Fuzzy Logix

Introduction

Fuzzy Logix develops high -performance analytics solutions for Big Data. Fuzzy Logix offers in-database and GPU-based analytics solutions built on comprehensive and growing libraries of over 600 mathematical, statistical, simulation, data mining, time series and financial models.

Contents

• Introduction
• History
• Software
• DB Lytix
• Fin Lytix
• Tanay Zx and Rx
• Supported Database Platforms
• Hardware
• Industry Use
• References

History

Fuzzy Logix was formed in 2007 by Partha Sen and Mike Upchurch who met while working at Bank of America and shared a goal of making analytics pervasive. In 2008 Fuzzy Logix released DB Lytix, the first complete and commercially available library of in-database analytics. FIN Lytix was released in 2010 and was the first comprehensive library of in-database financial models. In 2010, Aperity OEM’d Fuzzy Logix models to run in their analytics and CPG software SaaS solutions. In 2011, Quest (now Dell) released Toad for Data Analyst (Data Point) that included Fuzzy Logix’s models running in MySQL. In 2012, Fuzzy Logix introduced Tanay ZX, the first comprehensive library of analytics that run in NVIDIA GPUs as well as the Tanay Analytic Appliance, a hardware and software solution for GPU-based computing. In 2013, software that allows R users to run analytics in GPUs, Tanay Rx, was released. The company was started in Charlotte, NC, USA, where their headquarters are located today. Fuzzy Logix has offices in Richmond, VA, Cupertino, CA and in the UK and India and has reseller partners in Mexico, Sweden, Japan and China.

Software

Fuzzy Logix offers four software products DB Lytix and Fin Lytix are comprehensive libraries of in-database analytic models. The libraries leverage the user defined function (UDF) capability available in database platforms. The software is available on multiple database platforms. Since data movement from the database is minimized and database platforms are growing increasingly powerful, in-database models run 5X to 100X faster than models that use multi-tiered analytics platforms. Tanay Zx and Rx are GPU- based analytic libraries that run in NVIDIA GPUs. Using GPUs to accelerate analytics can result in 10X to 1,000X improvements in performance over multi-tiered platforms.

DB Lytix

Fuzzy Logix released the first comprehensive library of in-database models, DB Lytix in 2008. The library had been under development since 1998. The library includes mathematical, statistical, data mining, simulation and classification models.

Fin Lytix

Fuzzy Logix released the first comprehensive financial library FIN Lytix, in 2010. The library contains models for equity, fixed income, foreign exchange, interest rate and time series models that are used by the financial services industry for risk management, pricing and portfolio optimization.

Tanay Zx and Rx

In 2012, Fuzzy Logix released the first comprehensive library of GPU-based analytics designed to run on NVIDIA GPU Tesla and Kepler hardware. Tanay Zx contains for mathematical, statistical, data mining, simulation and classification models as well as equity, fixed income, foreign exchange, interest rate and time series models. Tanay Rx allows R users to call the models in Tanay Zx by writing R code. This allows R models to run using GPUs.

Supported Database Platforms

Aster Data, Informix, Netezza, IBM PureData Systems, MySQL, ParAccel, SQL Server, Sybase IQ and Teradata.

Hardware

The Tanay Zx Appliance is a high-performance computer that contains one or more GPU cards. It is available in either rack mounted or server configuration that includes either the Tanay Zx or Rx library.

Industry Use

Fuzzy Logix solutions are effective in optimizing business process performance by utilizing mathematical modeling based risk management. Industries like Marketing, Healthcare, Insurance, Digital Media services, Financial Services (Investment and Retail Banking, Brokerage Houses, Stock Exchanges, Hedge Fund Management) are some examples. Same techniques and superior performance can potentially be utilized much more broadly for solving complex problems in other industries and organization (Government programs, Educational institutions, Research) when there is a need for running Analytics on Big Data using complex models. Solutions are derived from Predictive modeling of behavior in assessing risk and modeling an optimal system.