FPGA
Connect with Intel® experts on FPGAs and Programmable Solutions
215 Discussions

2023 Q1 Intel® & LRL Webinar: Blazingly Fast Self-Searching Storage Accelerates Big Data Analytics

Sabrina_Gomez
Employee
0 0 1,781

Reimagine data utilization with neuromorphic processing.

Lewis Rhodes Labs, Inc. is shipping a novel Self-Searching Storage Array that incorporates a neuromorphic processor to search data with unparalleled speed and low power consumption, re-imagining big data utilization.

Modern companies and government organizations all require the ability to rapidly identify critical information in big data collections.  High-volume search latency directly correlates to advantages in profitability and threat protection. Existing search methods struggle to support the demand of rapidly expanding data storage.  Indexing doesn’t scale, ingest barely keeps up, and the cost to store and use data are skyrocketing.

Data lakes accommodate rapidly accumulating data volumes, but zettabytes of data are now accumulating into a post-Moore’s Law world.  As data lakes get bigger and computing advances stall, data utility plummets. When collections sit unused, data lakes become the problem instead of the solution. Extreme Search is an enterprise storage with indwelling search capacity that scales data utility on pace with data accumulation. 

Lewis Rhodes Labs implemented their high-performance, low-power neuromorphic processor on the Intel Agilex® FPGAs directly next to solid state drive (SSD) in storage.  The result is Extreme Search - enterprise storage servers with indwelling search capacity, fully scalable from TBs to PBs of storage within a distributed file system.  The resulting solution completes content searches of structured and unstructured data in minutes compared to hours with traditional methods, independent of query complexity, data type, or data volume.

Extreme Search is a disruptive solution to the challenges of modern data access and utilization.  Systems can be remotely accessed for search without moving data. Indexing is unnecessary. Real-time access to sparse target data within massive data volumes is a game-changer for time-sensitive analytics such as high-speed trading, risk analysis, compliance, fraud detection, and cybersecurity. 

Extreme Search: Optimize Data Utility for Big Data Collections

  • At the edge where high volumes of raw data sitting at remote locations are painfully under-utilized, but moving data for analysis is prohibitive.
  • For cybersecurity challenges when indications of compromise need to be identified in minutes not days, and saving months of log data as active storage is overly expensive.
  • When specialized artificial intelligence (AI)/ machine learning (ML) tasks are backlogged by the long latency of data preparation, Extreme Search combs through all the data and labels the subset of useful “needles” that optimizes your analysis.
  • For huge collections of unstructured human and machine generated data in healthcare so insights can impact outcomes.
  • For market analysis and risk assessment in FinTech, where speed of decision making directly impacts profitability.

Improving data utility and your staff’s efficiency is crucial in today’s modern enterprise and government agency.  By having the data you need, when you need it, you can make the right decisions immediately…not days or months later.

Join Lewis Rhodes Labs and Intel on March 22 for an in-depth discussion on how this technology can be used today to help you re-image your data utilization. Register Here.

LRL Logo dark.png

About the Author
Sabrina is Director of Marketing, FPGA Platforms and has over 20 years of experience serving a wide breadth of markets including data center, communications and industrial with a strong technical background in FPGAs. She has a Bachelor of Science degree in Electrical Engineering from San Jose State University.