Latest news with #AMDInstinctMI350
Hindustan Times
16-06-2025
- Business
- Hindustan Times
AMD Advancing AI event: 6 key details you need to know
At AMD's Advancing AI event in San Jose, California, United States, the brand unveiled a range of hardware as well as software-centric announcements, which at large serves as the vision for its open AI ecosystem. Here are the key announcements: AMD Instinct MI350 series GPUs announced: AMD announced the Instinct MI350X and Instinct MI355X GPUs and platforms, which the company claims allow for a four times generation-on-generation AI compute boost for better AI solutions across various industries. End-to-end open standards rack-scale infrastructure: At the keynote, AMD also showed its open standards rack-scale AI infrastructure, which is already rolling out with AMD Instinct MI350 series accelerators, 5th Gen AMD EPYC processors, and AMD Pensando Pollara NICs in hyperscaler deployments, including Oracle's Cloud Infrastructure. The company says the broad availability for the same is going to be the second half of 2025. AMD unveils its next-generation AI rack called Helios: AMD stated that this is going to be built on the next-gen AMD Instinct MI400 series GPUs. When you compare this to the former generation, it can allow for 10 times more performance when running inference on mixture-of-experts models. AMD announced the broad availability of its Developer Cloud: The company says this is purpose-built for fast, high-performance AI development and allows for developers to have access to a fully managed cloud environment with tools, allowing for fast development. The company says the combination of ROCm 7 and the AMD Developer Cloud enables the company to expand access to next-generation compute. It is already collaborating with AI leaders like Hugging Face, OpenAI, and Grok. AMD's latest version of its AI software stack, ROCm 7: The company says that this is to serve the growing demands of generative AI and the demand for more compute. The company says that this is going to enhance the experience for developers and features improved support for industry-standard frameworks and more. AMD also revealed its partner ecosystem: Seven out of ten of the largest AI model builders, including the likes of Meta, OpenAI, Microsoft, and xAI, are now on board with AMD for training their AI models, the company said. For instance, it was detailed how the Instinct MI300X is used for Llama 3 and for Llama 4 inference. Sam Altman, OpenAI CEO, also discussed hardware, software, and more. Other companies like Microsoft and Red Hat also joined them. MOBILE FINDER: iPhone 16 LATEST Price, Specs And More
Techday NZ
20-05-2025
- Business
- Techday NZ
Dell launches new AI innovations for enterprise & research
Dell has announced a range of advancements in enterprise AI infrastructure, solutions and services to support organisations seeking to adopt and scale artificial intelligence. The company reported that 75% of organisations view AI as central to their strategy, but high costs and data security concerns remain significant obstacles. Dell aims to address these challenges by simplifying deployment, reducing expenses and enabling secure, scalable AI adoption through its expanded Dell AI Factory, enhanced infrastructure and an expanding partner ecosystem. The Dell AI Factory, launched a year ago, has received more than 200 updates and now supports AI workloads at any scale with new infrastructure, software improvements and collaborations with partners such as NVIDIA, Meta and Google. The company states its approach to on-premises AI inference can be up to 62% more cost effective for large language models compared to public cloud options. Among the notable product introductions is the Dell Pro Max Plus laptop, equipped with the Qualcomm AI 100 PC Inference Card, which the company states is the world's first mobile workstation to include an enterprise-grade discrete NPU. This platform is intended to provide rapid, secure on-device inferencing for large AI models, facilitating edge deployments outside traditional data centres. The Qualcomm AI 100 PC Inference Card offers 32 AI-cores and 64GB of memory to support engineers and scientists working with sizeable data models. Addressing the energy demands of AI workloads, Dell introduced the PowerCool Enclosed Rear Door Heat Exchanger (eRDHx), designed to capture 100% of IT-generated heat and reduce cooling costs by up to 60% compared to current solutions. This innovation supports water temperatures between 32°C and 36°C and enables increased data centre density, allowing organisations to deploy up to 16% more racks of dense compute without raising power consumption, and provides advanced leak detection and unified management features. For high performance computing and AI, Dell's PowerEdge XE9785 and XE9785L servers will support AMD Instinct MI350 series GPUs, promising up to 35 times greater inferencing performance. Both liquid- and air-cooled versions will be available to further reduce facility cooling costs. In terms of data management, Dell's AI Data Platform now includes updates designed to improve access to structured and unstructured data, with Project Lightning, a parallel file system, reported to deliver up to two times greater throughput than alternatives. Enhancements to the Data Lakehouse further streamline AI workflows for use cases like recommendation engines and semantic search, and the introduction of Linear Pluggable Optics aims to lower power use and boost networking efficiency. Dr Paul Calleja, Director of the Cambridge Open Zettascale Lab and Research Computing Services at the University of Cambridge, commented: "We're excited to work with Dell to support our cutting-edge AI initiatives, and we expect Project Lightning to be a critical storage technology for our AI innovations." Dell has also broadened its partner ecosystem to include on-premises deployments with platforms such as Cohere North, Google Gemini, Glean's Work AI platform and Meta's Llama Stack, as well as joint solutions with Mistral AI. The company is providing enhancements to its AI platform with AMD and Intel technologies, including upgraded networking, software stack improvements, container support and integration with Intel Gaudi 3 AI accelerators. Updates to the Dell AI Factory with NVIDIA include new PowerEdge servers supporting up to 192 NVIDIA Blackwell Ultra GPUs per standard configuration and up to 256 per Dell IR7000 rack with direct to chip liquid cooling. These advancements aim to simplify data centre integration, speed up rack-scale AI deployment and are reported to deliver up to four times faster large language model training compared to the previous generation. The PowerEdge XE9712, featuring NVIDIA GB300 NVL72, targets efficiency at rack scale for training and is said to offer up to 50 times more inference output and five times improvement in throughput, with new PowerCool technology supporting power efficiency in high-demand environments. The company intends to support the NVIDIA Vera CPU and Vera Rubin platform in future server offerings. In networking, Dell has extended its portfolio with new PowerSwitch and InfiniBand switches, deliver up to 800 Gbps of throughput, and are now supported by ProSupport and Deployment Services. Further software platform updates include direct availability of NVIDIA NIM, NeMo microservices and Blueprints, plus Red Hat OpenShift integration on the Dell AI Factory with NVIDIA. To streamline AI operations, Dell has introduced Managed Services for the AI Factory with NVIDIA, providing 24/7 monitoring, reporting, upgrades and patching for the stack, supported by Dell's technical teams. Michael Dell, Chairman and Chief Executive Officer, Dell Technologies, said: "We're on a mission to bring AI to millions of customers around the world. Our job is to make AI more accessible. With the Dell AI Factory with NVIDIA, enterprises can manage the entire AI lifecycle across use cases, from training to deployment, at any scale." Jensen Huang, Founder and Chief Executive Officer, NVIDIA, added: "AI factories are the infrastructure of modern industry, generating intelligence to power work across healthcare, finance and manufacturing. With Dell Technologies, we're offering the broadest line of Blackwell AI systems to serve AI factories in clouds, enterprises and at the edge." Jeff Clarke, Chief Operating Officer, Dell Technologies, stated: "It has been a non-stop year of innovating for enterprises, and we're not slowing down. We have introduced more than 200 updates to the Dell AI Factory since last year. Our latest AI advancements — from groundbreaking AI PCs to cutting-edge data centre solutions — are designed to help organisations of every size to seamlessly adopt AI, drive faster insights, improve efficiency and accelerate their results." Christopher M. Sullivan, Director of Research and Academic Computing for the College of Earth, Ocean and Atmospheric Sciences at Oregon State University, said: "We leverage the Dell AI Factory for our oceanic research at Oregon State University to revolutionise and address some of the planet's most critical challenges. Through advanced AI solutions, we're accelerating insights that empower global decision-makers to tackle climate change, safeguard marine ecosystems and drive meaningful progress for humanity."
Yahoo
06-02-2025
- Business
- Yahoo
AMD pulls up the release of its next-gen data center GPUs
AMD says that it plans to launch its next major data center GPUs, the AMD Instinct MI350 series, sooner than originally announced. During the company's Q4 2024 earnings call Tuesday, AMD CEO Lisa Su said AMD plans to sample the MI350 with "lead customers" this quarter and "accelerate" production shipments to "mid-year." "So, we had previously stated that we thought we would launch [the MI350] in the second half of [2025]," Su said. "And frankly, that bring-up has come up better than we expected, and there's very strong customer demand, so we are actually going to pull that production ramp into the middle of the year, which improves our relative competitiveness." AMD's data center revenue pales in comparison to rival Nvidia's, which regularly eclipses tens of billions of dollars per quarter. But AMD has gained ground in the last year, acquiring and retaining customers, including Meta, Microsoft, and IBM. AMD says that sales of AMD Instinct chips reached over $5 billion in 2024, and the company expects its data center segment to grow "double digits" in 2025. Last year, data center revenue made up around half ($3.9 billion) of AMD's overall revenue ($7.1 billion), the company reported on Tuesday. "I believe that the demand for AI compute is strong," Su continued. "[MI350] will be a catalyst for the data center GPU business … We see [the data center] business growing to tens of billions as we go through the next couple of years." Investors seemed generally pleased with AMD's fourth-quarter results. The company's stock was up 4.58% as of publish time.
