Founder Showcase: PAF’s Smart Deployment and Uber Performance

June 14, 2011 at 9:38 am 2 comments

By Terence Craig

I am going to be attending the Founders Showcase in San Francisco this Wednesday.  We were looking for a warm-up event before we put the company back in fundraising mode, and thought that this venue would be a nice low key but quality event to get back in the groove after being completely focused on all things product for a while.  Well, it’s definitely going to be a quality event but low key it’s not.  Not only was the event sold out but the demand was such that they offered more tickets, moved to a larger venue, and that venue is now sold out!  The same thing happened at the Mongo show that I spoke at a couple of weeks ago.

I am excited about this event. It involves three of the guys that have changed the way the venture game is played for the better (see my previous post).  Adeo Ressi is the organizer, and the keynotes are going to be delivered by Mark Suster of “Both Sides of the Table” and the man who is turning seed funding on its ear, Naval Ravikant. It will be a great opportunity to see what other entrepreneurs are doing, meet some partners and future customers, and show off our technology.

Speaking of technology, the show couldn’t come at a better time – we are going to show publically for the first time two of our most significant platform upgrades to date.

The first is the most flexible deployment model we have seen in Big Data Analytics. The PatternBuilders Analytics Framework (PAF) now supports 4 deployment modes (mentioned previously here) each designed to meet a unique set of customers needs:

Deployment Type Cloud On Premise Hybrid Dedicated Colocation
Description PatternBuilders deployment is located entirely in the cloud and maintained by PatternBuilders staff. PatternBuilders deployment is stored on premise and maintained by internal staff. A PatternBuilders deployment that contains all metadata that would make the numeric analytics data useful, is kept on premise at the
customer data center but the numeric analytics results are stored encrypted in the cloud or at a dedicated data center.
PatternBuilders deployment is kept at a PatternBuilders colocation facility with dedicated hardware.
Customer Benefits Cheapest and most scalable model for a customer. Complete control over their data and regulatory or legal compliance. Complete control over their data, regulatory or legal compliance, all without requiring the overhead of maintaining the hardware
infrastructure required for Big Data.
Best performance with no investment in an IT infrastructure.
Customer Type SMB, pilots, companies with a lack of available IT resources. Security conscious customers with experienced IT staff and a high quality data center. Security conscious customers constrained by lack of IT or hardware resources or that simply want to focus on their core competencies. Customers who need the maximum performance but want to keep IT investment down.
Example Industries/
Sectors
SMBs, Social Media Financial Services, Medical, Regulated industries, Defense Any Hedge Funds, Retail Suppliers

The second feature we will be highlighting at the show is linear fan out performance of our analytics engine.  Our analytics servers now scale linearly, both in the cloud and on dedicated hardware.  Slowdowns caused by calculating results or generating hierarchical indexes are a thing of a past. For example, on a single Rackspace cloud server we are performing analytics on one of our demo domains that features 8 years of intraday stock ticker data in a mash up with weekly petroleum prices.  Even with just the one server, the performance is pretty impressive at 2,000 analytics  created per second.  Adding a second and third server effectively doubles and triples the values respectively (4,000 and 6,000).  This speed up continues as you add machines. In our colo, on dedicated machines, the calculation performance for one server approaches 9,000 analytics results per second. Just as in the cloud, as we add machines, the numbers go up accordingly – 18,000 results per second on two machines, 27,000 on three.

Don’t take our word for it – come by our table (we are in the silver section) and let us show it to you running live. We think PAF’s new UI,  scalability, and easy customization is going to blow a lot of people away on Wednesday and if not, would like to know why so we can fix it.   Even if you are not into “Big Data” stop by and say hi.  We love to hear about what other folks are working on.

Entry filed under: General Analytics, Technology. Tags: , , .

Data Security Reminder: Strong Passwords are Your Best Line of Defense McKinsey Study: Some Great Stats to Chew On

2 Comments Add your own

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Trackback this post  |  Subscribe to the comments via RSS Feed


Video: Big Data Made Easy

PatternBuilders Corporate

Follow us on Twitter

Special privacy section!

Enter your email address to subscribe.

Join 61 other followers

Recent Posts

Previous Posts


%d bloggers like this: