In a groundbreaking collaboration, VAST Data, an AI data platform provider, and Run:ai, an AI workload orchestration company , have joined forces to redefine AI operations on a large scale. This partnership introduces a full-stack solution that encompasses computing, storage, and data management, addressing the crucial requirements of enterprises venturing into expansive AI initiatives.
Run:ai specializes in streamlining NVIDIA accelerated AI infrastructure deployment across various cloud environments, including private, public, and hybrid setups. By implementing dynamic workload scheduling and innovative GPU fractioning techniques, Run:ai significantly enhances GPU allocation and utilization, thus optimizing AI project efficiency. The platform caters to diverse needs, ranging from facilitating interactive environments for data scientists to improving large-scale training and ensuring reliable, scalable inference. With its Open Architecture, Run:ai ensures a future-proof platform that seamlessly integrates with a wide array of industry-leading tools and technologies.
On the data management front, the VAST Data Platform consolidates storage, database, and containerized compute engine services into a unified, scalable software platform. Specifically engineered to empower AI and GPU-accelerated tools in modern data centers and clouds, VAST Data Platform offers a robust foundation for data-intensive organizations.
Renen Hallak, CEO and co-founder of VAST Data, emphasized the holistic approach to AI operations enabled by this partnership, stating, “Our collaboration with Run:ai goes beyond traditional AI solutions, integrating all essential components for an efficient AI pipeline. This announcement provides organizations worldwide with a blueprint for delivering more efficient, effective, and innovative AI operations at scale.”
Together, VAST Data and Run:ai offer organizations:
– Comprehensive Visibility for Resource and Data Management: The integration between the VAST Data Platform and Run:ai delivers a holistic AI solution, offering enterprises complete visibility across compute, networking, storage, and workload management in their AI operations.
– Cloud Service Provider-Ready Infrastructure: Recognizing the pivotal role of Cloud Service Providers (CSPs) in making GPU resources accessible, VAST and Run:ai present an efficient blueprint for deploying and managing AI cloud environments. This partnership ensures secure, enterprise-grade AI environments across shared infrastructure, with a focus on compute security and data isolation.
– Optimized End-to-End AI Pipelines: From data ingestion to model training and inferencing, organizations can accelerate data processing using NVIDIA RAPIDS Accelerator for Apache Spark and other AI frameworks available with the NVIDIA AI Enterprise software platform. The VAST Data Platform facilitates high-performance data pre-processing, enabling seamless AI pipeline optimization.
– Streamlined AI Deployment and Infrastructure Management: Run:ai simplifies resource allocation with fair-share scheduling and simplified multi-GPU distributed training, ensuring equitable access to compute resources for diverse data science teams. Meanwhile, the VAST DataSpace facilitates easy data access across geographies and multi-cloud environments, with robust encryption and access controls.
Omri Geller, CEO and co-founder of Run:ai, highlighted the significance of equitable compute resource access and extended visibility across the AI pipeline. He stated, “Our collaboration with VAST emphasizes unlocking maximum performance potential within complex AI infrastructures and significantly extends visibility and data management across the entire AI pipeline.”
The VAST Data Platform, coupled with Run:ai’s Open Architecture, seamlessly integrates with NVIDIA AI Enterprise and NVIDIA accelerated computing, enabling organizations to accelerate development, scale AI infrastructure, and optimize compute costs effectively.
By combining NVIDIA’s leading AI computing capabilities with Run:ai’s dynamic AI workload orchestration and VAST’s disruptive AI data platform, organizations can optimize resource utilization and gain enhanced control and visibility across compute and data layers.
VAST + Run:ai blueprints, solution briefs, and demos will debut at NVIDIA GTC 2024. To learn more about VAST and Run:ai, visit VAST Data at Booth #1424 during NVIDIA GTC.
