A collection of 117 posts tagged with "AI/ML"
Tl;dr: We train a small LLM to become good at reasoning with reinforcement learning (similar to the process that led to Deepseek R1) all against AIStor AIHub, an on-premises model repository. Based on the great GRPO demo by will brown. Motivation: A growing requirement for teams is the need for an organized, secure, "single source of truth"
Read more
In today’s AI-driven enterprise landscape, resource optimization has evolved from a desirable goal into an operational imperative. As organizations scale their artificial intelligence initiatives to meet rising demands for innovation, the efficient orchestration of compute resources directly shapes operational performance and model precision. The forthcoming integration of NVIDIA GPUDirect Storage (GDS) with MinIO AIStor is a co-engineered solution slated
Read more
The Arm architecture is revolutionizing the hyperscale cloud, propelled by its Total Cost of Ownership (TCO) advantages—lower power consumption and reduced cooling requirements—that enable sustainable, high-performance computing at scale. Industry leaders like AWS, Azure, and GCP are embracing Arm to drive their latest compute instances for AI training, harnessing its efficiency to meet the demands of data-intensive workloads.
Read more
Building upon AIStor's robust AI capabilities, MinIO's PromptObject has been enabling users to interact with their data through natural language queries as described here. PromptObject transforms how users interact with stored objects by allowing them to ask questions about their data's content and extract information using natural language—eliminating the need to write complex
Read more
Modern enterprises seeking to leverage AI capabilities often face a significant hurdle: the complex deployment and management of GPU infrastructure in their Kubernetes environments. MinIO's AIStor addresses this challenge head-on by integrating the NVIDIA GPU Operator, revolutionizing how organizations deploy and manage GPU resources for AI workloads. Through automated GPU setup, driver management, and resource optimization, this integration
Read more
MLflow Model Registry allows you to manage models that are destined for a production environment. This post picks up where my last post on MLflow Tracking left off. In my Tracking post I showed how to log parameters, metrics, artifacts, and models. If you have not read it, then give it a read when you get a chance. In this
Read more
In several previous posts on MLOps tooling, I showed how many popular MLOps tools track metrics associated with model training experiments. I also showed how they use MinIO to store the unstructured data that is a part of the model training pipeline. However, a good MLOps tool should do more than manage your experiments, datasets, and models. It should be
Read more
In the rapidly-evolving world of artificial intelligence (AI) and machine learning (ML), speed and scalability are paramount. The ability to process massive amounts of data in real-time is a critical requirement for organizations looking to leverage AI/ML for competitive advantage. Whether it's training large machine learning models, running complex inference tasks, or scaling data pipelines, the performance
Read more
This post first appeared on The New Stack on January 16th, 2025. Often, while accessing the legitimacy of a new technology receiving a lot of hype, studying existing core capabilities and history is helpful. If the new technology in question is not based on existing or imminent capabilities, we can label it as “hype” and move on. Another litmus test
Read more
2025 has inherited a slew of geopolitical concerns that started years ago. U.S. Foreign policy, U.S. - China Relations, China’s geopolitical maneuvers, Conflicts in the Middle East, Russian Ukraine war, and cybersecurity threats. Additionally, new leadership in the United States adds to the uncertainty created by these concerns. And, as if all this were not enough, the
Read more
The evolution of data roles never stops—first, we were all data scientists, then data engineers, and now, DataOps engineers. But is DataOps really new, or just a fresh take on the same mission: delivering business value through data?
Read more
In November of 2023, Amazon announced the S3 Connector for PyTorch. The Amazon S3 Connector for PyTorch provides implementations of PyTorch's dataset primitives (Datasets and DataLoaders) that are purpose-built for S3 object storage. It supports map-style datasets for random data access patterns and iterable-style datasets for streaming sequential data access patterns. The S3 Connector for PyTorch also includes
Read more
Earlier this month, Amazon held their re:Invent conference in Las Vegas, Nevada, from December 1st to 5th - a 5-day event. If you have never been to a re:Invent conference, then the word that describes it best is “huge” - not just in terms of the number of attendees (60,000) but also the breadth of topics covered.
Read more
In November of 2023 Amazon announced the S3 Connector for PyTorch. The Amazon S3 Connector for PyTorch provides implementations of PyTorch's dataset primitives (Datasets and DataLoaders) that are purpose-built for S3 object storage. It supports map-style datasets for random data access patterns and iterable-style datasets for streaming sequential data access patterns. In a previous post, I introduced the
Read more
How does Exness handle massive data volumes and demanding AI/ML workloads? By moving to an on-prem infrastructure powered by MinIO. From scaling their data lake to managing traffic peaks of 200 Gbps, MinIO supports their AI workflows, disaster recovery, and more.
Read more
Your DevOps Engineer’s customer should be your AI/ML Engineering Team. The DevOps Engineer is there to ease the friction points in infrastructure so AI/ML folks can focus on the task at hand. Any issues that come with the infrastructure should be the responsibility of the DevOps Engineer.
Read more
Almost a year ago (actually 11 months ago), I wrote about the “Starving GPU Problem” and how the horsepower of Nvidia’s Graphic Processing Units (GPUs) could be so powerful that your network and your storage solution may not be able to keep up - preventing your expensive GPUs from being fully utilized. Well, in those short 11 months, a
Read more
As AI workloads drive cloud costs through the roof, many companies are rethinking their approach. Moving select AI tasks back on-prem offers a path to predictable costs, improved performance, and stronger data control.
Read more
Interoperability is the key to building a flexible, future-ready AI data stack. As proprietary systems lock down innovation and drive up costs, open tools like S3-compatible storage and multi-format table systems offer the freedom to scale and adapt.
Read more
MinIO recently surveyed 656 IT leaders as part of a primary research initiative with User Evidence. The results were very interesting and underscore the massive sea change we are seeing in the enterprise, both around the movement to object storage and the interest in using object storage as the primary building block for an organization’s AI initiatives. We will
Read more
