Be part of us in our return to New York on June fifth to accomplice with executives to discover complete strategies for auditing AI fashions for bias, efficiency, and moral compliance throughout organizations. Discover out how one can get entangled right here.
Knowledge is the lifeblood of recent AI techniques, however making it out there for these advanced workloads is kind of a problem. Weka, a startup working to simplify this job by making knowledge repeatedly out there on demand with a novel AI-powered providing, immediately introduced that it has raised a $140 million Sequence E spherical due to robust buyer curiosity. The funding, which comes solely from present traders, boosted the corporate’s valuation to $1.6 billion — double its November 2022 valuation.
“Few might have predicted how shortly the AI market would take off… However when [generative AI] “In December 2022, international demand for Weka knowledge platform software program has elevated dramatically, with massive enterprise clients and analysis organizations seeking to speed up their AI initiatives,” Jonathan Martin, president of Weka, instructed VentureBeat.
The corporate plans to make use of the contemporary capital in quite a lot of areas, primarily to additional improve its platform, a software program answer that eliminates knowledge bottlenecks present in legacy architectures and creates a “dynamic pipeline” that repeatedly feeds knowledge to GPUs and workloads AI, thereby rising their effectivity and sustainability.
What does Weka convey to the desk?
Whereas enterprise leaders proceed to reiterate their dedication to trendy workloads reminiscent of generative synthetic intelligence, downstream groups battle to convey these tasks to life resulting from a scarcity of knowledge and the gaps created by legacy architectures. A typical generative AI pipeline revolves round a number of phases of copying datasets, creating bottlenecks that gradual studying processes and burn extra power.
Occasion VB
AI Impression Tour: AI Audit
Be part of us once we return to New York on June 5 to talk with senior executives, delve into methods for auditing AI fashions to make sure equity, optimum efficiency and moral compliance throughout organizations. Safe your spot at this unique invitation-only occasion.
Request an invite
Based in 2013, Weka solves this drawback with what it calls a “dynamic knowledge pipeline.” Basically, the corporate’s software program knowledge platform makes use of a novel zero-copy structure that eliminates time-consuming copying and accelerates every stage of the AI pipeline in order that GPUs are consistently receiving knowledge. Finally, this allows fashions to be educated sooner and extra effectively, resulting in sooner insights and improved enterprise outcomes.
“By enabling organizations to simplify their IT stack for demanding AI- and GPU-intensive pipelines, delivering important value financial savings and velocity, Weka clients can get to market sooner than their opponents at decrease prices. Our important efficiency enhancements additionally lead to important financial savings within the power required to run GPU servers, making Weka probably the most sustainable strategy to run massive AI tasks,” defined Martin.
On the coronary heart of the Weka Knowledge platform is the scalable shared parallel file system WekaFS. It interfaces instantly with Peripheral Part Interconnect Specific (PCIe) drives connected to non-volatile reminiscence Specific (NVMe). It handles a variety of knowledge sorts and sizes and I/O profiles – and delivers 10x the efficiency of legacy community connected storage (NAS) and 3x the efficiency of native server storage.
“The Weka Knowledge Platform is designed for purchasers with advanced knowledge challenges and demanding knowledge environments, together with massive enterprises, cloud service suppliers, analysis establishments, media corporations, AI/ML corporations and startups, IoT functions and monetary providers corporations , who cope with productiveness. subsequent era intensive workloads reminiscent of AI, ML, HPC, quantum computing, 16K media and VFX,” Martin defined.
When it comes to sustainability, the president claims that the platform’s capacity to enhance GPU utilization with dynamic pipelines permits clients to avoid wasting 260 tons of CO2e (carbon dioxide equal) for each petabyte of knowledge saved.
Important progress and a street forward
Whereas the AI era wave began after the rise of ChatGPT nearly two years in the past, Weka has been making ready for this period because the starting. Consequently, as an alternative of growing its product in line with the wants of the market (similar to different gamers), the corporate aggressively sells to clients who match its doorstep.
“We had been already centered on modernizing the enterprise knowledge stack by growing an answer that might help the velocity, scale, simplicity and resiliency calls for of immediately’s high-performance workloads like AI/ML. Not solely had been we ready for the change, however we had been additionally forward of the curve,” famous Martin.
The corporate at the moment has greater than 300 clients, together with 12 of the Fortune 50. Some notable AI corporations utilizing the Weka platform embrace Stability AI, Midjourney, ElevenLabs, AI Safety Heart, and AI/GPU cloud service suppliers reminiscent of as Iris Power (IREN). ), Utilized Digital, NexGen Cloud and Yotta. On the monetary facet, annual recurring income from the corporate’s software program subscription mannequin has doubled year-over-year and now exceeds $100 million. Martin predicts that it’s going to triple and even quadruple this fiscal 12 months.
With this funding, Weka will enhance its money reserves from earlier rounds and work on increasing the enterprise to satisfy the demand for AI infrastructure following the generative AI growth. This, Martin mentioned, contains investing in analysis and improvement, enhancing the info platform and investing in buyer success initiatives.
The corporate expects to develop its international workforce of 400 by not less than 25% by the tip of this monetary 12 months. Different outstanding gamers competing with Weka within the distributed file system house are VAST Knowledge, Nutanix, IBM, Dell Applied sciences, Qumulo, and Pure Storage.
