NeuroBlade, NeuroBlade, a pioneer in data analytics acceleration, has been honored with the highly esteemed Most Innovative AI & Big Data Analytics Application Award at the distinguished Flash Memory Summit 2023. This recognition celebrates NeuroBlade’s advancements in Data Analytics acceleration technology, specifically its groundbreaking SQL Processing Unit (SPU) designed to transform data analytics performance on petabyte-sized datasets.
“The global data analytics market segment is expected to reach $301.8 billion by 2030, rising at a market growth rate of 26.8% CAGR. This explosive growth is fueled by the rapid development of artificial intelligence, IoT, hybrid cloud, and edge computing applications, demanding high performance processing,” said Jay Kramer, Chairman of the Awards Program and President of Network Storage Advisors Inc. “We are proud to recognize the NeuroBlade SQL Processing Unit which maximizes throughput at petabyte-scale. This solution makes possible the ability for queries to run multiple orders of magnitude faster than industry standard server systems. The end result is a solution which can empower businesses to make better decisions and help analyze customer trends and satisfaction. This can lead to new and better products and services.”
“We are delighted to be the recipients of this prestigious award at the Flash Memory Summit,” expressed Elad Sity, CEO and cofounder of NeuroBlade. “This achievement underscores our dedication to advancing the field of data analytics and our commitment to providing organizations with game-changing solutions. Our SQL Processing Unit (SPU) represents a breakthrough innovation that redefines the performance capabilities of AI and big data analytics applications.”
The NeuroBlade SPU ensures data bandwidth matching for analytics processing at storage I/O speeds. It facilitates high throughput data reads, fully utilizing PCIe bandwidth, and employs pipeline compute execution to reduce the output result size pushed back to memory/CPU, complementing existing compute and storage infrastructure. To enhance scalability and performance, the NeuroBlade Hardware Enhanced Query System (HEQS), a rack-mounted server, integrates multiple SPUs. Leveraging high-bandwidth NVMe flash storage and PCIe Gen 4.0/5.0 networking, it allows optimal query execution and peak storage performance utilization.
In the Flash Memory Summit submission, NeuroBlade demonstrated the NeuroBlade HEQS G100, a rack-mounted server built with 32 PCIe Gen4.0/5.0 NVMe SSD drives, 512GB memory, and eight NeuroBlade SPUs. The above setup was compared to a standard server with 128 CPU cores running a software-only solution. The results showed almost full utilization, 80%, of the SSD I/O throughput, up to 23x query performance improvement compared to the software-only alternative. Moreover, the demo showcased a remarkable 21x improvement in consistent Query Data Scan Rate (QDSR), driven by the fact that NeuroBlade’s architecture can process queries at a constant rate, regardless of query complexity.
NeuroBlade’s SQL Processing Unit (SPU) introduces unparalleled efficiency, scalability, and performance to data analytics tasks. This revolutionary technology has the potential to reshape industries such as healthcare, finance, federal, and more.
SOURCE: Businesswire