About this event
As AI models continue to grow in size and context windows expand, GPU memory has become a critical limitation for achieving fast, efficient inference. When context memory capacity is exceeded, organizations experience increased latency, context recomputation, reduced throughput, and lower GPU utilization.
Join MinIO and NVIDIA for a technical discussion on how MinIO MemKV addresses the growing challenge of inference context memory, including the underlying infrastructure demands of AI inference.
Learn how MemKV provides a distributed, high-performance context memory layer that extends GPU memory capacity using RDMA-connected, memory-mapped NVMe storage.
Attendees will learn how MinIO MemKV:
Date/Time:
September 17, 2026
Time: 1:00 pm (EDT), 10:00 am (PT)
Hosted by
Expertise level technical instruction and curriculum design for object storage, enterprise Linux, and related technologies, including security, distributed computing, AI/ML, clustering, and performance. Previously an engineer at Sun Microsystems and Red Hat.
Patrick Riel is a Storage DevTech with more than a decade of experience across cloud, data center, and Kubernetes environments. His work spans networking, storage, AI, and infrastructure, with a focus on agentic AI, open-source contributions, and inference acceleration using technologies such as NVIDIA Dynamo, NIXL, and KV cache offloading. Outside of work, Patrick is an avid golfer and skier.
Adit Ranadive is a Senior Software Architect in NVIDIA’s Networking Software Advanced Development Group, where he develops systems technologies for high-performance AI inference. His current work spans NIXL, NVIDIA Dynamo, and distributed KV cache management systems, with an emphasis on resilient data movement for distributed inference across a range of storage and networking technologies. His earlier work includes using DPUs both to enable I/O stream processing for high-performance computing applications and to provide a common abstraction layer across diverse cluster storage technologies. He holds a Ph.D. and an M.S. in Computer Science from Georgia Tech and previously led high-performance network virtualization initiatives at VMware.
MinIO is the data and memory foundation for enterprise AI. AIStor and MemKV unify every layer of the data stack, from agentic AI and inference context memory to tables and objects across core, edge, and cloud. Each layer is built for the speed, scale, and economics that AI and analytics demand.
