Netris lands $15M Series A from a16z to speed up AI neocloud deployments
▼ Summary
– The AI boom has led many new businesses to launch data centers, but setup is complex and time-consuming, leaving expensive GPUs idle.
– Netris offers software that automates data center setup and operations, reducing time to market for neoclouds by managing network switches and providing multi-tenancy.
– Unlike large operators like Equinix or AWS, small neoclouds lack resources for custom automation, making Netris’s solution a key problem-solver.
– Netris’s platform is vendor-agnostic, hardware-accelerated, and used in over 35 GPU clusters, including by Nvidia-recommended customers like Lightning AI and Foxconn.
– Netris raised $15 million in Series A funding from Andreessen Horowitz, using algorithms (not AI) for deterministic network configuration.
The race to build AI data centers has never been more intense, but getting a new facility online remains a grueling, months-long ordeal. Even after securing the essential hardware,GPUs, network switches, and storage,operators must painstakingly configure every component, manage complex customer requirements, and start generating revenue. The clock is ticking: every day a GPU cluster sits idle represents a massive financial drain.
Netris, a network automation startup, believes it has the answer for the growing wave of neocloud providers. Its software runs directly on network switches and connects to a centralized platform that automates setup, configuration, and day-to-day operations. By providing network abstraction, the platform allows hardware configurations to be swapped out as needed. It also enables hardware-level isolation for multi-tenancy, letting neoclouds safely serve multiple clients from the same infrastructure.
For years, the data center game was dominated by giants like Equinix, Microsoft, and Google. These incumbents solved their networking challenges by hiring massive engineering teams or building proprietary automation tools. Smaller neocloud operators, however, lack those resources. “As a GPU cluster operator, you need to make configuration changes to every link, every day,” explained Netris CEO Alex Saroyan. “At traditional data centers, they used SDN [software-defined networking], but SDN is falling short because it’s a software technology. For AI, software is not okay. The amount of traffic is so high that everything must be hardware accelerated. You need something like SDN, but completely hardware accelerated. This is what we do, and we’ve been doing it for eight years.”
Saroyan emphasized that Netris is vendor-agnostic, working with standard networking equipment from both Nvidia and AMD server environments. The company’s technology has already won a powerful advocate: Nvidia. Two years ago, after a demo, the chip giant began recommending Netris to its own customers. Today, the platform is active in more than 35 GPU clusters worldwide, supporting roughly one million GPUs. Clients include Lightning AI, Foxconn, Visionbay, Hewlett Packard Enterprise, Tensorwave, and Telus.
To accelerate its momentum, Netris has secured a $15 million Series A investment from Andreessen Horowitz (a16z), as exclusively reported by TechCrunch. Notably, the company does not rely on AI for its core operations. “We started way before AI,” Saroyan said. “We understood the challenge early on and developed this algorithm early. AI is not deterministic. It’s good for creative work, but for changing many thousands of switch configurations, you don’t need to be creative. You need to be very persistent and repeatable.”
a16z partner Guido Appenzeller will join the company’s board. With the fresh capital, Netris plans to expand its engineering and sales teams, add support for more hardware vendors, and enhance the capabilities of its automation algorithm.
(Source: TechCrunch)