Community hooked up storage (NAS) specialist QNAP is getting in on the sting synthetic intelligence (edge AI) increase, asserting add-in modules that hook up with appropriate QNAP NAS models so as to add three tera-operations per second (TOPS) of compute for on-device pc imaginative and prescient and different machine studying workloads.
“With the fast development of edge AI, the demand for real-time knowledge processing and evaluation on the edge has elevated,” says QNAP product supervisor Andy Chung in help of the corporate’s newest launch. “The mixture of QNAP NAS and the QAI edge AI accelerators not solely supplies a high-performance storage platform but additionally considerably boosts the computing effectivity of AI purposes, bettering total cost-effectiveness.”
QNAP has introduced the launch of two new accelerators for its NAS units, the M.2 QAI-M100 (above) and USB QAI-U100 (high). (📷: QNAP)
The 2 accelerator modules — the M.2 B+M-key QAI-M100 and USB 3.2 Gen. 1 Kind-A QAI-U100 — are, successfully, single-board computer systems primarily based round a Rockchip RK1808 system-on-chip that includes two Arm Cortex-A35 cores working at 1.6GHz and together with an on-board neural processing unit (NPU) rated at a claimed three tera-operations per second (TOPS) of minimum-precision compute. Their design has been tailor-made for compatibility with QNAP’s personal AI Core software program working on its QTS or QuTS Hero working programs. Each models embrace 1GB of devoted reminiscence and a passive heatsink for cooling.
In keeping with the corporate’s inside testing, including a QAI edge accelerator to a QNAP TS-673A NAS enhance the efficiency of object recognition within the QuMagie Sensible Album utility by as much as 36 per cent, with facial recognition pace boosted by as much as 22 per cent. For individuals who want much more energy, the corporate says a second accelerator might be added — although it has not said whether or not this supplies a linear doubling in efficiency or if there are overheads concerned.
Extra data on the QNAP QAI-M100 and QAI-U100 is accessible on their respective product pages; each units had been exhibiting as out-of-stock on the official QNAP retailer on the time of writing, with pricing confirmed by resellers at $59 for the QAI-M100 and $69 for the QAI-U100.
