Congratulations to Greenplum and Scott Yara, Bill Cook, and Luke Lonergan in particular! It has been quite a roller coaster ride over the last 10 years and there were a number of times we stared at the abyss only to come back stronger. This is a story of great people and incredible perseverance. The great news is that we leveraged two strong trends early on – the era of big data and the need for cheaper and better solutions and the fact that hardware is a commodity and the value is in the software. We also leveraged the open source database platform PostgreSQL as the initial foundation for our technology. After all these years, I am glad to see that EMC and others have caught on to both of these facts.
Amazon has taken off with its cloud compute infrastructure but there still have been some limitations from an enterprise perspective. Mainly, some enterprises are concerned about keeping their data private, about reliability, and storage costs over time. Any enterprise looking at potentially leveraging the cloud would love to have a hybrid solution which allows them to manage their own internal cloud and then burst over to a public cloud for either automated failover, extra storage, or to port an application over after using an internal platform for development. Sun seems to get it as evidenced by their announcement today to offer their own cloud computing platform. Key here is that it will be interoperable with Amazon S3 and its platform.
"Sun anticipates that the cloud scene will feature many clouds, both public and private, that are interoperable and driven by different application types. Applications eyed for deployment on Sun Cloud include Web 2.0 applications, social networking systems, gaming applications, and anything that needs the scale of the Web, said Tucker. Departmental applications are envisioned as well.
"What we're introducing in New York here is we're talking about our public cloud," for developers, Tucker said. Sun has seen a lot of interest in cloud computing from enterprises, he said. "It’s getting very rapid uptake at least in the large enterprises today," said Tucker.
What is interesting is that their is a little known startup with great open source technology called Eucalyptus which is helping drive some of this initiative. Eucalyptus will be the software that will allow the Sun cloud to interoperate with other platforms and services. With this open source platform, companies can now deploy apps on their own cloud and use Amazon or other cloud services for high availabilty or extra storage without vendor lockin. Congratulations to Rich Wolski and team as they have made tremendous strides during the last 6 months. I was just with them in New York yesterday and believe they are on to something big.
Congratulations to Bill, Scott and team on our new $27mm round of funding led by Meritech and including Sun Microsystems and SAP Ventures. You guys have been pushing the envelope since I have known you and delivering some spectacular results to boot. It is nice to see our team and product get validated with a significant round of funding so we can continue our battle to bring our customers a better, faster, and cheaper way to access and analyze massive volumes of data. When we made our first investment years ago, our fundamental bet was that a new approach was needed to deal with exponential data growth driven by network computing and internet applications. We certainly had some fits and starts tackling this data problem by utilizing a software-only approach built on top of open source software and delivered on commodity machines, but with this funding and our continued customer momentum, we are certainly on the right track. For more on this investment, read the following quotes from Jonathan Schwartz, CEO of Sun Microsystems, and Nina Markovic, head of SAP Ventures:
"Alongside Sun’s acquisition of MySQL, our investment in Greenplum is further evidence of our commitment to the open source database community and marketplace," said Jonathan Schwartz, CEO and president, Sun Microsystems. "Postgres has been a critical part of our support offering to customers, and Greenplum’s leverage of Postgres to disrupt the proprietary vendors with breakthrough business intelligence solutions creates opportunity for their investors, and more importantly, our mutual customers."
"We invested in Greenplum because we’re seeing a growing demand for scalable database technologies to support analytical and data-driven applications," said Nino Marakovic, head of SAP Ventures. "From a technology perspective, the Greenplum database is very strong and complementary to our offerings. We share the vision of enterprises harnessing ever-growing data repositories to make optimal business decisions in real time."
I always like to say that the "trend is your friend," and it is pretty clear that one of the most powerful trends in the technology industry is the commoditization of existing markets which are currently served by high-priced, proprietary vendors. In addition, it is also quite clear that companies that can leverage the web for sales, marketing, and even product delivery (downloads or SAAS) can have some significant advantages. When I look at the enterprise landscape, I am not necessarily looking for the cheap solution, but rather a disruptive one that will allow a company to offer orders of magnitude improvement in performance, price, and delivery. In addition, there are a few must-have characteristics companies should possess in order to get me interested:
1. large projected market-new emerging markets are welcome as long as we can see the opportunity ahead.
2. capital efficient business models – leverage frictionless sales and the web (try before you buy model, low barrier to usage, downloads, etc.) to create a more efficient and less costly sales and marketing machine. Also leverage the commoditization trend to deliver products faster, cheaper and better.
3. disruptive technology – orders of magnitude improvement in price, performance, and delivery
A great example of a company meeting a number of those characteristics is portfolio company Greenplum (yes, full disclosure, I am on the board and may be biased in my opinion 🙂 ). Greenplum is leveraging the power of commoditization to turn the data warehousing market, traditionally led by proprietary vendors like Teradata, upside down. Rather than rave about Greenplum, I thought I would share a recent article from Bill Inmon, a well known data warehousing analyst who some view as the father of data warehousing:
And with that explosion of data comes a corresponding increase in the costs of data warehousing. In particular, storage costs and the cost of the infrastructure required to support the storage needs are rising. The hardware vendors love to say that storage costs are going down all the time. This appeases the manager who has to pay large sums for the storage infrastructure. Storage costs may be decreasing at a factor of X, but the demand for storage is increasing at a rate of Y, and Y is a lot bigger than X.
It is reputed that one hardware vendor is selling storage for data warehouses at the rate of approximately $750,000 for a terabyte of storage.
So along comes Sun Microsystems and Greenplum with an offer you cannot ignore. How about $35,000 for a terabyte of data up to 24 terabytes?
If you are planning for a data warehouse in your future, you should take a close look at the Sun/Greenplum offering. No, let me say that a little bit more strongly – you cannot afford to not take a look at the Sun/Greenplum offer – not unless you enjoy throwing your corporate resources away.
It is about time that someone lowered the dreadful cost of data warehousing. Some of the leading vendors have been shameful in their gouging of customers. So the Sun/Greenplum offer comes as a godsend.
The offer is so good that in fact, you can afford to buy and install Sun/Greenplum, try it out, and if it doesn’t work, for whatever reason, use the gear for some other purpose. At the price ratio of $750,000 for one terabyte versus $35,000 for a terabyte of data up to 24 terabytes – you simply have to try this offer.
So you may ask yourself how we are able to offer that kind of pricing, 20x cheaper than some competitors, and still get profitable? Well first, we are leveraging a hybrid sales model where partners like Sun help drive the high end opportunities and our open source street cred and our creation of Bizgres.org helps fuel the download model. In addition, rather than build expensive proprietary hardware solutions, we are leveraging the power of commodity boxes and clusters to deliver better performance at a fraction of the cost of existing competitors. In addition, rather than start from scratch we have built some proprietary extensions on top of PostgreSQL, a leading open source database, to make it BI ready. So combine lower costs to build with a highly leveraged sales model and you can quickly see why we can offer the pricing that we do and build a great business from it. There is nothing like leveraging a powerful trend, so if you are an entrepreneur building a company with many of the characteristics outlined above, I would love to hear from you.
Congratulations to Greenplum (full disclosure: portfolio company) as it announced its first referenceable customer, Frontier Airlines, last week. To refresh your memory, Greenplum develops software that allows customers to deploy terabyte scale datawarehouses leveraging PostgreSQL at significant price/performance advantages over exsiting solutions. Building credibility is an important step for startups and getting referenceable customers and hiring industry talent are two surefire ways to do that. Here is a quote from Robert Rapp, CIO of Frontier and former CIO of Southwest Airlines, from a Charles Babcock Information Week article:
Frontier CIO Robert Rapp says the airline’s yield management process runs on Bizgres MPP. The system predicts the yield or profit that Frontier will receive on various flight combinations and ticket prices. The system helps Frontier determine where to offer seats at bargain prices and where to avoid what might turn out to be a competitive bloodletting, with no one profiting, says Rapp, the former CIO of Southwest Airlines, a pioneer of low-priced flights.
"Greenplum allowed us a very economical solution for a mid-sized airline. There are large amounts of parallelism in the system," says Rapp. A comparable but higher end commercial system used by retailers such as Wal-Mart comes from Teradata, a unit of NCR Corp. "Greenplum was available at 20-30 times less" than such a system."It was available at a very nice price point for us," adds Rapp.
Congrats to the Greenplum team on reaching this significant milestone and I am sure that this Frontier Airlines story is one that the company and I will be hearing about for a long time, in every sales presentation and pitch. As I have said before, it is important to make sure your first 5 customers are highly referenceable (extremely happy with your solution and influential in the community to get the market’s attention) so you can significantly leverage those first relationships to establish market credibility and even help close some of your sales prospects.
It seems that the open source business model has been top of mind for many in the technology industry as of late. First comes Checkpoint’s attempted purchase of Sourcefire and now comes Red Hat’s announcement that it will acquire JBoss. The acquisition price of $350mm is pretty sweet validation for the open source model considering that the multiples are about 20x trailing revenue ($20mm estimated revenue in 2005) and 6-7x forward ($50-60mm estimate for 2006). And on top of that the company only raised $10mm which means it was incredibly capital efficient. That being said, we have to remember that this is not going to change the corporate IT landscape overnight. First Red Hat may end up competing with many of its partners like IBM who have helped validate Red Hat by offering the muscle and handholding of the IBM brand and employees. Secondly, just because Red Hat’s name is on it does not mean that CIOs will immediately change their buying decisions. As I mention in an earlier post in 2004, Red Hat has needed to find more avenues for growth and what better way to do that than moving up the stack from the OS. Here is an excerpt from my post in 2004:
It seems that many of the bigger open source players are building out their own stacks ala Microsoft and others in the pursuit of growth and profits like traditional closed-sourced software companies. Isn’t this the antithesis of what open source stands for? Rick Sherlund, Goldman’s software analyst, says that it makes sense from a financial perspective since it allows vendors to cross-sell and lock-in the customer – customer retention is a good thing after all, isn’t it? While all of the open source players did their best to dodge this question and claim that they are really open, MySQl was the only company that really seemed credible here as its goal was to be part of everyone’s stack, including the Microsoft .NET one. JBoss and RHAT clearly seemed to be building their own middleware and open source stacks while at the same time claiming an open architecture.
Fast forward 18 months later and you have the first move in that model – Red Boss. Sounds like Microsoft? I thought part of the reason technologists bought open source was to not be locked in to any one vendor. This will be interesting to see as the need for revenue, growth, and profits drives some of the larger open source players and to see if they continue to remain 100% truly open. Should I tweak the JBoss app server just a tad to make it work better on Red Hate vs. Suse or .NET? Let’s watch how Red Boss balances the need to meet Wall Street expectations for quarterly numbers with the need to make its customers happy by helping them avoid proprietary vendor lock-in.
We are all enjoying the benefits that come with the commoditization of existing hardware and software infrastructure. It is true that it costs exponentially less to launch a business today versus five years ago. We are all smarter, broadband penetration is reaching critical mass, and open source and commodity hardware have become reliable alternatives to proprietary architectures and closed systems. As we all move forward with our web-based operations, it is clear that scaling the back-end infrastructure still remains a formidable challenge. There have been many an instance of popular services going down – remember Typepad, Salesforce.com, and del.icio.us as a few examples. With scaling the backend also comes a need to learn more about your users and their interactions. Data mining and analysis is becoming a big thing to not only help companies create better services but also to generate more revenue per user. In addition, for many web companies extreme data driven applications are the core of their services. Think about Zillow, Technorati, and services like Indeed which are dynamically driven services based on aggregating, crawling, and filtering millions of pieces of data. However, the fast growth of many a web-based operations combined with the need to mine the data leaves a big hole in the revolution of the cheap. Web-based operations need an open source way and cheaper option to scale their database needs, move to a data warehousing architecture without breaking the bank, and scale with user growth leveraging commodity infrastructure. Enter Greenplum (full disclosure-Greenplum is a portfolio company and I am on the board) which just released its GA product Bizgres MPP for data warehousing leveraging the best of the open source PostgreSQL database. We have been working on the code for the past 18 months, and I am quite proud of the team for having delivered the release. Greenplum is taking the best of the open source database PostgreSQL and rebuilding some of the core functions like the query optimization, execution, and interconnect. We are allowing anyone to build a shared nothing architecture ala Google to scale their backend to multiterabyte sized systems leveraging cheap hardware. It is free to run on a single machine but if you want to run a massively parallel option we charge a fee per CPU.
Dana Blankenhorn from ZDNet gets it:
This is a problem a lot of Web 2.0 start-ups like Technorati, Bloglines and Flickr are facing, and projects like Drupal will face soon. They were built with open source tools, but then find they need to "graduate" to something like a data warehouse. And there’s old Oracle, telling them there’s nothing from an open source supplier that can deliver what they need. Share with us, they say, you don’t have any choice.
Well, now there is a choice. Greenplum CTO Luke Lonergan said that O’Reilly Media, one of Greenplum’s early customers, graduated from mySQL to PostgreSQL with Greenplum and got a
100%100 times improvement in database access speed across a 500 Gigabyte database. Other Web 2.0 start-ups, and projects, can do the same thing.
"The price of conversion is where the pain is," said Yara, "but look at how fast some of these projects grow." While mySQL was smart in building on a lightweight Web base, more and more users and projects will find the need to graduate, and face proprietary FUD from major vendors saying they have to pay the "monopoly tax" in order to grow.
I truly believe the next battleground will be based on scaling the back end and more importantly mining all of that clickstream data to offer a better service to users. Those that can do it cheaply and effectively will win. The tools are getting more sophisticated, the data sizes are growing exponentially, and companies don’t want to break the bank nor wait for Godot to deliver results. Given these trends, I suggest downloading Greenplum’s Bizgres MPP and let me know what you think.
I have looked at a number of open source projects over the last year and mostly agree with Bill Burnham’s comments that many of these open source plays are "marketing gimmics for startup companies." Many of these companies are trying to start a new project from scratch, hoping to build a community brick by brick. In addition, without the ability to create a community, it is hard to build a real sustainable revenue model. Finally, open source does not matter if there is no customer need for the solution. That being said, I am quite excited about the relaunch of one of my portfolio companies, GreenPlum, which is bringing the power of open source to enterprise business intelligence. (Stop reading if you are not interested in a pitch for a portfolio company)
Quite simply, Greenplum is using an open source database optimized with supercomputing architecture to bring terabyte scale datawarehousing to enterprises. Leveraging this architecture, Greenplum will be able to offer significant price performance benefits over existing BIG IRON solutions. In addition, Greenplum is working with Josh Berkus and the PostgreSQL community to launch a new project, Bizgres, whose goal is to build a complete database system for BI exclusively from free software. From a business perspective, what I like about our strategy is that we are building off an already existing and strong community of PostgreSQL developers. Secondly, rather than pursue a broad platform play for all databases, we are focusing on a large but focused market in BI. We believe this is a great way for open source to enter the enterprise as the market is riddled with expensive solutions, BI is a top 3 initiative in most enterprises, data is growing like a weed in most places, and because we are not asking CIOs to bet their transaction systems on open source. Finally, our revenue model is not based fully on a support/service play. The open source DeepGreen product will target small-medium sized businesses or anyone with data marts and reporting apps in the 10-300 gigabyte range. GreenPlum will sell licenses for any company that wants to to deploy the DeepGreen MPP product to scale to multi-terabyte environments. While it is yet another spin on open source, I am quite excited about what GreenPlum is doing and truly hope that by leveraging the success of PostgreSQL, staying focused on a targeted market, and employing a dual license model that the company will be able to rise above the noise. As I have mentioned in a previous post, one of the clear benefits of open source, especially if you leverage an existing community, is to reduce the friction in the sales and marketing process.
Last year at this time, I was at Demo in Arizona watching a couple of my portfolio companies launch new products and networking with other VCs and entrepreneurs. Given my travel schedule of late, I decided to go to Linuxworld in Boston for a day and follow Demo from many of the bloggers like Jeff Nolan. It seems that the consensus view from Demo was that there were lots of interesting products but nothing that blew the audience away. I, too, can say the same about Linuxworld. After a few meetings in the morning, I decided to walk the expo hall to see the various offerings. I saw my fair share of companies that sold into the high performance computing (HPC) market with various clustered file servers, data replication, and workflow application software. I also saw a number of companies offering tools to better manage deployment and performance of Linux boxes. Then there were a few companies selling enterprise applications like document management platforms and antivirus and antispam software on Linux-not terribly exciting. Finally, there were various companies going after the desktop Linux market with operating systems and applications-while I found some of them intriguing, it is still quite early.
One area I did like was the market for software compliance. As we move to a componentized world where developers increasingly build in pieces of software from a variety of sources, how does a company know what they are using and from whom and more importantly what the licensing rights are for those components. 2 early stage companies going after this space are Palamida and Black Duck software. I had a chance to speak with one of the founders of Palamida, Theresa Bui Friday, and came away quite impressed. The Palamida software works like an antivirus scanner looking into code and checking against its compliance database to catalog your code base, identify whose components you are using, and then providing the user with the associated license and contact information. Increasingly IP compliance is becoming a big deal, especially when you talk to CIOs, and incorporating this type of automated scanner early in the development process can save customers a ton of headaches and potential dollars from law suits. I view this market as part and parcel with the source code scanning market. Increasingly, secure coding is being built into the QA process and companies are coming out with automated scanners to check for vulnerabilities before products go to GA. According to Reflective and NIST (full disclosure I am an advisory board member) it costs less than $0.10 to scan code early in the development process and up to $1,000 per line of code once a product is in GA.
Jonathan Schwartz from Sun has a good post about the nature of developers, and why building a relationship with them is key to creating opportunities.
One of the smartest software execs I’ve worked with had a saying, Developers don’t buy things, they join things." That’s been a pretty focusing statement for us over the years, and as we enter the new year, you should expect 2005 to be one in which we place an ever heightening focus on our dialog with the community, and the developer community in particular. And not simply maintaining the dialog we have today, but finding new constituencies, and expanding our reach. Establishing a relationship with a developer is all about starting a conversation – one that always flowers. And often into opportunity.
I totally agree with Jonathan on this. The key, however, for any small company is to do this economically and efficiently. Let me give you an example. Let’s face it-many companies selling into enterprises end up going through some "pilot" or "beta" period where a sales prospect’s developers and technologists get to use the software and deploy it on a trial basis. When I look at a sales pipeline, I always want to know who in the organization the company is selling into and why. You see, I have more often than not seen a number of early stage companies selling into enterprises but not selling high enough to the people with budget. In other words, the vendor ends up getting excited about the number of pilots in the market, many of which are with technologists who by nature like to try things and rarely end up buying. The vendor spends an inordinate amount of time reaching out to the developer or technologist to set up a pilot and then leaves with no defined criteria on when the pilot ends and how it automatically converts into a sale. The developer uses the product, sucks up lots of our resources, and moves on to the next new technology. While it is imporant to court developers and technologists in the sales process since they typically have to give the technical buy-off and can just as easily squash an opportunity, it is not a great and economical use of time to have your most expensive direct sales resources and sales engineers doing this.
Enter the web and the open source movement. Sure, "try before you buy" works if your users can download the software for free either on a trial basis, say 90 days, or if you open source a version of your product and build a real community. One of my portfolio companies is laying the foundation and groundwork to open source some of its software to help build a community and buzz around its product. We know that developers and technologists are key to the sales process. We want developers and techies to download and use the product and bang on it. However, we just want to reach them in an easier, more efficient way. Why have our most expensive sales resources do this when we can leverage the web? We want to build community around the product, gather great feedback, and land and expand our relationship with the developer. We hope this open source strategy will work as we build a relationship with the developers who ultimately will drive decision making from the bottom-up while our expensive sales reps can reach the execs with budget from the top-down. Hopefully, the two ends will meet in a selling process with less friction. We shall see. I will keep you posted as this experiment evolves.