Nvidia reports Q2 data center revenue surging as AI adoption drives demand for GPU compute infrastructure.
Nvidia's data center business is experiencing explosive growth, with revenue climbing 83 percent year-over-year to $760 million in the company's second quarter, positioning it as the company's fastest-growing segment. Bob Venero, CEO of Future Tech and an Nvidia partner headquartered in Holbrook, N.Y., attributes this surge to widespread adoption of GPU acceleration across industries. "We're seeing a ridiculous uptick around GPU utilization, whether it's deep learning or high-end graphics," Venero said, working with Fortune 100 customers. This expansion reflects what Nvidia CEO Jensen Huang characterized as "the widening gap between demand for computing across every industry and the limits reached by traditional computing," which drove the company's overall 43 percent year-over-year revenue growth in Q2. "Developers are jumping on the GPU-accelerated computing model that we pioneered for the boost they need," Huang stated.
The data center expansion is being fueled by multiple sources. According to CFO Colette Kress, growth was "driven by hyperscale demand as internet services increasingly leverage artificial intelligence," with AI applications spanning search, recommendation, fraud detection and voice. High-performance computing also contributed significantly, gaining traction in oil and gas, financial services and transportation. Huang identified transportation as a particularly promising vertical, emphasizing the GPU-accelerated computing requirements for ride-hailing and autonomous vehicles. "In each of those instances you need high-performance computing to use machine learning to make that perfect match between riders and drivers for ride-hailing calls," Huang explained, noting that "every single autonomous vehicle company will need more GPU-accelerated servers."
Nvidia unveiled its next-generation Turing GPU architecture at SIGGRAPH, positioning it as a breakthrough for both visualization and gaming markets. The architecture features built-in ray-tracing support for cinema-quality real-time graphics and Tensor Cores for AI inferencing. "Turing is a giant leap forward and the greatest advance for computing since we introduced CUDA over a decade ago," Huang said. The company's first Turing products—the Quadro RTX 8000, Quadro RTX 6000 and Quadro RTX 5000 GPUs—will ship in the fourth quarter for the visualization market, with gaming variants expected to launch at Gamescom. Meanwhile, Intel announced its first dedicated GPU arriving in 2020, bolstered by hiring former AMD talent.
Beyond data center, Nvidia's other business segments showed mixed performance in Q2. Gaming, the company's largest business, grew 52 percent year-over-year to $1.8 billion, making it the second-fastest growing unit. Professional visualization expanded 20 percent year-over-year to $281 million, while Automotive grew 13 percent year-over-year to $161 million. The OEM segment, however, declined 54 percent year-over-year to $116 million, attributed to reduced cryptocurrency mining demand.
Though Nvidia exceeded Wall Street expectations with Q2 revenue of $3.1 billion and net earnings of $1.94 per share ($1.2 billion in net income), the company's reduced guidance for Q3 proved disappointing to investors, sending the stock down as much as 5 percent in after-hours trading Thursday.