The world demands us to be available 24/7 and businesses are asking the same of their analytics. Apache Hadoop, Spark, Kafka, Kudu, and others are modernizing the data platform, providing real-time analytics and insights.
Shant Hovsepian, CTO and co-founder of Arcadia Data, discussed strategies for enabling a real-time enterprise during his session, titled “Four Ways to Scale Interactive Analytics and BI for Real-Time Insights” at Data Summit 2017.
Real-time visualizations are used in critical businesses, Hovsepian explained. However, no one has time to sit and stare at dashboards.
There are two key areas where enterprises rely on real-time information. This includes real-time alerting in industries such as cyber security and financial services, and monitoring data in real-time, such as online advertising or fraud detection.
A few examples of different technology offerings in this area include Kafka, Amazon Kinesis, Google, Spark, Flink, and Amazon Lambda.
Enterprises should invest in streaming first architecture to capture the most in real-time. All data comes in streams with message passing built in. Streams can also be replayed.
True real-time applications include reactive machine learning with Spark streaming, for example, it can identify suspicious network activity as it happens and alert the right people immediately.
“Truly streaming architecture will have everything in place to pull the human element in,” Hovsepian said.
Many conference presentations have been made available by speakers at www.dbta.com/datasummit/2017/presentations.aspx.