Blogs

Post a Comment

Github, StreamsDev and Quick Start–Oh my!

July 9, 2014

A scholarly article published jointly by professors from Brown University and the Massachusetts Institute of Technology highlight eight requirements for real-time stream processing platforms. The authors discuss the importance of keeping up with fast data. This is challenging because streaming data is full of imperfections. The data is sometimes delayed, missing, out of order and highly complex, without firm governance or tight standards. The value of streaming data is enhanced when it can be analyzed with data at rest and transferred while maintaining privacy.    

Datagram_v17-070214.jpg

To meet fast data requirements, InfoSphere Streams was launched well ahead of the big data craze in 2009 to give organizations the ability to:

  • Respond quickly to events and changing requirements
  • Continuously analyze data at rates vastly greater than existing systems
  • Adapt rapidly to changing data formats and types
  • Manage high availability, heterogeneity and distribution of the new stream paradigm
  • Provide security and information confidentiality for streaming data that is shared

Today, more than five years after the launch of InfoSphere Steams, the market for stream computing is really heating up. It is the next generation of big data technologies and technology giants such as Google and Amazon, as well as smaller players like DataTorrent, have launched stream computing offerings. Open source offerings including Apache Storm and Apache Spark are also available. 

How do these vendors stack up? Forrester Research is currently evaluating the market opportunity and will deliver the first ever Wave for Streaming Analytics in 2014. The International Technology Group (ITG) recently compared the cost, benefit and risk of InfoSphere Streams with Open Source Options. Other analysts such as Enterprise Management Associates and The 451 Group have evaluated InfoSphere Streams as a leader and front runner in the market.  

But, you will want to evaluate InfoSphere Streams for yourself. You will also want increase your skills in stream computing to get ahead.

  1. InfoSphere Streams Open Source Project on GitHub: IBM has decided to create an open source project for some Streams components to speed development of applications and harness the energies of the development community. In future releases, IBM expects to incorporate new function from the projects into Streams.   We believe a mix of open source and closed source is the best way to drive adoption in the marketplace, as seen by the success of open source offerings like Apache Web Server and Eclipse. The chief architect for InfoSphere Streams, Mike Spicer, recently spoke on a podcast about his vision for this project.
  2. The StreamsDev Community: A place for developers by developers. It is your direct channel to the Streams development team and a place to discuss, learn and share ideas. Join now and engage across five key areas: documentation, downloads, help, blogs and events.
  3. Quick Start Program provides a free, downloadable, nonproduction version of InfoSphere Streams that lets you explore and evaluate real-time analytics from IBM. It is ideal for companies that want to experiment with real-time analytics or extend existing capabilities. Review the Quick Start Brochure, download and go.
  4. Big Data University courses make stream computing manageable. Block a few hours on your calendar and participate in some no charge education including a Stream Computing 101 course complete with exercises and quiz questions. If you get stuck, post a quick question on StreamsDev!
  5. The Performance BenchMark is perfect for developers who want to understand the underlying architecture of InfoSphere Streams. The benchmark is a direct comparison of Streams and Storm, in a real world use case of email classification. The benchmark study found that InfoSphere Streams outperforms Apache Storm by 2.6 to 12.3 times in terms of throughput while simultaneously consuming 5.5 to 14.2 times less CPU time. 

And we aren’t stopping with this top five list—right now developers can join the early access program for our cloud based offerings: the Streams platform as a service is available for limited use on Marketplace and a Geo-Spatial Service powered by InfoSphere Streams is now available on Bluemix.

How will you engage? Tell us your favorite stream computing resource to add to the list.