AI Startup Granulate Gets $30M B Round Funding as it Gains Traction in Data Centers
Granulate Cloud Solutions, the AI-based optimization software startup, announced its second financing round in 10 months as adoption of its data center workload accelerator gains traction among enterprise customers looking to squeeze more performance out of underutilized IT infrastructure.
Tel Aviv-based Granulate announced a $30 million funding round on Wednesday (Feb. 3), raising its venture total through two rounds to $45 million. The Series B round was led by Red Dot Capital Partners with participation from existing investors. Dawn Capital joined the round as a new investor.
The company’s technology revolves around using software agents at the operating system level to gauge IT resource utilization to prevent wasteful over-provisioning. With most organizations running production workloads on Linux-based platforms, Granulate says there is a “misalignment” between data and transactional workloads and the way the Linux kernel distributes CPU, storage and networking resources.
“The way that you measure your production workload is by latency, response time, throughput, IOPS,” said Asaf Ezra, Granulate’s co-founder and CEO. That results in Linux-based systems with costly over-provisioning, “bloated” IT infrastructure and resulting resource underutilization, he said.
The startup claims its AI-based data center optimization framework finetunes workload prioritization, reducing response times by as much as 40 percent, while cutting the cost of scaling workload performance by up to 60 percent.
“The AI is around the data flow [and] it has to be effective enough across multiple use cases,” Ezra added in an interview. “We had to create a way to understand each workload’s data flow. That’s where the AI comes in because for each workload, it’s going to be different.”
That means consumption of processing, storage, memory and networking resources will vary based on the data flow of different application workloads. To accomplish this, the Granular agent tracks and learns those otherwise manual resource allocation flows for individual workloads.
The startup and its investors are betting that enterprise customers moving more workloads to the cloud are looking for ways to reduce public cloud spending, or at least gain a better return on the substantial annual investment when expanding data center capacity.
When “you’re able to cut back [data center costs] by 40 percent, 50 percent, sometimes 60 percent, this is a huge amount of money,” Ezra argued.
Since its previous funding round last April, Granulate claims healthy customer and revenue growth, with the number of CPU cores managed rising 10-fold to more than 300,000 cores.
“With the rapid adoption of usage-based public clouds, we’ve returned to a world—not unlike the mainframe era—where dramatic improvements in speed and efficiency of computing systems drop directly to the bottom line,” said Lonne Jaffe, managing director at Insight Partners, a Granulate investor.