Excelero Triples AI/ML and Hyperscale Customer Growth During FY19

Excelero Ltd. deployed 3x more new AI and ML customers during FY19, and also grew new hyperscale web customers using its NVMesh Elastic NVMe solution by 3x.

Revenue from new deployments represented 70% of its full year total, while 30% came from existing accounts. New customers secured during January 2020 continue the trend.

These results affirm the appeal of NVMesh as high-performance GPU storage to serve AI/ML and hypperscale applications. With 270% growth in enterprise AI deployments in the past 4 years, and 154% growth in 2019 alone (Tractica, 2019), data centre operators are re-evaluating their storage architectures to ensure that they support the high-throughput and low latency of AI/ML systems and associated transactional databases. NVMesh delivers up to 10x faster data processing for multi-server, multi-GPU compute nodes when working with massive datasets to carry out complex financial analysis, climate modellig and related workloads.

New AI/ML installations during 2019 include biotech and pharmaceutical firms and other marquee government-funded research institutions. In the hyperscale storage category, 2 of Excelero’s Fortune 1000 customers, both of whom have purchased additional NVMesh licenses annually for the past 3 years, also expanded their deployments in 2019. One new hyperscale deployment that rolled out in 2019 neared 5000 nodes.

“The AI revolution is real, hyperscale deployments are multiplying, and Excelero was fully focused on these key opportunities in 2019,” said Lior Gal, CEO and co-founder. “The word is out that when AI/ML deployments run into the GPU storage bottleneck, Excelero can set them right because our elastic NVMe architecture provides more bandwidth and I/Os per second while reducing latency. We’re positioned to capture even more of this high-growth segment in 2020.”

Other 2019 highlights include:

Collaboration on NVIDIA Magnum IO – In November debuted NVIDIA Magnum IO, a suite of software that enables greater data throughput on multi-server, multi-GPU computing nodes than previously possible. NVIDIA developed Magnum IO in collaboration with leaders in networking and storage, including Excelero. The GPUDirect Storage within Magnum IO enables researchers to bypass CPUs when accessing storage and quickly access data files for simulation, analysis or visualization. By combining GPUDirect Storage with NVMesh, users bypass CPUs all the way from GPU memory to NVMe devices, achieving frictionless access to NVMe performance for shared NVMe storage at local speed.
More partnerships – with its 100% channel sales model, the firm is expanding its partner network, and in 2019 added strategic relationships with Lenovo, Penguin, and QCT.
Subscription sales model – instead of perpetual licensing, the company adopted a subscription sales model in 2019, making it easier to buy and facilitate partner services.
More awards and patents including a FMS Best of Show 2019 for the Most Innovative Flash Memory Technology of the year; an International Business Awards Gold Medal Stevie Award, finalist in SearchStorage’s 2019 Products of the Year competition in the storage system and application software category, and a berth on the 2019 Entrepreneur 360 list for the third year in a row. During 2019 the firm also was awarded its third and fourth patent for heightened shared NVMe storage efficiency and an approach to tail latency, with 11 more patents pending.

Read also:
Excelero Grows Revenue 4X From FY17 to FY18
Closing multi-million dollar deals
17 Jan 2019 | Press Release