- Data Platform – Powered by Hadoop
- All the usability features of the Fintellix Data Platform – Standard Edition with the processing power of Hadoop
- Ready to deploy and use/load compared to the large amount of integration time bespoke Hadoop platforms take
- Algorithmically optimized querying engine using relational algebra for higher throughput
- Seamless redundancy, highly resilient and infinitely scalable architecture. Built on Hadoop 2.x framework.
- Complete abstraction to the end user from the underlying complexity of data storage and processing for a seamless user experience
- ANSI SQL Compliant – Easy integration with third party BI tools such as BO and Tableau.
Banks handle and store huge volumes of data in a structured and unstructured format, and with the number of users and users transactions ballooning, the data sizes has seen exponential growth. For storing and churning this huge amount of data, the traditional approach of enterprise data warehouse system, where the underlying data store is in an RDBMS, faces many challenges. Even with the rapid advancement in the field of database technology, the additional costs and complexity to scale RDBMS, either vertically by beefing up server or resources, or horizontally by partitioning, is huge.
Hadoop to the Rescue
Using Hadoop, which primarily enables distributed computing and storage on a farm of commodity hardware nodes, scaling horizontally is a very rewarding. The redundancy support, which Hadoop provides out-of-box via replication in HDFS, also helps in building a highly reliable system. Hadoop 2.0 eliminated the single point of failure (SPOF) in the Hadoop cluster by provisioning for multiple NameNodes and introduced YARN that has the ability to process various non-MapReduce programs, making the framework much more versatile and programmable.
Apache Hadoop powered Fintellix Data Platform is built to amass and process huge amount of data. The cornerstone of the design is the use massively scalable hybrid data architecture, combining traditional RDBMS as metadata store with scale out clusters running data crunching Hadoop nodes.
The platform embeds Hortonworks Data Platform (HDP) an enterprise ready Hadoop solution.