processing unit, Voltage regulator, ISC High Performance, ROCE, Voltage, CPU, Multimedia, HPC, InfiniBand, Temperature, Memory, Video game

This server supports four dual-socket nodes within a compact 2U form factor, featuring a total of eight cutting-edge Intel/AMD processors.

Key Points: 
  • This server supports four dual-socket nodes within a compact 2U form factor, featuring a total of eight cutting-edge Intel/AMD processors.
  • The K24V2 incorporates advanced cooling technologies optimized for extreme computing density, offering both cold plate liquid cooling and air cooling configurations.
  • View the full release here: https://www.businesswire.com/news/home/20240513994533/en/
    The K24V2 server enables support for four dual-socket nodes within a 2U form factor, allowing for independent operations between nodes.
  • Each node is compatible with two 5th Gen Intel® Xeon® processors or the latest 4th Gen AMD EPYC™ processors.