Silicon Valley chip startup Enfabrica, specializing in networking chips for artificial intelligence data centers, announced on Tuesday that it secured $125 million in venture capital, with strategic investment from Nvidia.
Enfabrica, co-founded by former Broadcom and Alphabet’s Google executives, is part of a broader trend reshaping data centers to support generative AI technologies, akin to ChatGPT. Nvidia’s semiconductor chips are central to this transformation.
However, Nvidia’s GPU chips face an issue: they occasionally remain inactive because the networks connecting them cannot deliver data quickly enough.
To address this challenge, Enfabrica has developed a network chip that reconfigures data centers in innovative ways. This Enfabrica chip establishes a network resembling a hub-and-spoke system, enabling Nvidia GPUs engaged in data processing to access data from multiple sources seamlessly, avoiding performance bottlenecks.
Rochan Sankar, Enfabrica’s co-founder and CEO, explained that this approach significantly enhances GPU utilization. As a result, the same computational workload can achieve with approximately half the number of chips, resulting in cost savings—an important consideration in the technology industry.
Sankar remarked, “It’s no secret to Nvidia or anybody else out there that in order for AI computing to become truly ubiquitous, the cost curve has to come down. The key here is that we enable those GPUs to be better utilized.”
Atreides Management led the Series B venture funding round, with founder Gavin Baker, a veteran of Fidelity Investments, securing a position on Enfabrica’s board. New backers in this round include IAG Capital Partners, Liberty Global Ventures, Valor Equity Partners, Infinitum Partners, and Alumni Ventures. Sutter Hill Ventures, an earlier investor, also participated in the round.
