Businesses interested in rolling out AI and machine learning need to pay attention to the goings-on at Hewlett Packard Enterprise’s (HPE) 2023 Discover event in Barcelona. At the event, the company unveiled an expansion in its AI initiatives, the centerpiece of which is a multi-faceted association with Nvidia, designed to optimize AI in business processes. This expanded partnership will encompass both hardware and software, aimed at boosting the performance of AI for enterprise workloads. Moreover, HPE will amp up its machine learning development environment with new features to improve customizability and creation of AI models. This powerful platform will be upgraded to be available as a full-fledged managed service, run through Amazon Web Services and Google Cloud. Additionally, HPE will offer AI-optimized cloud instances in HPE GreenLake and implement other enhancements to file storage for AI workloads.
According to Evan Sparks, HPE’s VP/GM for AI Solutions and Supercomputing Cloud, HPE’s goal, with these updates, is to create a comprehensive AI-native architecture, optimized from hardware to software that aligns with the company’s view that AI requires a fundamentally different architecture.
One of the new hardware systems under the agreement with Nvidia is the HPE ProLiant Compute DL380a, which is integrated with Nvidia L40S GPUs, Nvidia BlueField-3 DPUs (data processing units), and the Nvidia Spectrum-X Ethernet technology. Additionally, HPE will be working with Nvidia AI Enterprise and the NeMo framework software to provide further support to enterprise users. HPE intends to focus on enabling businesses to “lean into” generative AI workflows by releasing new features for tasks such as prompt engineering, retrieval augmented generation (RAG) and fine-tuning pre-trained models.
AI isn’t useful for organizations if it can’t operate by drawing insights from their data. So, HPE GreenLake for File Storage is being updated to give significantly greater performance, capacity expansion and throughput to cater to the needs of AI workloads. This is crucial as organizations are increasingly demanding hundreds of petabytes of data storage capacity to hold the huge language models they’re creating. As Patrick Osborne, SVP/GM, HPE Storage, put it, the updates to HPE GreenLake for File Storage are an evolving means to address customer “most challenging needs in the AI workloads space.”
HPE plans to increase its presence in the AI space with these announcements. The goal is to give businesses the tools to maximize AI’s potential and in turn keep them competitive in the market. These offerings are a part of HPE’s endeavor to ensure the full support of AI initiatives for businesses. For those interested in participating, The AI Impact Tour promises a forum for enterprises to engage in the AI community.
