Data is essential for modern AI systems, yet providing it for complex workloads remains a challenge. Weka, a startup dedicated to addressing this issue, announced a $140 million Series E funding round, driven by notable customer interest. This latest investment, sourced entirely from existing investors, boosts Weka’s valuation to $1.6 billion—double its value from November 2022.
“Few predicted the rapid growth of the AI market...but when generative AI surged in December 2022, demand for Weka’s data platform exploded as enterprises and research organizations sought to accelerate their AI initiatives,” stated Jonathan Martin, president of Weka.
The newly raised capital will primarily enhance Weka’s platform, a software-driven solution designed to eliminate data bottlenecks caused by outdated architectures. It establishes a “dynamic pipeline” that continuously feeds data to GPUs and AI workloads, increasing efficiency and sustainability.
Weka's Innovative Solutions
Though enterprise leaders express commitment to modern workloads like generative AI, teams often struggle due to data silos and gaps within legacy systems. Traditional generative AI pipelines typically involve multiple data copying steps, creating bottlenecks that slow training processes and increase energy consumption.
Founded in 2013, Weka addresses these challenges with its “dynamic data pipeline.” The company’s platform utilizes a unique zero-copy architecture, accelerating each phase of the AI pipeline to keep GPUs constantly supplied with data. This approach enables faster and more effective model training, yielding quicker insights and improved business outcomes.
“By simplifying the IT stack for demanding AI and GPU-intensive processes, Weka empowers customers to enter the market faster and at a lower cost. Our performance enhancements also lead to significant energy savings for GPU server operations, making Weka a sustainable choice for large AI projects,” Martin explained.
Central to the Weka Data platform is WekaFS, a scale-out, shared parallel file system that directly interfaces with PCIe-connected NVMe drives. It accommodates various data types and sizes, delivering 10x the performance of legacy network-attached storage (NAS) and 3x that of local server storage.
“The Weka Data Platform is ideal for enterprises with complex data needs, including large corporations, cloud service providers, research institutions, media companies, AI/ML startups, IoT applications, and financial services firms deploying performance-intensive workloads like AI, ML, HPC, and VFX,” added Martin.
On the sustainability front, Martin claims that enhanced GPU utilization through dynamic pipelines enables customers to save 260 tons of CO2e (carbon dioxide equivalent) for every petabyte of stored data.
Significant Growth and Strategic Outlook
As the generative AI landscape has evolved since the advent of ChatGPT nearly two years ago, Weka has long been prepared. Instead of merely adapting to market demands, the company is proactively engaging customers seeking its solutions.
“We designed our enterprise data stack to meet the speed, scale, simplicity, and sustainability needs of modern, performance-intensive workloads like AI/ML. We were not only ready for this transition but ahead of it,” Martin remarked.
Weka currently serves over 300 customers, including 12 Fortune 50 companies. Notable AI firms leveraging Weka’s platform include Stability AI, Midjourney, and ElevenLabs, along with GPU cloud services like Iris Energy, Applied Digital, and Yotta. The company’s annual recurring revenue from its software subscription model has doubled year-over-year, surpassing $100 million, with predictions to triple or quadruple in the current fiscal year.
With this funding, Weka will bolster its cash reserves and scale operations to address the growing demand for AI infrastructure post-generative AI boom. This includes investments in research and development, platform enhancements, and customer success initiatives. The company anticipates increasing its workforce of 400 by at least 25% by the end of this fiscal year. Key competitors in the distributed file system arena include VAST Data, Nutanix, IBM, Dell Technologies, Qumulo, and Pure Storage.
Ranking
