According to a recent LinkedIn post from VAST Data, the company is using the NVIDIA GTC conference to highlight its CNode-X platform for demanding AI workloads. The post describes CNode-X as a GPU-accelerated system that works with the VAST AI OS and NVIDIA cuVS and cuDF libraries to offload indexing tasks from CPUs.
Claim 30% Off TipRanks
- Unlock hedge fund-level data and powerful investing tools for smarter, sharper decisions
- Discover top-performing stock ideas and upgrade to a portfolio of market leaders with Smart Investor Picks
The LinkedIn post cites performance metrics including a 20x speedup for GPU-accelerated SQL queries, a 4.5x faster vector indexing lifecycle, and sustained ingestion of 1 million vectors per second into the VAST vector database. These proof points suggest VAST is positioning its storage architecture as an integrated compute layer for high-volume RAG and multi-agent AI systems.
For investors, the emphasis on GPU-centric data infrastructure points to VAST’s strategy to align closely with accelerated computing trends driven by NVIDIA’s ecosystem. If the platform gains traction among enterprises building complex AI applications, it could support higher software and systems revenue and deepen VAST’s relevance in the competitive AI data infrastructure market.
The focus on unifying storage and compute functions may also differentiate VAST from more traditional storage vendors that rely heavily on CPU-based architectures. This approach could enhance the company’s appeal to customers seeking lower latency and higher throughput for AI workloads, potentially strengthening its position in large-scale AI and analytics deployments.