Ryft Unlocks Value of Big Data with High Performance Analytics Platform

Ryft has unveiled the Ryft ONE, a commercial 1U platform designed to enable faster business insights by simultaneously analyzing up to 48 terabytes of historical and streaming data at 10 gigabytes/second or faster.


"The performance required to unlock the value of big data demands a new type of architecture that optimizes compute, storage and IO," said Des Wilson, CEO of Ryft. "The Ryft ONE will shift the market from legacy x86 architectures to a platform purpose-built for real-time analysis.”

“We are totally focused on being able to provide one of the first streaming and historical analysis solutions that can do them both simultaneously because it is the coupling of those two capabilities that make for a pretty unique solution for our potential customers,” added Bill Dentinger, vice president of products at Ryft.

The Ryft ONE will be available in early Q2 2015, as a hosted or on-premises solution, and the Ryft Early Access Program is now accepting applications. The actual GA release is about a month away, said Dentinger.

Real-Time Insights

The Ryft ONE enables data scientists and business analysts to gain insights from data in real time, replacing hundreds of high-end servers and decreasing operational costs by 70% with a single Ryft ONE, using less power than a hair dryer, to store and analyze the equivalent of the contents of Wikipedia in 4.5 seconds, without any data indexing, preprocessing, tuning or partitioning.

According to the vendor, the Ryft ONE combines advances in both hardware and software to boost big data analysis. Components include: a massively parallel, hardware-accelerated compute architecture that speeds historical and streaming data analytics to speeds of 10 gigabytes/second and faster, prebuilt algorithm components such as exact search, term frequency and fuzzy search that accelerate the development and execution of applications.

“Beyond price and performance, the other thing is ease of use,” said Dentinger. “From our 10 years of experience in working with the government on some of their big data systems, we have learned they have got a very strong talent base and they are able to use the products. However, as you go to the more mainstream, having it be user-friendly is very important and in many cases where customers are dealing with data that has a high velocity or the data has a very short half life, the speed is really important.”

Another aspect of the product that is also critical, Dentinger said, “is the ability for customers to copy their data on to our box and have it be immediately available - so the whole process of extract, transform, and load, which is a big issue in most of the big data solutions today, has been minimized.” 

Linux Front End and Open API

Ryft ONE features a Linux front end and open API. ““That is really important to us,” said Dentinger. “You will see all our API documentation, all of the RESTful API - all of this will be open for our customers because the truth is that without the hardware that software isn’t useful so there is not a lot of intellectual property there, but that allows customers to basically take the best of the art - as we start rolling out new customers and using it in their own deployments - without any additional work. We are using the standard Linux server as the front end so all the standards that interwork with a Linux Ubuntu 14.04 OS come out-of-the-box.”


The Ryft ONE also analyzes SSL encrypted data without added latency to protect sensitive data without sacrificing performance. Data can be stored either encrypted or unencrypted, with no impact to analytics performance. “In our particular case, our 48 terabytes’ worth of storage performs exactly the same whether encryption is on or off," Dentinger explained. "We actually need that as well because our 48 terabytes is provided by slide-in cartridges -  8 that have 6 terabytes in them - and the truth is that someone could slide one of those out and walk away with some very critical data, so we need that security aspect, as well. However, I think that today while customers are beginning to recognize that security is important, most of them have not taken that leap because of the barrier to being successful based upon performance. We don’t have problem.”

Performance benchmarks for term frequency, exact search and fuzzy search are provided at