Cloudera Whatsapp
Total Page:16
File Type:pdf, Size:1020Kb
LOS ANGELES | SAN FRANCISCO | NEW YORK | BOSTON | SEATTLE | MINNEAPOLIS | MILWAUKEE August 13, 2013 Michael Pachter Steve Koenig (213) 688-4474 (415) 274-6801 [email protected] [email protected] PRISM … Progress Report for Internet and Social Media In This Issue: Cloudera, WhatsApp Cloudera Big Data analysis firm founded by four tech veterans from Facebook, Yahoo, Google and Oracle. Developed applications based on open source Apache Hadoop to process large data sets. Has raised $141 million over five rounds of funding. Annual revenue estimated at $100 million. Market expected to grow at 54% CAGR over next five years. WhatsApp Mobile instant messaging service with 250 million global monthly active users. Subscription-based revenue model; no advertisements or virtual goods. Mostly self-funded, with only $8 million raised in four years. STRATEGIES GROUP THE INFORMATION HEREIN IS ONLY FOR ACCREDITED INVESTORS AS DEFINED IN RULE 501 OF REGULATION D UNDER THE SECURITIES ACT OF 1933 OR INSTITUTIONAL INVESTORS. WEDBUSH PRIVATE COMPANY Wedbush Securities does and seeks to do business with companies covered in its research reports. Thus, investors should be aware that the firm may have a conflict of interest that could affect the objectivity of this report. Investors should consider this report as only a single factor in making their investment decision. Please see page 10 of this report for analyst certification and important disclosure information. Cloudera Cloudera offers scalable and efficient data storage and analytics solutions to companies in the enterprise, Internet and government sectors. The company, based in Palo Alto, was launched in 2008 by three top engineers from Google, Yahoo, and Facebook (Christophe Bisciglia, Amr Awadallah, and Jeff Hammerbacher, respectively) who joined former Oracle executive Michael Olson to address problems inherent in analyzing large volumes of data quickly. Since 2008, Cloudera has gained widespread adoption, bringing open-source Apache Hadoop (highly available distributed object-oriented platform) and distributed storage to companies of all sizes. Its revenue is estimated to be in the range of $100 million annually, and employee headcount was 320 in February of this year (according to The Register). Product Hadoop, created by Doug Cutting, a Cloudera employee who named the technology after his child's stuffed elephant, is an open source Java framework that was created from Google’s MapReduce and Google File System papers; it enables distributed parallel processing of huge amounts of data across inexpensive commodity servers. According to Cloudera, much of the data produced globally is unstructured and was virtually unusable in any meaningful or systematic way until the development of Hadoop. With Hadoop, users can consolidate data – structured, unstructured, log, picture, audio, email and other communication – from disparate systems and analyze information as though all of the data were on the same platform in the same format. Hadoop supports distributed applications running on large clusters of computers processing enormous amounts of data. This differs from prevailing centralized computing systems, because the bottleneck that previously existed when transferring massive amounts of data from storage to processing is removed. As a result, Hadoop enables companies to leverage large volumes of preexisting data to better understand customers, markets, operations, products and relationships through distributed parallel processing. The platform has been especially popular with social networks and other Internet companies because of the vast amounts of data these firms generate and the low cost of Hadoop processing systems. According to a study by eBay, the global volume of corporate data doubles every 1.2 years, and prior storage platforms could not scale to meet the rising demand for more efficient and cost-effective data management needs. Subsequently, distributed processing has grown increasingly important, placing increased emphasis on the monitoring of IT system performance. According to CEO Olson, current relational enterprise data warehouse (EDW) costs are close to $20,000/TB, compared to $1,000/TB to $2,000/TB for Hadoop clusters, including hardware (via InformationWeek). We believe that Hadoop’s downward pressure on pricing provides a significant catalyst as companies shift their reliance away from EDW to Hadoop clustering for extract, transform and load (ETL) needs, business intelligence and analytics, and reporting. Cloudera’s ability to easily scale and integrate with virtually every data storage platform on almost any hardware is a central reason for its adoption by large enterprises. Business Model Cloudera has monetized its software by developing Cloudera Enterprise, a commercial software and support package offering that caters to enterprise customers with large amounts of data. Cloudera Enterprise includes CDH (Cloudera’s Distribution for Apache Hadoop), the industry’s most popular means of deploying Hadoop. The company offers customers an annual per-node subscription license to Cloudera Manager and technical support that allows a host of services to manage and analyze its data. Although the company does not provide official pricing, it told The Register packages range between $2,000 and $4,000 per node depending on features. A free version of Cloudera Enterprise offers limited functionality and caps the number of managed nodes at 50. Cloudera’s Desktop Management platform allows copying and browsing of data files stored on a cluster as well as creating, running and saving jobs for reuse and/or customization. The Desktop Management platform also monitors Hadoop cluster performance. Financials Cloudera does not release financial figures, but an article by Date Round Amount Investors GigaOm suggested that the Mar-09 Series A $5 million Accel, SV Angel, Diane Greene, Qi Lu, etc. company generated approximately Jun-09 Series B $6 million Accel, Greylock $100 million in revenue in 2012 and has grown approximately Oct-10 Series C $25 million Meritech, Accel, Greylock tenfold since 2009. The company Nov-11 Series D $40 million In-Q-Tel, Ignition, Accel, Greylock, Meritech was valued at $700 million in its latest round of financing in Dec-12 Series E $65 million Accel, Greylock, Ingnition, In-Q-Tel, Meritech December 2012 (according to an article by AllThingsD). The investment was led by Accel Partners and included existing shareholders Greylock Partners, Ignition Partners Meritech and In-Q-Tel. Previous rounds are listed above. 2 In Chairman Olson’s company blog post announcing Reilly as the new CEO, Olson said he expected the company to stay independent. Business Insider has suggested that the hiring of Reilly was a sign that the firm was preparing to go public. Other news outlets have expressed similar opinions, including Forbes and VentureBeat. In an interview with The Register, Vice President of Products Charles Zedlewski said that the company was “steadily moving out of the startup phase.” Market The market for Big Data solutions is expected to grow rapidly over the next five years. Transparency Market Research estimates that the global Big Data analysis market was worth $6.3 billion in 2012 and will grow to $48.3 billion by 2018 at a CAGR of 40.5%. Over 50% of this market came from software and services. The largest contributing sectors were healthcare, financial services and government; healthcare is expected to be the fastest growing. The market for Hadoop is also growing quickly with a 54.7% CAGR. In 2012, Hadoop’s addressable market was around $1.5 billion, and is expected to grow to $20.9 billion in 2018 (according to Transparency). This should allow Cloudera to grow for the next several years. Although adoption of Hadoop and other open-source software products has gained momentum among large enterprises, many companies are still wary of implementing them on a large scale, preferring proprietary commercial software instead. Many industries are still evaluating the use cases around Hadoop. To increase market penetration Cloudera must work closely with current and prospective clients to determine Hadoop’s optimal fit within their respective data ecosystems. Competition While Cloudera competes with more mainstream data storage and analytics providers, its main competitors are Hortonworks, a company formed by Yahoo in 2011, and MapR, a San Jose, California-based startup. Hortonworks received $50 million in new funding in June 2013 to accelerate application development. MapR received $32 million in March 2013, bringing its total funding to $61 million. Since Cloudera was an innovator in the Hadoop space and maintains several of the platform’s early designers and programmers, it should prove difficult to for new entrants to upstage the incumbent, which has accumulated over 150 customers (according to The Register) including eBay, Groupon, Rackspace, ComScore, Trulia, Samsung, LinkedIn, Facebook, and Twitter. The interest in Hadoop and other big data platforms has spurred bigger tech companies such as Oracle, IBM, Hewlett Packard, EMC, and Dell to form partnerships with firms like Cloudera. In January 2012, Oracle and Cloudera announced a partnership to provide an Apache Hadoop distribution and tools for Oracle’s newly announced Big Data Appliance. Other competitors include Google’s BigQuery, KarmaSphere and EMC. Technology The Hadoop Distributed File System (HDFS) breaks incoming files into blocks and stores each block in distinct locations. The replication