98 Enterprise AI in action: 5 real-world use cases powered by object storage Welcome to part 2 of our AI blog series. In Part 1, we untangled the alphabet soup of artificial intelligence. We explored AI, ML, DL, NLP, LLMs, and RAG in clear, actionable terms — showing how each element fits into today’s evolving IT strategies. Now, let’s kick it up a notch. In this article, we’re highlighting 5 real-world enterprise AI/ML use cases — all powered by scalable, metadata-rich object storage from Scality. First, we’ll showcase why object storage is being hailed as a natural fit for many AI initiatives. You’ll walk away with practical examples of how the right data foundation can unlock real-world AI success and, hopefully, be inspired to try your own AI/ML project. Why AI loves object storage You don’t have to take our word for it — industry analysts and technology leaders agree: object storage and AI are a perfect match. Articles like this one from The New Stack and TechTarget’s deep dive lay it out clearly. To summarize, software-defined object storage, or on-premises object storage, is: Built for massive scale: No need to worry about outgrowing your infrastructure. Optimized for unstructured data: Expertly handles data without fixed formats, including video, images, text documents, logs, and sensor data, which traditional databases struggle to manage. Flexible for on-prem AI workloads: A big win for reducing public cloud egress fees and improving data locality. Rich in metadata: Essential for training AI/ML models effectively. Equipped with immutability features: Ensures your training datasets remain secure and compliant. Now that we’ve got our bearings on the bigger picture of why experts are calling object storage a powerful and common-sense choice for AI, let’s take a closer look at the key attributes that organizations are leveraging with Scality object storage. What makes it particularly well-suited for this new wave of data-intensive innovation? Designing object storage for AI/ML: Scality’s approach Scality’s RING platform takes everything great about object storage and dials it up for enterprise-scale AI/ML workloads. Here’s how: Hardware vendor agnostic: Customers have the freedom to deploy on their choice of industry-standard x86 servers and storage media. No vendor lock-in, no forced appliance purchases — just full control over your infrastructure. Extreme scalability: Patented multiscale architecture allows you to scale independently in more dimensions — by far — than any other storage vendor. While a few modern storage vendors can disaggregate across 2 dimensions, RING scales independently across these 10 dimensions: Applications: Supporting multiple workloads within the same infrastructure Capacity: Expanding total storage volume as data demands grow Storage compute: Allocating computing power to match workload intensity Metadata: Scaling metadata operations to maintain searchability and efficiency S3 objects: Managing ever-growing object counts without performance degradation S3 buckets: Allowing for an increasing number of storage containers S3 authentications per second: Scaling security processes to match global access needs Throughput: Ensuring fast, uninterrupted data movement Objects per second: Supporting high transaction rates for real-time applications Systems management: Allowing IT teams to manage complex environments with ease. Get a deeper dive on these dimensions and their benefits here: https://www.solved.scality.com/multidimensional-scaling-cloud-storage/. Durability & availability: With SLA-backed 14 nines of data durability and 100% uptime guarantees, your data is safe, secure, and always on. Geo-resilience: Deploy across data centers for built-in disaster recovery and business continuity. Microsecond latency: The performance of our latest RING XP architecture is tuned for real-time applications. Tiered data movement: AI training and inference data move fluidly between hot and cold tiers, keeping costs under control without compromising performance. Metadata search: Intelligent metadata tagging and search enable AI workloads to quickly locate and access relevant data. Support for hybrid & multi-site architectures: Run across on‑premises, service provider, and edge environments with consistent storage services across locations. SEC17a-4 and FINRA compliance: Built-in immutability features meet the highest regulatory standards. ISO27001 and ISO 42001 certified: Security is baked in, not bolted on. Being ISO 27001 certified demonstrates a systematic and proactive approach to securing our software products, ARTESCA and RING, building trust with clients through validated security practices and continuous improvement methodology. We are also looking to add ISO 42001 certification to demonstrate our commitment to robust AI security practices. Published in December 2023, ISO 42001 is the first international standard for establishing, implementing, maintaining, and continually improving an artificial intelligence management system (AIMS). It provides a framework for organizations that develop, deploy, or use AI-based products and services to manage the unique risks and opportunities associated with AI. Future-proof architecture: Scality RING architecture was built to handle the extreme flexibility demands of cloud-scale storage — and that same adaptability makes it ideally suited for today’s AI data pipelines. But this isn’t a lucky coincidence. It’s the result of a deliberate design philosophy: Build storage that can evolve to meet any future need — even the ones we haven’t imagined yet. Tight integrations with the AI ecosystem: RING has integrations with key ISVs within the AI ecosystem — from databases and SQL query tools (Starburst, Dremio, Trino), to RAG and vectorial databases (LangChain, Milvus, Weaviate), to AI/ML frameworks (PyTorch, TensorFlow, JAX), ML ops and experiment tracking, data orchestration/pipelines etc. Five use cases: Real-world AI in action So, what does all of this look like in practice? Let’s dig into 5 real customer stories. These examples illustrate the transformative impact of AI and data lake technologies across diverse sectors, from healthcare and transportation to pharmaceuticals and aerospace. *Due to customer privacy or preference, some customer names have been omitted. AI use case #1: SeqOIA accelerates genomic research As part of France’s national genomic medicine initiative, SeqOIA plays a critical role in integrating whole genome sequencing into healthcare for patients with cancer and rare diseases. To support this mission, the lab built an AI-powered data lake on Scality RING, managing nearly 10 petabytes of genomic data. This enables fast, reliable access to massive datasets across thousands of compute nodes, powering everything from raw sequencing to deep analysis. With a two-tier storage architecture combining all-flash access for hot data and long-term storage in Scality RING, the scalable platform fuels accelerated research and more precise diagnostics. From detecting subtle genetic mutations to guiding personalized treatments, SeqOIA’s data-driven approach is transforming patient care through AI and advanced genomics. “We have solved our analytics processing needs through a two-tier storage solution, with all-flash access of temporary hot data sets and long-term persistent storage in Scality RING. We [use] RING to protect the petabytes of mission-critical data that enable us to carry out our mission of improving care for patients suffering from cancer and other diseases.” – Alban Lermine, IS and Bioinformatics Director, SeqOIA Read more about the SeqOIA case study here. AI use case #2: Fraud detection at a large US global bank A leading U.S. bank operating thousands of branches nationwide uses cloud-scale Scality RING object storage to power its massive Splunk-based Security Information and Event Management (SIEM) data lake. Splunk is a critical component of the bank’s fraud detection and prevention strategy, requiring 100% availability and uncompromising data durability. To meet these demands, the bank needed to store approximately 40 PB of Splunk SmartStore data with: Active/active deployment across 2 geographically separated sites Object replication within 2 minutes between the sites One-year data retention Altogether, this translates to roughly 80 PB of storage after erasure coding and overhead. To overcome the limitations of traditional storage systems, the bank turned to Scality RING, enabling the centralization of massive volumes of security logs and event data in a scalable, cost-effective object store. 14 nines durability, 100% availability with geo-distribution This shift has allowed the bank to: Ingest and analyze data at unprecedented speeds Significantly reduce threat detection and response times Improve alert fidelity and reduce false positives Strengthen its overall security posture and operational efficiency Dramatically reduce financial losses tied to fraud The inherent resilience and high availability of Scality object storage have also bolstered the reliability of the SIEM platform, ensuring continuous monitoring and analysis without compromise. RING now serves as the critical foundation for a modernized, AI-ready SIEM strategy that keeps pace with today’s evolving threat landscape. AI use case #3: Transportation pioneer powers autonomous vehicles A leading car manufacturer leverages Scality object storage to power its self-driving and crash detection AI. The scalable data lake ingests billions of sensor readings and telemetry from connected vehicles, forming a rich foundation for training sophisticated AI algorithms. By analyzing historical data, they are able to: Predict and potentially prevent collisions Develop increasingly precise autonomous driving features Enhance crash detection systems The agility of object storage enables seamless data access and analysis, accelerating innovation towards safer and smarter vehicles. AI use case #4: One of the largest insurance companies in North America accelerates claims and fortifies cyber resilience To enhance both security and analytical capabilities, a leading US insurance provider utilizes object storage as the foundation for its claims processing. All multimedia claims data submitted via their app is securely stored in their private cloud object store. This enables proactive threat detection, allowing them to identify and neutralize ransomware, viruses, and other malicious threats, safeguarding both their infrastructure and customer information. Moreover, the robust and high-performing object storage underpins their AI data lake, empowering sophisticated analytics engines. This strategic adoption of object storage has also simplified their storage infrastructure, leading to faster and more efficient claims processing. AI use case #5: Global biotech company revolutionizes biopharma research A leading genomics customer has revolutionized its research platform by integrating a high-performance filesystem with a scalable object store. The filesystem provides low-latency, high-throughput access crucial for demanding tasks such as DNA sequence analysis and complex computational modeling for drug discovery and development. Simultaneously, the object store serves as a cost-effective and durable repository for the large amounts of datasets generated, including raw sequencing data and processed results. This two-tiered architecture significantly: Accelerates research workflows Enhances the speed and accuracy of diagnostic processes Empowers the development of personalized medicine approaches by enabling efficient analysis for disease risk prediction and the identification of novel therapeutic targets. What’s next? Launch your first AI/ML initiative These stories show what’s possible, but they’re just the beginning. AI/ML is not one-size-fits-all, but one thing’s clear: the foundation matters. Without a scalable, reliable data layer, even the most advanced models struggle to deliver results. If you’re exploring how to launch your own AI/ML project — or aiming to boost performance by modernizing your data foundation — this practical guide to building a smarter chatbot is a great place to start. It’s an achievable way to gain hands-on experience with AI, lay the foundation for more advanced initiatives, and up-level your customer experiences. Whether you’re experimenting with a small LLM, automating customer insights, or developing next-gen analytics, we’re here to help you build smarter, faster, and with confidence. Other AI resources: A primer on the concepts of AI: ML, LLMs, DL, NLP, GenAI, and the rise of RAG Stop building dumb chatbots: The RAG + Scality RING solution Multidimensional scale: 10 must-have data storage dimensions to power your AI workloads The AI storage problem you didn’t see coming — and how Scality RING already solved it
