2011 February Meeting

Our next meeting will be on February the 17th at CoSpace Austin from 6:30 - 9pm.

As usual we'll have plenty of Pizza, Beer and Tacos. This event is free and open to everyone. We have lots of people that come that are new to Hadoop and Big Data.

Agenda

6:30 - 7:00 : Meet and Greet (Austin's Pizza, Quality Beer and Tacos)

7:00 - 7:45 : "Flume" - Jonathan Hsieh, Cloudera - @jmhsieh

 
Jonathan Hsieh is a Flume Committer and the Project Lead. He'll be presenting on how you can use Flume to centrally manage the collection, aggregation and distribution of huge amounts of log data. He'll also cover the latest release of Flume (0.9.3) and conclude with a demo .

7:45 - 8:15 : "Graph processing at Web Scale with Google Pregel and Golden Orb" - Zach Richardson, Ravel - @jzachrichardson

Google has recently published a new white paper on Pregel, a graph processing runtime capable of running arbitrary graph algorithms at web scale. Zach has taken the specifications described in the paper and built Golden Orb, an Apache licensed open sourced equivalent that runs on Hadoop, and is in process to be submitted to the Apache incubator. 

Location

Directions to CoSpace are available here 

Sponsorship

This event wouldn't have been possible if it was not sponsored by the awesome folks at RackSpace and Cloudera

2011 January Meeting

Our January Meeting will be the Data Day Austin Conference on January 29th. Our user group is co-hosting the event along with GeekAustin and Riptano. See the event link for registration and location details - http://datadayaustin.eventbrite.com/ 

The Schedule is still being finalized, but you can see the speakers being added in realtime at the EventBrite link. The conference is hands on. i.e. Bring your laptops. The intent is that when you leave you will actually be able to do the things being presented. 

These skills are in extremely short supply and we want to cement Austin's position as a premiere destination for Big Data. This is why we are putting this on. Quite frankly, the amount of people available with these skills is not commensurate with the demand for them, This
is the perfect event to bring your coworkers and friends along to in order to help them become proficient in working with Big Data. 

I'm also hoping this event will be a catalyst that helps grow our local Big Data community for this coming year. 

Lastly, some Big Data events on the horizon: 

O'Reilly Strata Conference - Feb 1-3, Santa Clara, CA - http://strataconf.com/strata2011 

SXSW Interactive - March 11-13, Austin, TX (Featuring a Local Big Data Panel) - http://sxsw.com/interactive 

Data Cluster Party - March 13, Austin, TX - http://dataclusterparty2011.eventbrite.com/ 

Structure Big Data - March 23, New York City - http://events.gigaom.com/bigdata/ 

2010 November 11th Meeting

Our next meeting will be held on the 11th of November at CoSpace Austin.

This is going to be the last meeting of the year and its going to be awesome. We have 3 great sessions and loads of pizza, tacos and beer. Consider it the geek start to the festive season, especially if you're sporting a sweet Movember Stash.

This event is free and open to everyone. We have lots of people that come that are new to Hadoop and Big Data.

Agenda

6:30 - 7:00 : Meet and Greet (Austin's Pizza, Quality Beer and Tacos)

7:00 - 7:45 : "Apache Cassandra 0.7" - Jonathan Ellis, Riptano, Apache Cassandra PMC Chair - @spyced

This session will cover Cassandra and its main use cases and also provide a response to recent mentions to Cassandra deployments in the technology press. The session will go on to discuss the new features in the 0.7 release and then conclude with a demo/code walkthrough of how to actually use Cassandra in the real world, showing how to load data into Cassandra and then access it via the Cassandra API from another application.

7:45 - 8:15 : "Million Fold Mashups and the Shape of Data" - Flip Kromer, InfoChimps - @mrflip

Typically, AHUG talks focus on the tools and methods for working with massive data. I'd like to focus instead on our stock in trade, the data itself. I'll describe interesting and unexpected themes in the public and commercially available data that's out there; new mechanisms for the creation of humanity-scale datasets; and the challenges and opportunities of wiring that data up into a connected millionfold-mashup whole.



8:15 - 8:45 : "Saving lives with Big Data - the Chilean Earthquake Crisis" - Steve Watt, IBM - @wattsteve

This session will provide an explanation and demonstration of a solution our team created using the Apache Big Data stack that mines social media (such as Twitter) in the event of a Crisis (The Chilean Earthquake) allowing one to divine actionable information from your ad-hoc network on the ground and present it in a contextually relevant, human comprehensible format on a map.



Location

Directions to CoSpace are available here

Sponsorship

This event wouldn't have been possible if it was not sponsored by the awesome folks at RackSpace and Riptano

2010 September 23rd Meeting

Our next meeting will be held on the 23rd of September at Pervasive

This month's meeting focus is on Performance and Throughput for Big Data. This event is free and open to everyone. We have lots of people that come that are new to Hadoop and Big Data.

Agenda

6:30 - 7:00 : Meet and Greet (Austin's Pizza, Quality Beer and Tacos)

7:00 - 7:30 : "Realtime Classification at Scale"

Click Forensics helps large ad networks and online advertisers improve traffic acquisition and monetization. Next to working with Hadoop and very large data sets, some of our customers have strong real-time requirements. In this presentation, Click Forensics will outline how they created a real-time classifier that allows for 1 billion scoring transactions per server per day on commodity hardware with response latencies under 1 ms.

7:30 - 8:00 : "Revolutionary performance in Big Data crunching: New benchmarks using a Hadoop Appliance, and leveraging HBase and Pervasive DataRush together"

Mike Hoskins, Pervasive CTO, describes his Innovation Lab’s work on big data analytics, illustrated with Genomic and Weblog benchmarks – reflecting both absolute throughput, and green-efficiency. We’ll look at how we are combining Pervasive DataRush (delivers fine-grained, thread-level massive parallelism on multicore, with auto-scaling and simple dataflow programming model, see: www.pervasivedatarush.com), with HDFS and HBase. The result: 200-300x price/performance gains over traditional cluster implementations.

Location

Pervasive is located at (See Map):

12365B Riata Trace Parkway

Austin, TX 78727

512 231 6850

Sponsored By

Big Data Companies in Austin

This is a list of some of the companies working with or on Big Data Technologies in Austin. Contact me to be added to this list.

Click Forensics, Inc.
Technologies: Hadoop, Hive, HBase, Mahout
Application: Data mining and machine learning with very large online advertising data sets

IBM
Technologies: Hadoop, Cassandra, Zookeeper, Hive, Pig

InfoChimps
Technologies: Hadoop, Cassandra, Pig, WuKong

Pervasive Software
Technologies: Pervasive DataRush, Hadoop, Pig and Hive

Potomac Fusion - Hiring!
Technologies: Hadoop, Zookeeper, Map/Reduce, HBase

RackSpace
Technologies: Apache Cassandra and Hadoop

Riptano
Technologies: Apache Cassandra Committers and PMC Lead. Provide their own Cassandra distribution and services.

Maxpoint Interactive
Technologies: Hadoop, Cascading, HBase and Nutch

2010 August Meeting - 8/26

The August Meeting will be held on the 26th of August at CoSpace Austin.

This month's meeting focus is on tooling and productivity around Big Data and we have two special presenters flying in specifically to present to our group.

Please RSVP HERE so I can properly cater for everyone. It is free and open to all.

Agenda

6:30 - 6:50 : Meet and Greet (Austin's Pizza, Quality Beer and Tacos)

6:50 - 7:00 : Introduction to Hadoop and Big Data (Steve Watt)

7:00 - 7:30 : "Business Intelligence and Open Source"

Will Gorman (VP Engineering - Pentaho) will be presenting on the the challenges and architecture of big data analytics including the integration of structured and non-structured data. The presentation will also include a demonstration of Pentaho Data Integration’s new capabilities for Hadoop.

7:30 - 8:00 : "Techniques, Solutions and Tooling for Map/Reduce"

Abe Taha (Karmasphere) will be presenting on the challenges of writing Map/Reduce applications for Hadoop and will discuss a variety of techniques, solutions and tooling to make the developer more productive.

Location

CoSpace is located at 911 West Anderson Lane, Austin, TX which is at the confluence of Anderson Lane, Lamar, and 183. They are in Suite 203, which is upstairs, right in the middle of the retail center. Parking is free and there is ample. In back, up front, or even in the covered parking at the East end of the center.

Sponsored By

Pentaho and Rackspace

July Recap

Great turnout ! We had close to 50 people. It was a great meeting of the minds and most folks stayed until almost 9pm. If you haven't made it over yet, you are missing out. All are welcome. We now have the two presentations on video. We should have a more professional setup for recording the sessions at the next meeting.

"Graph Processing with Pregel" - Zach Richardson

"Extractiv" - Shion Deysarkar

Slides from the Presentation

Extractiv

View more presentations from Steve Watt.

There was some interest in learning how to decompose algorithms into Map/Reduce. Take a look at Jimmy Lin's (free in electronic format) book - Data-Intensive Text Processing with Hadoop

Also, the videos of the sessions from the Hadoop Summit are now available and they contain quite a few presentations on Graph processing

Thank you to our awesome sponsor

See you all at the next one !