Apr 11 2019
Big data references the amount, type, and speed of data analytics for purposes such as ecommerce metrics, business intelligence, research trends to detect patterns. The challenge is data storage and the best approach to synthesize the information. Machine learning involves interpreting data as it is processed to learn from the data, as a foundation to allow the machine to essentially teach itself over time.
Petabyte-scale parallel workloads are becoming more common and present problems for real-time data analytics that cannot be solved with conventional server architectures.
Forbes provides a staggering summary of the proliferation of big data including the fact that within five years there will be over 50 billion smart connected devices in the world, all developed to collect, analyze, and share data. And, this mind-boggling statistic: By 2020, 1.7 megabytes of new information will be created every second for every human being on the planet. More data has been created in the past two years than in the entire previous history of the human race. Couple those facts with this one: Less than 0.5% of all data is ever analyzed and used. https://www.forbes.com/sites/bernardmarr/2015/09/30/big-data-20-mind-boggling-facts-everyone-must-read/#85222de17b1e
Currently, up to 50% of the energy in data centers is consumed by moving data from storage to servers. Our Newport Computational Storage platform provides record-breaking acceleration for storage-intensive parallel applications- scaling storage, processing and linear interconnectivity. Our 16TB Newport U.2 NVMe SSD is the world’s most advanced U.2. SSD. We embed a multi-core ARM processor and dedicated hardware to accelerate machine learning in the Newport Computational Storage controller ASIC. Additionally, the package is 15mm and consumes less than 10 watts of power. Add more Newport SSDs to add more capacity and processing power. By moving server processing capabilities into storage devices, we minimize data movement which improves application performance, reduces power consumption, and lowers costs.
NGD Systems, in partnership with the TechUnplugged.IO team, will be at AI Expo in Olympia, London, April 25 and 26. Stop by our booth and contact me for more information regarding our products and how computational storage can address your real-time analytics needs Scott.Shadley@ngdsystems.com.