Fwd: Austin Hadoop and Big Data User Group - Sept 23rd Meeting
Yinon Bentor
Some of you might be interested in this month's Hadoop Users Group
meeting. Last time, I found one of the tools presented quite
interesting and useful. The first talk in this month's meeting looks
like it could be as well.
I won't send any more of these out unless it is super-relevant to the
group, so if you want to stay aware of these meetings, I suggest you
subscribe to the group's mailing list here:
http://groups.google.com/group/austinhug
-Yinon
---------- Forwarded message ----------
From: Steve Watt <watt...@yahoo.com>
Date: Mon, Sep 13, 2010 at 8:42 PM
Subject: Austin Hadoop and Big Data User Group - Sept 23rd Meeting
To: AustinHug <aust...@googlegroups.com>
Hi Folks
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. Also, I will be posting
the videos from last months session to our website
(austinhug.blogspot.com) later this week.
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:
12365B Riata Trace Parkway
Austin, TX 78727
512 231 6850
This meeting is sponsored by Pervasive
I look forward to seeing everyone there !
Steve Watt