At first thought, the idea of deploying a GPU within a NAS for AI computing might seem unconventional. However, as demonstrated by tech reviewer Jordan Ranous on StorageReview, it can yield remarkable outcomes. Ranous opted for QNAP’s TS-h1290FX NAS unit, featuring an AMD EPYC 7302P CPU, 256GB DRAM, 25GbE connectivity, and extensive PCI slots, perfect for accommodating an internal GPU and substantial storage capacity.
By integrating an Nvidia RTX A4000 GPU into the TS-h1290FX and setting it up for AI tasks using the Virtualization Station within the NAS, Ranous effectively streamlined AI workflows. The NAS environment proved capable of handling the demands of running AI software with the GPU augmentation.
Nvidia’s Chat with RTX
The utilization of Nvidia’s ChatRTX software suite was pivotal in enhancing the AI capabilities within the NAS. This software facilitated AI interactions through a GPT-based Large Language Model (LLM) empowered with a tailored dataset, enabling quick and context-aware responses while upholding privacy and security standards.
The installation process, as outlined by StorageReview, involved ensuring hardware compatibility, inserting the GPU, updating QNAP firmware and software, setting up the operating system within the Virtual Machine (VM), configuring GPU pass-through, installing GPU drivers within the VM, and verifying the functionality. Through these steps, the NAS was transformed into a potent AI computing platform, showcasing its suitability for businesses seeking cost-effective AI solutions.
Ranous highlighted the feasibility and affordability of enhancing a QNAP NAS with a GPU for AI purposes, stating, “We’ve shown that adding a decent GPU to a QNAP NAS is relatively easy and inexpensive. We put an A4000 to work, and with a street price of about $1050, that’s not bad when you consider Virtualization Station is free and NVIDIA ChatRTX is available at no charge.”
