Newsletters




Real-time Decisioning with Machine Learning


Bookmark and Share

Machine learning continues to grow with the advent of new technologies and solutions. It has emerged as a truly revolutionary technology innovation across all verticals and industries; according to analyst firm IDC, spending on machine learning will grow from $12B to $57.6B by 2021.

However, to confidently move machine learning models into actual production requires the right infrastructure that allows enterprises to perform complex analytics on streaming data in real-time, with predictable performance and low latency.

In a recent DBTA webinar, Doug Jauregui, senior solutions engineer, VoltDB, explored how intelligent real-time decisioning technology in machine learning helps operationalize analytics to take the next critical step in realizing machine learning goals.

More data has been created in the past two years than in the entire previous history of the human race, Jauregui explained.

By 2020 about 1.7 megabytes of new information will be created every second for every human being on the planet, he said. Within five years there will be over 50 billion smart connected devices in the world, all developed to collect, analyze, and share data.

The VoltDB platform can quickly use data to ingest, analyze, and overlook transactions, according to Jauergui.

VoltDB was designed to deliver blazing fast performance, enact intelligent decisions in real-time and offer enterprise-class capabilities such as:

  • The Ability to process millions of transactions per second at sustained low latency as a Stream processing solution.
  • Deploying Machine learning models on incoming streams of information to make true real time decisions with full ACID compliance and full persistence if necessary as a RTD solution.
  • Can function for edge processing or rendering as a Smart Cache solution.
  • The ability to run back and run forward for Blockchain or Distributed Ledger solutions.

By using Machine Learning the process of the rules definitions can be automated through the use of mathematical expressions that are modeled or generated from historical data that is queried for training and classification. PMML is fast becoming a standard: XML based markup to define the mathematical models.

VoltDB offers this:

  • Clustering
  • Decision Tree
  • Regression
  • General Regression
  • Support Vector Machine
  • Association rules
  • Transformations

An archived on-demand replay of this webinar is available here.


Sponsors