Since Gartner coined the term HTAP in 2014 to describe a category of databases, both the technological innovation and market need for them have grown. HTAP, or Hybrid transactional/analytical processing, defines databases that can perform both transactional processes (OLTP) and analytical processes (OLAP). The integration of OLTP and OLAP in a single database provides a more straightforward alternative to traditional approaches that try to utilize relational databases such as Oracle, DB2, SQL Server, MySQL, and Postgres, alongside data warehouses such as Teradata, Netezza, and Vertica. The rise of HTAP is paving the way for modern, adaptive applications that can benefit from the lower latency of a hybrid approach.
Additionally, the analyst firm Forrester has created a similar term that they call Translytical Processing, defining a single database for transactions and analytics.
Benefits of HTAP Databases
In a recent report from Gartner the “Market Guide for HTAP-Enabling In-Memory Computing Technologies,” the authors detailed several main benefits; the most prominent being that data does not have to be transferred from an operational database to a data warehouse for large analytical jobs. This feature allows transactional data to be readily available for use with historical data when analytics jobs are run. It also reduces, and can even eliminate, the need for multiple copies of the same data. With no need for a data warehouse, users can bypass the Extract, Transform and Load (ETL) process and save themselves hours or days’ worth of time that would have otherwise been spent waiting for data to transfer.
Uses of HTAP Databases
HTAP can be used to add machine learning, real-time operational reporting and dashboarding, and reactive capabilities to new or existing applications. These more intelligent applications are always on, can react to new data as it streams in, and can even improve and adapt as they receive more data and run more workloads.
HTAP use examples include:
- Clinicians can be alerted to dangerous conditions for patients from always-on, reactive healthcare applications that analyze Electronic Medical Records (EMR) in real time
- Marketers can create programs that learn and adapt to the previous buying behavior of site visitors as well as their last three clicks
- Financial services companies can detect real-time fraud based on purchasing history and real-time transactions
The advanced foundation that HTAP provides allows these applications to automatically improve over time with more data. By integrating machine learning models with HTAP, applications can become adaptive.
Splice Machine provides HTAP capabilities at a more affordable price than legacy scale up systems, but with less operational complexity than other open source offerings. Built on scale-out hardware and soon to be offered as a cloud service on Amazon AWS, Splice Machine eases the complexity involved in creating adaptive applications that enhance your business.