5 critical success factors for streaming data pipelines


Earlier this year, our partner, Syncsort published the results of their 2019 Data Trends survey.

A key finding; IT is grappling with data delivery and value

Only nine percent of respondents called their organisation very effective at getting value from data, with many struggling to make data accessible to users.

What is Data Streaming?

Data streaming is the continuous, high speed transfer of data at a steady rate.

This allows real time analytics to be performed against the data stream without the data having to be stored – very important when dealing with emerging huge data sets.

Technologies supporting streaming data pipelines present an opportunity to deliver data for advanced analytics, AI and machine learning applications such as fraud detection, real time shopping/recommendations, or managing financial assets based on real time changes in the stock market.

The bottom line?

The fresher the data the quicker your business can react to changing circumstances.

What are the challenges that must be addressed for enterprise data streaming?

  1. Adherence to enterprise data governance and security standards
  2. Guaranteed data delivery
  3. Integration to leading streaming platforms
  4. Support multiple environments, including on premise and hybrid cloud environments
  5. Provide agility and flexibility

Adherence to enterprise data governance and security standards

Strict data protection regulations, such as PoPI and GDPR, are becoming the norm world wide. Your streaming data solution must integrate with common data security frameworks such as SSL/TLS encryption and Kerberos to ensure the security of the data in motion.

In addition, decision making is dependent on a clear understanding of how data is sourced, and what changes have been made to it. Metadata from your streaming data solution should be preserved and presented to consumers – for example through integration with the Kafka schema registry. This helps to support end to end lineage of the streaming data pipe.

Guaranteed data delivery

By definition, streaming data is continuous flow of data that is not stored.

If connectivity fails (through a network, source, target or application server failure) your streaming CDC solution must manage this issue by automatically restarting the data stream at the correct point once connectivity is restored.

Integration to leading streaming platforms

The modern enterprise may source data from numerous environments – including the mainframe, various relational databases, big data platforms and data lakes, and emerging streaming platforms such as Kafka.

Rather than sourcing separate solutions for each, organisations should seek a single solution that can support all required source/target combinations

Support multiple environments, including on premise and hybrid cloud environments

Sources and targets may exist on premise, in the Cloud, or in any combination of the above.

The reality is that real time data streaming solutions must support a variety of application typologies including and combination of: one-way, two-way, bidirectional, distributed or cascaded.

This capability means that you will be able to cope with both current and future data streaming requirements without expensive rework

Provide agility and flexibilility

A global shortage of technical resources combined with the need for IT departments to cut costs mean that solutions that enable you to deliver streaming data pipelines quickly without specialist skills, coding or tuning are critical – particularly when working in complex organisations with multiple data environments.

How can we help?

Master Data Management partners with Syncsort – leaders in Big Data for Big Iron – to provide an enterprise strength CDC and data integration capability that delivers on all of these critical success factors, and more.

Watch the video below for a brief overview and call us on +27 11 485 8456 to set up a meeting

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.