Breaking AI Barriers at GTC 2025
Industry-First Storage Certification & Augmented Memory Grid Innovation
New Augmented Memory Grid: Enabling the Token Warehouse for AI Inference
Introducing WEKA’s Augmented Memory Grid—dramatically lower token costs and accelerate the inference pipeline.
WEKA Unleashing AI Reasoning with NVIDIA Blackwell
WEKA has achieved certification as a high-performance data store for NVIDIA GB200 (Blackwell) deployments.
Unlock the Full Potential of Your
AI Innovation
Don’t let delays in data diminish your progress.
Leading AI organizations are transforming their NVIDIA environments with the WEKA Data Platform™ to supercharge model training, reduce inferencing costs, and effortlessly scale to meet AI demands.
Ready to accelerate your AI journey? Schedule a meeting and discover how WEKA can propel
your innovation.
Meet with WEKA at NVIDIA GTC!
Take your NVIDIA-powered AI infrastructure to the next level with WEKA.
Schedule a meeting with us at GTC to discover how WEKA’s high-performance data platform eliminates bottlenecks, accelerates AI workloads, and empowers innovation—while lowering costs. Learn how leading enterprises are leveraging WEKA + NVIDIA to slash AI token costs, optimize data pipelines, and reduce overall AI infrastructure expenses—all while delivering maximum performance.
WEKA Session Catalogue
Wednesday, March 19th
3:20 – 3:35 PM PT
Blueprint for Scalable AI: Deploying RAG Applications with WARRP
Explore WARRP, a modular AI RAG Reference Platform built on WEKA with NVIDIA NeMo and NIMs, enabling up to 20x faster RAG-based AI inference. Learn how it simplifies deployment, enhances portability, and delivers enterprise-ready performance.
Add to Your ScheduleWednesday, March 19th
4:00-4:15PM PT
Accelerating KV Cache Inferencing using WEKA Augmented Memory Grid
Blueprint for Supercharging LLM Inference with PagedAttention over RDMA and WEKA
Discover how PagedAttention over RDMA (PAoR) enhances LLM scaling by reducing latency and boosting throughput with WEKA’s high-performance storage. Learn to optimize multi-node inference with vLLM and TensorRT-LLM for AI workloads.
View Session DetailsThursday, March 20th
1 PM PT
Maximize GPU utilization with high performance storage from WEKA with Amazon SageMaker HyperPod
Learn how WEKA and AWS optimize SageMaker HyperPod clusters to maximize training performance and GPU utilization. Explore real-world case studies, including Stability AI, and discover best practices for overcoming storage bottlenecks in foundation model development.
Thursday, March 20th
2:30 or 2:50 PM PT
From Dribble to Deluge: Supercharging AI with WEKA’s Game-Changing Architecture
Discover how WEKA’s AI-Native Data Platform on Azure eliminates data bottlenecks, transforming silos into high-performance pipelines. Learn how it delivers 10-100x speed boosts for AI, ML, and HPC workloads across hybrid and multi-cloud environments.
