Innovation is at the core of Dell Technologies DNA, and our AI and industry experts are dedicated to guiding you every step of the way. We’re excited to explore the robust features of our AI-Ready Data Platform that are at the forefront of powering new generative AI (GenAI) use cases. Recently, we launched our AI Anywhere on Data Everywhere vision event, where we made a number of major announcements.
Envisioning the Modern Data Lakehouse
Our vision for an open modern data lakehouse addresses the escalating challenges organizations face in managing the exponential growth of data. With data sprawling across multiple environments, the proliferation of data sources and an ever-expanding selection of formats and tools, the need for flexibility has never been more critical. Traditional data lakehouses, often constrained by their design, fail to meet these evolving needs. They tend to work against the natural flow of your data, require unnecessary data movement and lock you into proprietary data formats that hinder interoperability.
We believe in a different approach—one that aligns with your data gravity, simplifies data discovery and querying by allowing in-place operations and champions the use of open table formats like Iceberg and Delta Lake. This approach ensures freedom from vendor lock-in, fostering a more open and flexible data environment.
Our partnership with Starburst marks a significant step toward realizing this vision. By integrating Starburst’s capabilities with Dell’s leading compute and storage solutions, we are laying the groundwork for a high-performance, scalable lakehouse. This foundation extends from the edge to the core and into the cloud, further delivering on our multicloud strategy.
Open Ecosystem of Partners: The Highlights
Understanding that data toolset diversity is as varied as our customers’ goals, we are embracing the entire data landscape. This comes to life through our partnerships with data industry leaders such as Databricks, Snowflake, Teradata and Cloudera. Our approach is not just about developing our solutions; it’s about creating an open ecosystem that embraces broader toolsets and data sources across AI workflows. Here are some recent partnership highlights:
-
- Introduced a groundbreaking reference architecture for cloud-native AI workflows, we are leveraging Databricks and Mosaic ML AI toolkits in tandem with our APEX File Storage for a dynamic, cloud-native experience. This initiative underpins our vision of AI Anywhere on Data Everywhere.
- With Teradata, we have rolled out enhanced platforms and ecosystems focusing on data warehousing and active archives. Launched early last year, these solutions merge Teradata’s software with our hardware expertise, and we’ve received great customer feedback.
Optimizing AI Across Multicloud Environments
In today’s rapidly evolving digital landscape, we recognize that an overwhelming majority of companies—nearly 87%¹—are adopting multicloud strategies due to dispersed data. This trend highlights the need for data flexibility, be it on-premises, at the edge or across various cloud environments. To address this, we give you the freedom to store and process your data in the most optimal location for your AI use case, ensuring efficiency, scalability and seamless access regardless of where your data resides.
Last year, we took a significant step forward by introducing the PowerScale OneFS scale-out software for public cloud, adding to our robust multicloud portfolio. Building on this foundation, we are excited to announce a major expansion of our cloud portfolio through a strategic partnership with Microsoft to deliver our APEX File Storage for Microsoft Azure solution. This collaboration enables customers to leverage Dell’s software-defined file services within Azure’s cloud. More importantly, the solution seamlessly integrates with Microsoft’s AI services, including OpenAI and Azure AI Vision. Our goal is to give customers freedom of choice so they can innovate faster with AI, and APEX File Storage for Microsoft Azure continues to deliver on this customer promise.
Delivering AI-Ready Performance
We are thrilled to share how we’ve taken our PowerScale storage to the next level, ensuring your AI workflows run faster so you can deliver insights for your business. We’ve made significant upgrades to both the hardware and software components of PowerScale, focused on minimizing disruptions and optimizing performance for AI workflows. Here are some highlights:
-
- We’re set to become the first storage provider to achieve NVIDIA Ethernet SuperPOD certification—a clear indicator of our commitment to leading-edge technology and excellence.
- Our recent OneFS 9.7 software update is another milestone, offering up to an 80% boost in storage performance.² This enhancement is crucial for accelerating GenAI processes, from model training to fine-tuning, and is available to our current PowerScale users.
- We just released our new all-flash storage nodes, specifically designed for AI. These nodes offer non-disruptive scaling and up to 2x the speed in streaming reads and writes,³ significantly improving GPU efficiency by reducing idle time and speeding up data processing. Additionally, our innovative smart client feature optimizes GPU client performance by intelligently distributing traffic, enabling data streaming capabilities of over 40 GB per second to a single client.4
Looking Ahead: Innovations for 2024
The future is bright with promising AI developments on the horizon in 2024. We invite you to watch the replay of our AI Anywhere on Data Everywhere event for a glimpse into what lies ahead. Embarking on your AI journey with Dell means having a dedicated team of AI and industry specialists by your side from the outset. From strategy formulation to data preparation and deployment, we’re committed to transforming your AI aspirations into tangible results. We look forward to hearing more about your AI use cases!
Connect with your Dell or partner representative today and take the first step toward realizing your AI-driven innovations.
1 Flexera, 2023 State of the Cloud Report, n=750.
2 Based on Dell internal analysis. Compared PowerScale OneFS 9.7 with OneFS 9.4
3 Based on Dell preliminary testing. Oct 2023. Compares the latest generation PowerScale all-flash nodes with OneFS 9.7 vs. the previous generation PowerScale all-flash nodes with OneFS 9.4. Actual results may vary.
4 Based on Dell preliminary testing. Oct 2023. Peak bandwidth using multipath client driver and limited by client network.
