Potential Markets for Plura

As mentioned in our last post, we’re focusing on customer acquisition for the start of 2009.  To give you an idea of what types of customers we’re looking at, I thought I would give a quick rundown of what we’re currently working on or exploring:

Web-scale analysis: Applications that utilize the bandwidth on our nodes fit Plura very well.  This is a feature of Plura that tends to get glossed over, but is probably the coolest thing we can provide.  On top of the compute power our nodes have, they also have their own bandwidth.  Any application that uses this bandwidth gets access to thousands of separate Internet connections, which can be used to access the web in a myriad of ways.  One of our sister companies, Computational Crawling, is developing a platform that takes advantage of this feature.

Financial modeling: Although a lot of financial algorithmic trading relies on low latency, there are also a wide array of possibilities for more robust modeling.  For example, some modelers utilize genetic algorithms to create sophisticated models.  Such techniques fit well on Plura due to the low data:compute time ratio inherent to these algorithms.  One of our customers is developing such a package.

Bioinformatics: We’re really interested in exploring bioinformatics more.  Based on some brief research, there seems to be a lot of applications in this field that utilize parallel computing.  In particular, we’ve looked at BLAST and similar algorithms.  It appears that the BLAST algorithm can be broken into parallel components.  Unfortunately we haven’t looked at this field enough to know for sure, but it’s something we may consider in the future.

Oil and Gas Exploration: While some seismic analysis algorithms will not work on Plura, there are some that will, such as Kirchhoff migrations.  Developing a seismic implementation in Plura will take a good bit of work, but is a definite possibility.

In general, any application that uses a significant amount of compute time relative the data transferred or any application that can is embarassingly parallel will work very well on Plura.  We’ve mentioned just a few potential industries that can utilize Plura, but there are several more on top of these.  We’ll be looking for customers in one or more of these sectors in the coming year.

Happy 2009!

Happy new year to all our readers!  2008 was a big year for Plura.  Last year we managed to accomplish a lot of things, including:

• Proving Plura’s underlying technology and business model
• Acquiring several affiliates, including a few very popular sites and applications
• Getting featured on GigaOm and HackerNews
• Growing the size of our computing power to about 50,000 nodes

2009 figures to be an equally exciting year for us.  Since we’ve been able to flesh out a lot of our business on the affiliate side, we’re going to focus more heavily on existing and future customers.  Two of our beta customers will begin full-scale use of our computing power in the next couple of months.  We also plan to start increasing awareness of Plura to other potential customers and will be exploring various markets, such as oil and gas exploration, bioinformatics, and financial modeling, in this effort.  Of course, we will also be seeking out various non-profits that are in need of computing power for research projects.  If all of this goes to plan, we’ll be looking to expand the size of our compute network to accommodate the extra demand – possibly increasing our node count to 100,000 nodes or more.

We’ll keep the blog updated with news of our progress.  We hope everyone’s new year is fruitful and prosperous!

Plura awarded “Most Promising I.T. and Web 2.0 Company” at Rice Alliance Venture Forum

As mentioned in a previous post, we participated in the Rice Alliance I.T. and Web 2.0 Venture Forum this past Thursday.  I had the pleasure of representing our firm in the elevator pitches, where I gave a 90-second description of our business.  Based on this pitch, we were awarded “Most Promising I.T. and Web 2.0 Company”, along with 10 other firms (out of 55 total).  The event as a whole was a great experience.  We met a lot of interesting folks and got some great feedback on our business.

If you’d like to hear the pitch, you can do so here.

Private Beta Status

Thank you for your support during our private beta.  We’ve been overwhelmed by the interest in Plura.  At this point, we’re only accepting a portion of our affiliate applications into our private beta.  We’re trying to find unique affiliates that are providing different nodes than what we’ve already seen.  If there is something unique about your site, application, or users, then there is a much higher chance that you’ll be accepted into the private beta.  We are particularly interested in different types of desktop applications that can integrate with java and affiliates that have very high engagement sites (sites where users stay on a page for a long time).

Of course, for those affiliates that have not been accepted into our private beta, please be assured that we will remember you when we enter our public beta in the coming months.  If you want to think ahead about what will make your site earn more Plura revenue, please follow the tips here.

SuperComputing 2008 Conference

Five of us from Plura attended the SuperComputing 2008 conference in Austin.  Although we did not set up a booth, we still had a great week and talked to lots of potential customers and vendors about Plura and our coming infrastructure needs.  To our new friends, we hope you enjoyed the show as much as we did and had a safe trip home.

I’ve attended several SuperComputing conferences in the past and it is interesting how it has evolved.  Two years ago, GPUs and FPGAs were exploratory.  This year, they are everywhere.  Many people are predicting that the current GPU focus is temporary until Intel and AMD start embedding many smaller cores on the chip (see Intel Larabee for Intel’s version).

GigaOm makes a good point that SuperComputing is more and more about software.  Everyone knows how to build large clusters these days.  The software and infrastructure to tie them together to solve meaningful problems is the critical piece.  We’re working hard to solve real problems that are difficult on traditional infrastructures.

Also, one quick brag.  The new Top 500 list of fastest SuperComputers was released during the show.  Even in our limited private beta state, our current Plura capacity would place us in the top 10 SuperComputers in the world based on GFLOPs.  We’re pretty proud of this given that our private beta started less than a month ago.

Plura chosen for Company Showcase at Rice Alliance I.T. and Web 2.0 Venture Forum

We’re happy to announce that Plura was selected to present at the Rice Alliance I.T. and Web 2.0 Venture Forum.  As part of the forum, Plura will be giving a 90-second elevator pitch and showcasing our company at a booth.

Having to been to other Rice Alliance events in the past, it’s quite exciting for us to actually be presenting at one, albeit for a short time.  The audience at this event will probably contain several potential affiliates, customers, and other people who may be interested in Plura.

The event will be held at Rice University on December 11, 2008.  We encourage everyone interested to come!

Welcome New Private Beta Affiliates!

Our private beta officially opened this week!  We’ve added quite a few new affiliates this week from our various coverage.  Thanks for giving us a shot.  We’re working hard to make Plura a great way for you to increase revenue from your site.

We’d like you to know several things about Plura:
1) Payments – We are paying our affiliates by PayPal or check (your choice).  We pay once per month if your accumulated revenue is over $50.  We are targeting payments for the 5th day of each month right now (e.g. for October 1-31, we will pay by November 5).
2) Affiliate Management Portal – The AMP lets you control most of your account information.  It also lets you see your daily revenue totals.  It lags by 48 hours (e.g. earnings from October 27 are posted by midnight on the morning of October 30).  (edit: this originally read 25 hours)
3) Improving your revenue – Every affiliate wonders how to increase his Plura earnings even more.  We have a few tips

• If you are using website integration, make the Plura <iframe> visible.  Some versions of IE will not start java applets if it is invisible.
• If you have pages that get refreshed often, try using AJAX techniques to reduce refreshes
• If you are providing a service to your customers at no charge, ask your dedicated customers to leave your page up while they are using their computer.  This gives them a very nice way of making a donation to you without actually sending money and shouldn’t harm their typical computer experience.

4) Target sites – You will earn the most revenue from Plura if it is placed on high engagement sites where users linger or interact with a page for a long time.  Games, streaming audio/video, certain widgets, fantasy sports applets, or AJAXy sites that don’t do many page reloads are all great examples.  Your earnings per hit from typical static web content will not be very high unless you have people park on the page.

Featured on GigaOm!

We’re very excited to say that GigaOm has featured us in an article!  I had been talking with Wagner James Au, one of GigaOm’s contributors, for the past week or so about Plura after contacting the site in response to another article on monetizing casual web games.

Plura is a great fit for web games for a couple reasons.  First, it’s really easy to embed Plura on a page using website integration, which will give you static control over Plura, and with Flash or Java games using our other integration methods.  Second, because people tend to stay on a game site for several minutes, these sites can earn a significant amount of revenue.  If a site has about 10,000 players on it any given time, it can make up to $26,000 each month.  This figure scales up with the # of concurrent users (100,000 players => $260,000 and so on).  Finally, Plura revenue is incremental and non-competing to other revenue sources.  Games don’t have to replace ads or micro-transactions.

We’ve got a nice burst of interest from potential affiliates following the GigaOm article, which was pretty nice   We’re pretty excited here at Plura since we seem to be picking up momentum each week – let’s hope it continues!

Example applications using Plura

Plura is a unique blend of processing power and bandwidth.  It is not quite a cluster and has features that make it unique from grid computing projects like Folding@home.

I’d like to give some examples of problems that are being solved with Plura today.  I’ll post more details about these in the future:

Stock Market Simulations – Quant R&D is using Plura to analyze the stock market using an all-vs-all data strategy.  Quant’s problem boils down to wanting to run complex simulations on pairs of stocks from the entire stock market.  Each simulation is relatively expensive, but it can be broken up into pieces.  Quant uses 1-minute stock market data and one stock’s data over the time period they analyze is about 1MB compressed.  To stay within Plura node memory limitations and to minimize bandwidth needs, they want to reuse the stocks as much as possible.  So, they use Plura data groups to ensure Plura nodes get to reuse the stock data they have already downloaded as often as possible.  This means that a typical Plura node will download two 1MB pieces of data and will then proceed to work on WUs (work units) using those two pieces of data for quite some time (probably longer than the life of the node).

Custom Web Crawling – 80legs is using Plura to do distributed web crawling.  Rather than having data centers with very fat pipes, they use a portion of the bandwidth of the Plura nodes to crawl the web.  In order to improve the success ratio for each work unit, 80legs sends out Plura WUs with very few URLs to crawl.  The economics of this are incredibly dramatic compared to data center bandwidth.

Prime Number Search – As a sample application, Plura has created a distributed prime factoring engine that is doing pre-factoring for large Mersenne primes.  We are using this application to demonstrate a different data model for Plura.  This model has no specific data, but each WU contains a specific amount of factoring to do.  In the next weeks or months, we will release yet another prime number application that implements the Lucas-Lehmer test for Mersenne primes.  This will show yet another data model for Plura, so stay tuned for more information.

We have other customers that are in various stages of evaluating Plura usage for a variety of applications.  We’ll release details on these if and when they agree to do so.  In the mean time, if you want to explore the possibility of running any particular algorithm on Plura, feel free to contact us.

Comparing Plura to Amazon’s EC2 for High Performance Computing

Amazon has recently come out of beta with their EC2 service, as noted by several sources:

• http://www.techcrunchit.com/2008/10/23/the-cloud-is-shaping-up-amazon-beefs-up-ec2-bechtolsheim-shifts-his-attention-to-arista-networks/
• http://gigaom.com/2008/10/23/amazon-opens-windows-for-all-and-ec2-graduates/

EC2 has been extremely successful since its launch.

I’d like to do a quick comparison between Plura and EC2 for high performance computing.  (Note: All EC2 information was taken from Amazon’s website.)

Compute Performance

EC2 High-CPU Medium Instance nodes provide 1.7 GB of memory, 350 GB of instance storage and local network accesses between EC2 nodes at no charge.  Plura nodes vary in memory and size and we never use the disk.  Plura applications can request nodes with a minimum memory size if necessary.  Also, Plura nodes are not connected.  Each node knows only about its own tasks and can’t share with others.  This lends Plura to what’s called embarrassingly parallel applications of HPC.  That said, we have techniques for making some very difficult algorithms embarrassingly parallel.

Plura has a significant advantage over EC2 in terms of amount of available computing power.  Right now, the maximum # of nodes a customer can have on EC2 is about 1,000 to 3,000 nodes.  On Plura, customers have access to the entire node pool, which is currently over 50,000 nodes.  This gives Plura users significantly more compute power.

Cost

Each EC2 High-CPU Medium instance costs $0.20/hour.  This is equivalent to paying $1752/year for a  5-6 GHz CPU.  Plura charges approximately $100/year for a 2 GHz CPU (the average speed of nodes on our network).

The “conversion factor” between Plura nodes and EC2 nodes would be 2.5 Plura nodes = 1 EC2 High-CPU Medium Instance.  So Plura charges $250/year for the equivalent of 1 EC2 node, while Amazon charges $1752/year.

Side note: If you need to use EC2 nodes with Internet bandwidth, the $/year goes by $1000s/year for each node.  For now, Plura builds this in for free.

Conclusion

For HPC applications that need a lot of inter-node communication, EC2 will probably suit your needs better even at a higher cost.  However, if your application is suitable for Plura, you can save 7X on your compute costs.  If you need the equivalent of 1000 5 GHz nodes for a year, Plura will save you over $1.5 million ($250K vs $1.75M).