Combining the powerful Amazon EC2 F1 instance with easy-­to-­use Ryft technology gives organizations the ability to quickly and easily extract critical insights from all their data

ROCKVILLE, Md. — Ryft Systems Inc., a leader in powering faster, smarter business analytics applications, announced today that Amazon Web Services (AWS) customers running open source Elasticsearch on AWS can command performance up to 72 times faster than ever before by deploying Ryft technology on the new Amazon Elastic Compute Cloud (Amazon EC2) F1 instance. Ryft delivers more actionable insights by eliminating the data indexing, transformation and curation requirements of traditional compute architectures that cripple the ability to instantly mine data.

Capabilities provided by Ryft running on AWS give customers fast, simple access to data search and analysis with Elasticsearch. F1 instances offer customers the performance and feature gains afforded by FPGAs without requiring massive infrastructure investments or specialized FPGA knowledge. The magic of the faster performance and enhanced features is in the power of the abstracted heterogeneous compute technology provided by the F1 instance and the Ryft technology.

Performance benchmarking
According to Ryft’s benchmarking, Elasticsearch running on the Amazon EC2 F1 with Ryft software consistently performed in a fraction of the time it took to run the same query on the AWS C4.8xlarge instance type. Two benchmark tests using Ryft and F1 instances processed and completely searched 87GB and 175GB of log file data 63X and 72X faster, respectively.

 

*Benchmark comparisons of Ryft Cloud on F1 vs. Elasticsearch running on a commodity AWS EC2 – c4.8xlarge server
 

“Traditionally, companies wanting to take advantage of the performance benefits of new heterogeneous compute technologies had to deploy those technologies within their own data center infrastructure and take on all the associated costs and risks of that deployment. AWS’s investment in F1 changes all that,” said Des Wilson, CEO of Ryft. “AWS provides the cutting-edge platforms while Ryft makes these new technologies simple to use, which gives AWS customers the unique advantage of a low-risk, low-up keep solution with the highest performance. Ryft and F1 give users a substantial competitive advantage by allowing them to instantly mine their data to address changes in their target markets on the fly.”

Extending and expanding performance and functionality of traditional Elasticsearch
Ryft’s Elasticsearch is a preconfigured, ready-to-run image for the open source distributed real-time search and analytics engine. Searching raw files is as simple as running a normal Elasticsearch query — without the data indexing, transformation and curation.

 

Ryft seamlessly expands and extends the Elastic (ELK) stack
 

With Ryft and the new F1 instance, Elasticsearch users can eliminate data preparation and extend Elasticsearch to:

  • Accelerate workflows with the ability to deploy pre-index and post-index search
  • Speed search and analysis across unstructured data and JSON, XML, LOGs, CSV, TSV and other files with no transformation
  • Increase the power of edit distance with user-selectable changes to high-distance requests for fuzzy Hamming or Levenshtein
  • Enhance wildcard searches to include leading wildcard
  • Supercharge regular expressions and machine learning, coming later in 2017

To learn how to deploy Ryft with the F1 and to discuss how Ryft can help you power faster, smarter business analytics applications, contact us at www.ryft.com/free-trial.

About Ryft Systems Inc.
Ryft powers cloud-based and on-premise data analytics applications for faster, smarter data-driven decisions.  By making heterogeneous compute fast and simple, Ryft delivers instant insights into any data by eliminating the latency associated with data indexing, transformation and curation. Today, industry leaders rely on Ryft to quickly and simply unlock the value hidden in their data for real-time actionable insight.

Leave a Reply

Your email address will not be published. Required fields are marked *