Airmettle Logo
TM

AirMettle®

Accelerate Big Data Analytics by 100X or more

About AirMettle
AirMettle develops solutions that simplify and accelerate Big Data analytics by orders of magnitude — while reducing analytics memory, storage, compute, and networking costs. Our patented technologies integrate highly-scalable parallel processing to select, group, aggregate, and format data in the data lake — rather than moving vast amounts of data to analytic memory or a data warehouse for basic processing. And, we don't require exotic hardware; AirMettle works with industry-standard servers, storage, and networking solutions.

Analytical Data Platform:

AirMettle's Analytical Data Platform enables users to quickly obtain key insights from large data sets (records, multi-dimensional data, images, video) by analyzing the data in the storage tier. Our patented technology integrates massively parallel processing (including AI inferencing) of distributed data within a software-defined storage framework—accelerating analytics by orders of magnitude. By processing data in place, the platform eliminates the need to transfer large datasets to separate compute clusters for analysis, significantly reducing network traffic and latency, as well as analytics compute, memory, and storage requirements—making real-time analytics practical even at petabyte scales. For further information, the AirMettle User Guide is available for download, or contact info@airmettle.com

Hierarchical Multi-Dimensional Histograms (HMDH):

AirMettle’s Hierarchical Multi-Dimensional Histograms solution enables high-performance exploratory analysis of large, complex datasets. Our solution uses parallel processing of stored data to rapidly characterize floating-point and integer data distributions in up to four dimensions in a single pass, efficiently stores the results, and provides and API for advanced querying (e.g., range queries, percentile calculations, etc.). HDMH files are compact “fingerprints” containing statistical properties of massive datasets—including correlations between parameters. These compact histograms are typically over 1,000 times smaller than the raw data, while retaining precise counts and addressing at each hierarchical level. Histogram files also can be compared, facilitating the analysis of variations across datasets. HMDH has broad application in research, manufacturing, finance, and other industries. For further information see our post on LinkedIn or contact hmdh@airmettle.com

Patented AirMettle solutions:

  • Accelerate petabyte-scale data analytics by orders of magnitude to speed time to insight
  • Reduce network congestion, lower compute costs, and lessen — or even eliminate — data warehouse requirements.
  • Support a range of analytic solutions using industry-standard S3-compatible interface
  • Integrate with open source tools like Apache Spark, Presto, and Arrow.
  • Work with industry-standard servers, storage, and networking solutions.

National Science Foundation

Real-Time Smart Data Lake
Read More...

National Oceanic and Atmospheric Administration

Accelerated in-situ analysis of multi-dimensional data
Read More...

Multi-Dimensional Demonstration

Department Of Energy

Accelerated In-Storage Data Mining of Light Sources
Read More...

Platforms

Built on AWS and Oracle Cloud platforms

 Cloud-first

Selected member of the AWS Activate program

On-premises

Developing on-premises AirMettle solutions as a member of the NVIDIA Inception program.

Database

Multi-dimensional Data

AI Image Processing

Data Storage Science

DSS is a certified woman owned small business (WOSB) prime contractor for U.S. government projects.
ds-science.com

Aeon Computing

Aeon designs and deploys highly-integrated HPC, clustered computing, and storage solutions for computer-aided research and production.
www.aeoncomputing.com

iXsystems

iXsystems is a storage System Integrator and the developer of TrueNAS Open Storage.
www.ixsystems.com
Data Landscape
Future direction arrow

The future of data-centric business.

Data generated by business operations, infrastructure, and research, as well as commercial products and services, streams across the internet and internal networks into storage. This rapidly growing data comes in many forms, and increasingly is stored as very large objects in data lakes. Currently, most organizations report that they either already have implemented some form of data lake, or are actively evaluating the technology. Regardless of industry, companies that have implemented data lakes (and associated analytics) are outperforming their competitors.
magnify analytics

Challenges of insight

Over 80% of the execution time for analytics applications is spent on ETL (Extract/Transform/Load) and basic parsing tasks to find and extract relevant data sub-sets, rather than on processing the query itself. Moving large sets of raw data to the analytics platform for these basic operations is complex, expensive, and slow — delaying the adoption of transformative AI solutions, hampering the flow and analysis of scientific information, and hindering enterprises from gaining all the value possible from their big data assets.
The AirMettle Paradigm

We change the fundamental relationship between data analytics and storage

AirMettle's core innovation is integrating distributed parallel processing into a software defined storage platform to enable analysis of semi-structured content in storage. The value gained is greatly accelerating analytics while significantly reducing network and compute requirements.

AirMettle can transform disparate data formats into exactly those required by analysis tools, substantially reduce network traffic, and decrease the amount of expensive memory needed by higher-level analytics applications—dramatically accelerating all analytics from storage, especially in petabyte-scale environments.
Team
Donpaul Stephens
Best known as the founder of Violin Memory where he developed the original concept; hired and led the core team; raised venture capital from strategic, institutional, and individual investors; established key strategic partnerships in both supply chain and go-to market partners; and developed multiple accounts.
Donpaul Stephens
Founder and CEO
Matt Youill
Best known as one of Betfair's founding engineers where, as Chief Technologist, he built the platform that grew into one of the world's largest online gaming companies. Enabling this growth was an internally developed database that processed more transactions than the combined European stock exchanges and catalyzed the launch of Betfair's own financial exchange LMAX.
Matt Youill
Founder, Analytics
Chia-lin Wu
Bringing over 20 years of experience designing and implementing large scale consumer-facing products and services for organizations such as the US PTO and AOL. She is one of the principal developers of the AirMettle distributed object storage system.
Chia-lin Wu
Founder, Object Storage
Troy Trenchard
Brings executive management of products, marketing, and sales in startups to Fortune 100 companies — including Cisco, Citrix, Centurylink, and Sonatus. Over 30 years of tech experience defining, developing, managing, launching, marketing, and selling a wide range of innovative, customer-centric tech solutions and services for Enterprise, SMB, Consumer, and Service Provider customers globally.
Troy Trenchard
Chief Product Officer
Josh Fuhs
Josh Fuhs
Software Engineer
Neal Ekker
Former VP of Business Development for Texas Memory Systems, where he developed a comprehensive marketing and channel program which enabled one of the innovators in the solid state storage industry to double revenue in just a few years. Developing technical marketing collateral to facilitate AirMettle’s growth.
Neal Ekker
Business Communication
Ernest Villafana
Ernest Villafana
Software Engineer
Mohit Anand
Mohit Anand
Software Engineer
Rupin Jaira
Rupin Jaira
Software Engineer
Katsumi Kato
Katsumi Kato
Sales Rep, Japan
Junjie Hua
Junjie Hua
Dhyakesh Soundararajan
Dhyakesh Soundararajan
Address
AirMettle, Inc.
2700 Post Oak Blvd., 21st Floor
Houston, TX 77056
Say Hello
info@airmettle.com

Scroll Up
 Previous  All works Next 
×
Placeholder