Experts Discuss What Capabilities are Needed to Succeed with Real-Time Analytics

The ability to quickly act on information to solve problems or create value has long been the goal of many businesses.

However, it was until recently when new technologies emerged that the speed and scalability requirements of real-time analytics could be addressed both technically and cost-effectively by organizations on a large scale.

Today, close to half of DBTA subscribers are seeking real-time data capabilities to support a wide range of use cases, from more timely dashboards and reports, to fraud detection and IoT edge decision engines. 

DBTA held a roundtable webinar with Jamison Shaver, senior director, product management, Swim; Rob Hedgepeth, director, developer evangelist, MariaDB; and Rick Negrin, VP, product management, SingleStore, who discussed the key capabilities for succeeding with real-time analytics today.

Disruption accelerates digital transformation, Shaver said. Opportunities are created by organizations that continuously generate insights and actionable responses from real-time data streams.

Traditional data architectures fall short, with “store-then-analyze” approaches limiting real-time insights. Continuous intelligence is needed, according to Shaver.

Organizations need to continuously ‘listen to’ changes in data and project outcomes, process and analyze historical and streaming data in context, and always have an answer—storing raw data for later analysis is too slow. Enter Swim, a platform that can help users achieve continuous intelligence, Shaver recommended.

Swim offers the first open core, end-to-end enterprise platform for building and running continuous intelligence applications at scale, Shaver said.

Hedgepeth outlined the various challenges businesses are facing such as:

  • Applications have transactional and analytical queries: Constrained to limited, lightweight analytics that need full analytics to create competitive features
  • Applications with lots of customers, lots of transactions: Limited to current or recent transaction data (months) and need access to all historical data (years)
  • SaaS customers are becoming data-driven organizations: They don’t have access to their own data and they need to analyze it in unknown/unexpected ways

He suggested companies use the MariaDB platform. The solution provides versatility for transactions and/or analytics, relational and/or document, to any workload. MariaDB offers scalability, whether replicated and/or distributed on any scale. And the platform is a general-purpose database which can be configured to specialized database engines at any time.

Demands from data are growing... Yet, systems are struggling to keep up, Negrin said.

Companies need pure speed for fast transactions and analytics; fast ingestion for the high speed parallel load of data from a variety of sources; high concurrency with no locking and easy scale; and high reliability that meets strict availability SLAs.

SingleStore offers a cloud-native operational database built for speed and scale, Negrin said. SingleStore is the ideal database when you must deliver “Analytics with an SLA.”

SingleStore supports the creation and operation of models at scale across streaming and historical data. And SingleStore makes it easy to adopt a modern, cloud native approach to working with data.

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