Big Data Everywhere Event (12/08/2016 Los Angeles)

Invitation: 
Subhash Gowda, my colleague at IPS Group Inc., and I were invited by MapR Technologies to present our use case at the Big Data Everywhere event hosted and organized by MapR that was scheduled on December 08th 2016.

Event Schedule:
8:00am -- Registration and Breakfast
9:00am -- Introductions and Welcome (Brian Gates, Regional Sales Director, MapR)
9:15am -- The Keys to Digital Transformation (Jack Norris, SVP of Data and Applications, MapR)
10:00am -- Extending Open Source Big Data to the Enterprise with SAP HANA Vora (Daniel Rutschmann, Sr. Director, Database & Data Management Product Strategy, SAP)
10:30am -- Morning Networking Break
11:00am -- New Designs in Data Platform Reference Architectures (Ramesh Balasubramanian, Search and BigData, Qualcomm)
11:45am -- How Enterprises Are Using IoT Analytics to Drive Competitive Advantage (Daniel Gutierrez, Managing Editor, insideBIGDATA)
12:30pm -- Networking Lunch
1:30pm -- How to Manage Continuous Dataflows for a Customer 360 Application (Kirit Basu, Director, Product Management, StreamSets)
 
2:00pm -- Re-Platforming for IoT Scale (Subhash Gowda, Manager Web Applications & Analytics and Ashfaq Chougle, Software Development Manager at IPS Group)
2:45pm -- Running Large-Scale Big Data Platform at Rubicon Project (Devin Kramer, Director Infrastructure Engineering and Jesse Escobedo, Senior Systems Engineer, Rubicon Project)
3:30pm -- Event Ends

Preparation: 
Well we prepared our presentation got it blessed by marketing and management and submitted the final copy to the event organizers. Subhash had blessed the presentation with a wow factor, the IPS marketing video. It becomes very easy to give the company background with a video then speaking about it.

Drive to the event: 
MapR had provided us with the option of hotel stay, but we had our holiday party the earlier day so Subhash and I decided that we will leave early and drive to the event on the event day. Also, we had asked the organizers to schedule our session late so we will be there on time before time. As planned, we left early, dodged the traffic using the car pool lane and reached the event place that is Loews Hotel in Santa Monica at around 10:15 AM. We missed couple of sessions that were scheduled before the networking break, but we were able to do some networking in the break.

The Event:
The event ambience was wonderful, had also loved their earlier session which Subhash and I had attended in San Diego few months back. The audience present looked more interested in the Big Data and Hadoop technology.

For us the first event was 11:00 am one, which was New Designs in Data Platform Reference Architectures. It was an awesome session on how Qualcomm moved from document db to MapR-DB and how they were able to use elastic search to optimize searching over Mapr-DB. Also, it was good to know the enormous volume of data Mapr-DB was able to handle.

Next session at 11:45 was also an eye opener; talking more on the analytical trends going on in the market and more specific to IoT industry. We being from IoT industry, the session was good insight on what analytics IoT startup should look at and how they can implement it to have an edge over competition.

Network while lunch, is awesome. You get to speak with new people, get to share ideas, get to learn more.

After lunch break the session at 1:30 pm was another good insight on using Stream Sets to fetch data from multiple sources and how Stream Sets can help visualize and debug the data flow. I had earlier used Stream Sets, for implementing our change data capture (CDC), but we could not get thru, as Stream Sets release we were using around June 2016, had no support for Mapr-DB destination connector configured for cluster use. While discussion with Kirit, the Stream Sets presenter for the event, Kirit acknowledged that the recent release has Mapr-DB destination connector configured for cluster use. Anyways we had already created our own CDC. This tool is coming along good and looks really promising.

Next was our session. We were presenting our use case of how we are re-architecting our IoT infrastructure here at IPS Group Inc. I will be discussing this more in detail in my other blog on www.ToSqlOrNoSQL.com. I will update the link once its complete. Subhash kicked of with the wow factor the marketing video, he presented the business side of the use case, then I picked up with technical side of the use case. The whole session was awesome, could see the interest of people in the use case. Of course, it is expected that we will have more of this use cases where companies will be re-platforming and re-architecting their data model to fit the larger volume specially start-ups. The start-ups normally, for faster implementation build the RDBMS model, and once the volume of data is out of control start looking for DB models that can easily handle larger volume.

Few pictures:

I also enjoyed the question answer session. I was expecting a question asking me an example as to how it could fit into their data model. Also, another question about the data model mind set change. Well frankly, the Key-Value model looks really promising to me and that too Mapr-DB supporting more number of column families makes it very easy to store in de-normalized form.

After our session, I took a break for prayers and then rejoined it for the last session. The last session running large scale big data platform was another great session, where the company was running around with a cluster of 600 nodes. It was a interview kind of session where they were presenting their case in the form of interview. I missed some part of it but still the remaining part was packed with lots of info.

At the end of event it was really exciting to speak with more new people and discuss their ideas and use cases and give few recommendations.

I will be speaking more on the questions and my responses in the ToSQLOrNoSQL.com blog.

Seriously, looking for more such events!