Real time analytics with nodejs and azure
0 likes1,694 views
This document discusses using real-time analytics with Node.js and Microsoft Azure. Traditional analytics are not well-suited for real-time needs. The solution architecture presented captures data incrementally with Node.js, stores it in Azure for high-speed access, and delivers it interactively to users on any device. A demo of real-time call center analytics is shown, with cross-filtering of automatically updating metrics. The presentation aims to illustrate how this approach supports making immediate operational decisions from real-time data sources.
Real time analytics with nodejs and azure
- 1. Real-Time Analytics with Node.js and MS Azure JASON STEVENS & JON DE LA MOTTE CSG PRO INNOTECH OREGON 2013
- 2. Introductions Jason Stevens Senior Business Analyst at CSG Pro 20+ years experience in IT development, operations & management Jon de la Motte Senior Analytics Consultant at CSG Pro Specialist in data visualization and open source tools CSG Pro A leader in providing IT Solutions to the Portland community since 1993 Business Intelligence is one of our core practices
- 3. Presentation Overview Business Context: The New Reality of Real-Time Traditional Analytics: Best Fit? The Technical Challenge Solution Architecture Scenario User-participation Demo Q&A About CSG Pro
- 4. Business Context The New Reality of Real-Time Technology is continuing to accelerate the pace of business New micro-markets open constantly, to be tapped by the first to reach them Greater throughput on delivery processes has also increased the costs associated with slowdowns or stoppages The global reach of the Internet means that your customers, business partners and competitors are always active Staying on top of new opportunities and ahead of risks requires the most up-to-date information possible
- 5. Business Context The New Reality of Real-Time Business Decision-Makers are less likely to be at their desks than ever before They’re at customer sites, meeting suppliers, or on the shop floor (Or at conferences like this one!) The information that they need has to reach them where they are Real-time information isn’t real-time if you can’t readily use it until you get back to your desk Support for visualization and interactivity through mobile devices is key
- 6. Business Context The New Reality of Real-Time Real-time data sources might… Be external to your core business systems Appear as streams Have less structure than traditional relational databases Examples: Facebook Twitter Logfiles
- 7. Traditional Analytics Best Fit? Typically… Designed to integrate across many data sources, rather than a few key systems Report over relatively long periods of time, rather than hours or days May be a little stale, due to latency in ETL processes that feed them Due to data volumes, jobs may take some time to run or be batched User interfaces often designed for desktop viewing and may not be highly interactive
- 8. Traditional Analytics Best Fit? Traditional analytics serve traditional business needs well Help answer broad analytical questions Useful for evaluating past performance or planning for future performance But they weren’t designed to meet the needs of real-time business Help answer operational questions Useful for making immediate decisions Real-time analytics augment and enhance traditional BI
- 9. The Technical Challenge Making Real-Time Work 1. A lightweight, on-premise program that can incrementally capture data without impacting operational systems’ performance Optimized for real-time input (streams) or output 2. A high-speed data cache that is accessible via the Internet 3. A solid delivery mechanism
- 10. Just More Software? No, we need…
- 12. And Infrastructure to Support It
- 13. Solution Architecture NoSQL or SQL Database On-Premise Program Computers Cell Phones Tablets Modern web browsers
- 14. Solution Architecture Why use Node.js? Built on JavaScript Event driven Non-blocking input/output Built for high-performance, real-time delivery Lightweight and flexible Well suited for both the server and the worker Open source benefits Community packages Need Twitter/Facebook/Google authentication? Done.
- 15. Solution Architecture Why use Microsoft Azure? Secure integration with corporate networks and identity management Support for open-source platforms such as Node.js Proven web standards mentality Easily scalable across load, geography Straightforward code deployment
- 16. Demo: Call Center A customer’s impression of your business is often directly related to their experience when they call in – and these opinions are lasting You want to make sure that this experience is positive – and take immediate corrective action if not Modern ACDs make data available about current state – metrics like time in queue, percentage of calls that are abandoned, time to abandon, and time spent with a CSR Being able to see these data in real-time lets you improve customer loyalty and prevent expensive churn
- 17. Demo: Call Center Mock call center data Cross filtering Responsive layout Automatic data updating The “Simulate Call” fakes a call to the call center
- 20. About CSG Pro CSG Professional Services (http://www.csgpro.com) does whatever it takes to help organizations get the most value out of their enterprise data. We develop IT projects that improve business decision-making and efficiency throughout your organization, using leading Microsoft technologies to tap into your data – wherever it lies – to create maximum value.