Oracle unveils AMD-powered zettascale AI cluster for OCI cloud
The forthcoming zettascale AI cluster is designed to scale up to 131,072 MI355X GPUs, specifically architected to support high-performance, production-grade AI training, inference, and new agentic workloads. The cluster is expected to offer over double the price-performance compared to the previous generation of hardware.
Expanded AI capabilities
The new announcement highlights several key hardware and performance enhancements. The MI355X-powered cluster provides 2.8 times higher throughput for AI workloads. Each GPU features 288 GB of high-bandwidth memory (HBM3) and eight terabytes per second (TB/s) of memory bandwidth, allowing for the execution of larger models entirely in memory and boosting both inference and training speeds.
The GPUs also support the FP4 compute standard, a four-bit floating point format that enables more efficient and high-speed inference for large language and generative AI models. The cluster's infrastructure includes dense, liquid-cooled racks, each housing 64 GPUs and consuming up to 125 kilowatts per rack to maximise performance density for demanding AI workloads. This marks the first deployment of AMD's Pollara AI NICs to enhance RDMA networking, offering next-generation high-performance and low-latency connectivity.
Mahesh Thiagarajan, Executive Vice President, Oracle Cloud Infrastructure, said: "To support customers that are running the most demanding AI workloads in the cloud, we are dedicated to providing the broadest AI infrastructure offerings. AMD Instinct GPUs, paired with OCI's performance, advanced networking, flexibility, security, and scale, will help our customers meet their inference and training needs for AI workloads and new agentic applications."
The zettascale OCI Supercluster with AMD Instinct MI355X GPUs delivers a high-throughput, ultra-low latency RDMA cluster network architecture for up to 131,072 MI355X GPUs. AMD claims the MI355X provides almost three times the compute power and a 50 percent increase in high-bandwidth memory over its predecessor.
Performance and flexibility
Forrest Norrod, Executive Vice President and General Manager, Data Center Solutions Business Group, AMD, commented on the partnership, stating: "AMD and Oracle have a shared history of providing customers with open solutions to accommodate high performance, efficiency, and greater system design flexibility. The latest generation of AMD Instinct GPUs and Pollara NICs on OCI will help support new use cases in inference, fine-tuning, and training, offering more choice to customers as AI adoption grows."
The Oracle platform aims to support customers running the largest language models and diverse AI workloads. OCI users leveraging the MI355X-powered shapes can expect significant performance increases—up to 2.8 times greater throughput—resulting in faster results, lower latency, and the capability to run larger models.
AMD's Instinct MI355X provides customers with substantial memory and bandwidth enhancements, which are designed to enable both fast training and efficient inference for demanding AI applications. The new support for the FP4 format allows for cost-effective deployment of modern AI models, enhancing speed and reducing hardware requirements.
The dense, liquid-cooled infrastructure supports 64 GPUs per rack, each operating at up to 1,400 watts, and is engineered to optimise training times and throughput while reducing latency. A powerful head node, equipped with an AMD Turin high-frequency CPU and up to 3 TB of system memory, is included to help users maximise GPU performance via efficient job orchestration and data processing.
Open-source and network advances
AMD emphasises broad compatibility and customer flexibility through the inclusion of its open-source ROCm stack. This allows customers to use flexible architectures and reuse existing code without vendor lock-in, with ROCm encompassing popular programming models, tools, compilers, libraries, and runtimes for AI and high-performance computing development on AMD hardware.
Network infrastructure for the new supercluster will feature AMD's Pollara AI NICs that provide advanced RDMA over Converged Ethernet (RoCE) features, programmable congestion control, and support for open standards from the Ultra Ethernet Consortium to facilitate low-latency, high-performance connectivity among large numbers of GPUs.
The new Oracle-AMD collaboration is expected to provide organisations with enhanced capacity to run complex AI models, speed up inference times, and scale up production-grade AI workloads economically and efficiently.
Hashtags
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Techday NZ
2 days ago
- Techday NZ
Devart launches dbForge 2025.1 with AI assistant & new UI
Devart has unveiled dbForge 2025.1, presenting new features including an AI assistant, updated user interface, and expanded connectivity options across its database development tool suite. The new dbForge 2025.1 update introduces the dbForge AI Assistant, offering support for database professionals working with SQL Server, MySQL, Oracle, and PostgreSQL. With this release, Devart aims to address the evolving needs of those developing, managing, and optimising databases. dbForge AI Assistant is described as an AI-powered co-pilot designed to enhance SQL development. Its features include context-aware query generation, natural language conversion to SQL, query optimisation, and advanced troubleshooting for a broad spectrum of database engines. The new functionality seeks to allow users, regardless of their expertise, to interact with databases via natural language while maintaining precise output for complex queries. Oleksii Honcharov, Head of Engineering at Devart, commented on the introduction of the AI Assistant and the wider scope of the update: dbForge 2025.1 isn't just another version, it's a grand step forward for our entire product line. We believe technology should work for you, and with our AI-powered assistant, working with data is a breeze. This means fewer bottlenecks, high-quality answers, and smarter decisions across the board. The AI Assistant is available across the dbForge product family, providing users with several functionalities aimed at accelerating and simplifying SQL development. Amongst its capabilities, the assistant can: Generate context-aware SQL queries by analysing database metadata, without accessing actual data. Convert plain language commands directly into syntactically correct SQL queries, making database access more inclusive for non-experts or those seeking code-free interaction. Optimise existing SQL code with performance suggestions designed for quicker and more efficient queries. Offer explanations that deconstruct complex SQL statements into more comprehensible terms. Troubleshoot problematic SQL queries, diagnosing issues and suggesting or implementing fixes. Conduct error analysis with recommendations that account for the unique requirements and dialects of SQL Server, MySQL, MariaDB, Oracle, and PostgreSQL. Support an interactive chat interface that provides conversational assistance for SQL, general database topics, and dbForge products, removing reliance on separate documentation. The AI Assistant is fully integrated into all editions of dbForge products, including the free Express versions. New users receive a free 14-day trial period for the AI functions. UI and UX upgrades Enhancements to user interface and experience form another key aspect of dbForge 2025.1. The update features a redesigned interface that aims to be cleaner and more intuitive, improving navigation, panel management, and the discoverability of tools. These adjustments arise in response to feedback from users, with the goal of streamlining workflow and reducing potential friction within the application environment. Expanded analysis and compatibility Further improvements have been made to static code analysis capabilities in dbForge Studio for SQL Server and dbForge SQL Complete. The built-in T-SQL Analyzer receives an expanded set of rules, offering more customisation and deeper insight during code reviews. Aligning with evolving cloud integration requirements, dbForge 2025.1 introduces support for dedicated and serverless SQL pools within Azure Synapse Analytics. This enhancement is intended to improve data integration and management workflows for teams who rely on Microsoft's analytics service. Additionally, Devart has updated its support for PostgreSQL tools, which are now compatible with the recently released PostgreSQL 18. This ensures that users are able to leverage new database features as soon as they become available. Product availability dbForge 2025.1 is available for download or update across all core product lines, including dbForge Edge, the Studio series for SQL Server, MySQL, Oracle, and PostgreSQL, as well as dbForge SQL Complete and SQL Tools. Existing users can check for updates within their installed product, while new users may download and trial the full suite of dbForge tools.
Techday NZ
3 days ago
- Techday NZ
Ingram Micro launches Xvantage AI support for partners in ANZ
Ingram Micro has introduced the XvantageTM Enable | AI programme across Australia and New Zealand, aiming to provide structured support for customers and vendors seeking to adopt and scale AI-powered services and solutions. The XvantageTM Enable | AI offering comes as part of Ingram Micro's digital platform and features a combination of assessments, educational resources and vendor-specific enablement to assist technology partners through various stages of artificial intelligence adoption. The initiative is designed to help businesses improve customer experience, productivity and security using AI-enhanced technologies. According to Hope McGarry, Executive Managing Director, Australia at Ingram Micro, the programme is intended to demystify the process of bringing AI into existing operations while supporting partners with hands-on guidance and resources. McGarry stated: "At Ingram Micro, we're proud to support our partners through XvantageTM Enable | AI, a dynamic enablement program designed to accelerate AI adoption and drive real business outcomes. Whether you're just beginning your AI journey or scaling advanced solutions, our platform empowers you with curated learning paths, hands-on use cases, and seamless access to leading technologies. From intelligent automation to AI-enhanced cybersecurity and productivity, we're here to help you take advantage of the full potential of AI." The programme delivers several key resources for partners within the channel ecosystem, including AI Maturity Assessments, which help businesses evaluate their preparedness for adopting AI and uncover skill or capability gaps. Additionally, AI Basecamps offer foundational knowledge to enable effective delivery, and AI Growth Tracks provide step-by-step enablement alongside major technology vendors. Partners also have access to mission-specific data and AI services for secure and relevant data strategies, and immediate access to global centres of excellence for showcasing AI value to clients. Vanessa Sorenson, Chief Partner Officer ANZ at Microsoft, commented on the integration of the programme into partner offerings and its applicability across different experience levels and use cases, stating: "Ingram Micro's XvantageTM Enable | AI takes channel partner enablement to a new level with automated learning journeys and high-impact AI use cases across a variety of domains and lines of business. Regardless of the level of experience, Ingram Micro's program enables partners to deploy AI solutions from workplace productivity with Microsoft 365 Copilot to building AI Agents in Copilot Studio or implementing Intelligent Document Processing with Microsoft Azure AI services." The XvantageTM Enable | AI programme also covers AI-driven security solutions, providing a channel for partners to access, sell and deploy security offerings from leading vendors. Janet Docherty, Director, Channel and Alliances, ANZ, Palo Alto Networks, said: "XvantageTM Enable | AI provides Ingram Micro partners with a streamlined path to successfully sell and implement Palo Alto Networks' comprehensive AI-driven security portfolio. This program is key to empowering partners of all experience levels to maximise their opportunities and drive significant growth in the AI security market." With the broader release of XvantageTM Enable | AI in the Australian and New Zealand markets, Ingram Micro is targeting technology providers across the region, offering tailored engagement regardless of a partner's AI maturity or experience level. The programme is structured to address initial understanding through foundational learning, as well as advanced deployment of AI-enhanced business processes and security models. The initiative underscores the growing emphasis on AI as a driver of productivity, operational efficiency and risk management in the channel sector, with Ingram Micro's platform providing partners access to a blend of technical education, vendor collaboration and direct support in advancing market offerings. Follow us on: Share on:
Techday NZ
4 days ago
- Techday NZ
GenAI adoption surges in healthcare but security hurdles remain
Ninety-nine percent of healthcare organisations are now making use of generative artificial intelligence (GenAI), according to new global research from Nutanix, but almost all say they face challenges in data security and scaling these technologies to production. The findings are drawn from the seventh annual Healthcare Enterprise Cloud Index (ECI) report by Nutanix, which surveyed 1,500 IT and engineering decision-makers across multiple industries and regions, including the healthcare sector. The research highlights both rapid uptake of GenAI in healthcare settings and significant ongoing barriers around infrastructure and privacy. GenAI use widespread, but risks loom Among healthcare organisations surveyed, a striking 99% said they are currently leveraging GenAI applications or workloads, such as AI-powered chatbots, code co-pilots and tools for clinical development automation. This sector now leads all other industries in GenAI adoption, the report found. However, nearly as many respondents—96%—admitted their existing data security and governance were not robust enough to support GenAI at scale. Additionally, 99% say scaling from pilot or development to production remains a serious challenge, with integration into existing IT systems cited as the most significant barrier to wider deployment. "In healthcare, every decision we make has a direct impact on patient outcomes - including how we evolve our technology stack," said Jon Edwards, Director IS Infrastructure Engineering at Legacy Health. "We took a close look at how to integrate GenAI responsibly, and that meant investing in infrastructure that supports long-term innovation without compromising on data privacy or security. We're committed to modernising our systems to deliver better care, drive efficiency, and uphold the trust that patients place in us." Patient data privacy and security concerns underpin much of this hesitation. The number one challenge flagged by healthcare leaders is the task of integrating GenAI with legacy IT infrastructure (79%), followed by the continued existence of data silos (65%) and ongoing obstacles in developing cloud-native applications and containers (59%). Infrastructure modernisation lags adoption The report stresses that while GenAI uptake is high, inadequate IT modernisation could impede progress. Scaling modern applications such as GenAI requires updated infrastructure solutions capable of handling complex data security, integrity, and resilience demands. Respondents overwhelmingly agree more must be done in this area. Key findings also indicate that improving foundational data security and governance will remain an ongoing priority. Ninety-six percent agree their organisations could still improve the security of their GenAI models and applications, while fears around using large language models (LLMs)—especially with sensitive healthcare data—are prevalent. Scott Ragsdale, Senior Director, Sales - Healthcare & SLED at Nutanix, described the recent surge in GenAI adoption as a departure from healthcare's traditional technology adoption timeline. "While healthcare has typically been slower to adopt new technologies, we've seen a significant uptick in the adoption of GenAI, much of this likely due to the ease of access to GenAI applications and tools. Even with such large adoption rates by organisations, there continue to be concerns given the importance of protecting healthcare data. Although all organisations surveyed are using GenAI in some capacity, we'll likely see more widespread adoption within those organisations as concerns around privacy and security are resolved." Nearly all healthcare respondents (99%) acknowledge difficulties in moving GenAI workloads to production, driven chiefly by the challenge of integrating with existing systems. This indicates that, despite wide experimentation and early deployments, many organisations remain cautious about full-scale rollouts. Containers and cloud-native trends In addition to GenAI, the survey found a rapid expansion in the use of application containerisation and Kubernetes deployments across healthcare. Ninety-nine percent of respondents said they are at least in the process of containerising applications, and 92% note distinct benefits from cloud-native application adoption, such as improved agility and security. Container-based infrastructure is viewed as crucial for enabling secure, seamless access to both patient and business data over hybrid and multicloud environments. As a result, many healthcare IT decision-makers are expected to prioritise modern deployment strategies involving containers for both new and existing workloads. Respondents continue to see GenAI as a path towards improved productivity, automation and efficiency, with major use cases involving customer support chatbots, experience solutions, and code generation tools. Yet, the sector remains grappling with the challenges of scale, security, and complexity inherent to these new technologies. The Nutanix study was conducted by Vanson Bourne in Autumn 2024 and included perspectives from across the Americas, EMEA and Asia-Pacific-Japan.
