Open-source software has been a disruptive force in the IT industry for well over a decade. The disruptive potential of open-source software is far from exhausted, as new providers leverage it to expand the availability of solutions and delivery models.

• New technologies and product design concepts are beginning to enter the market through open-source software, in marked contrast to open-source software that was largely an attempt to duplicate commercial solutions.

• Open-source software will play an increasingly central role in the growth of cloud computing.

• The market will develop a more-mature view of open-source software as interest grows in understanding how collaborative development methodologies can be used in-house.

• IT organizations should explore next-generation technologies through open-source software. While such open-source solutions need to be carefully applied to appropriate low-impact projects, IT organizations can identify potential benefits without making significant vendor commitments in the process.

• Open-source software must move beyond a tactical sourcing option and become a strategically applied, in-house development methodology supported by software configuration and application life cycle management toolsets.

This document was revised on 19 April 2011. For more information, see the Corrections page on gartner.com.

This research does not constitute an exhaustive list of vendors in any given technology area, but rather is designed to highlight interesting, new and innovative vendors, products and services. Gartner disclaims all warranties, express or implied, with respect to this research, including any warranties of merchantability or fitness for a particular purpose.

For more than a decade, open-source software has been a disruptive force in the IT industry. Rather than being a single, well-defined trend, it has reached across the IT landscape in a series of overlapping waves.

Initially, open-source software was largely defined by the potential of collaborative development. The unique nature of the open-source license facilitated diverse, distributed development teams with no formal connections other than a commitment to produce software. Until now, these groups have been able to craft functional, robust solutions that have given established, commercial products strong competition. As one of our Cool Vendors, VoltDB shows open source is no longer being used largely for solutions that duplicate the capabilities (in part or in whole) of commercial solutions. Increasingly, cutting-edge technologies are appearing in the market through open-source offerings.

Open-source software has also played an instrumental role in the advent of the cloud computing phenomena. That role is set to increase as providers like OpenStack bring together a collection of open-source technologies into a massively scalable, cloud operating system. The role that open source is playing in cloud computing is also moving from behind-the-scenes supporting infrastructure to customer-facing solutions, as is evident by Cloud9 IDE, a provider offering next-generation development tools.

Open-source software has been the catalyst for some of the most significant business model innovations in the history of the IT industry; and it has yielded an array of alternative software solutions for enterprise IT organizations. Slowly, enterprise IT organizations are recognizing that the true disruptive potential of open source comes from interorganization collaborative development. If it's good enough for vendors, it's good enough for the enterprise. This will lead to a growth in products that support software configuration and application life cycle management — an area one of our Cool Vendors, CollabNet, has been focused on for more than a decade.

Amsterdam, Netherlands (www.cloud9ide.com)

Analysis by Eric Knipp

Why Cool: Cloud9 IDE is a Web browser-based integrated development environment (IDE) built on top of the popular Node.js event-driven JavaScript project. The project is open source and maintained primarily by Ajax.org, a Dutch company specializing in rich Internet application (RIA) technology. In late 2010, the cloud IDE project Mozilla Bespin was replaced with Ajax.org Cloud9 Editor (ACE), the foundation of Cloud9 IDE. Ajax.org offers a pay-per-use application development as a service (ADaaS) for programmers interested in using Cloud9 IDE as a cloud service, but there is no charge to use it for open-source projects (and it integrates with GitHub for that purpose). Anyone can download the ACE project and install it in a private Web infrastructure, or use it to create additional public cloud services. JavaScript developers can extend the project with new functionality, similar to Eclipse (for Java developers), because it is written entirely in JavaScript and has an open plug-in architecture.

Cloud9 IDE is part of the next wave of tools optimized for Web and cloud application development. IT leaders have expressed interest in hosted desktop environments for application developers; Web-based IDE delivery is the logical progression, as HTML's evolution increases the power of the browser. Today, Cloud9 IDE supports only JavaScript, HTML, Cascading Style Sheets (CSS) and Node.js running and debugging, making it useful to Web developers working in the presentation tier, building rich clients or working on server-side JavaScript components. There is support for other languages through syntax-highlighting. Run and debug support (including Ruby) is on the Cloud9 IDE road map. Whether or not Cloud9 IDE becomes the dominant Web-based IDE, it showcases the relatively newfound power of the browser and the continually increasing importance of the Web.

Challenges: Developers are already comfortable with tools such as Eclipse or Visual Studio, and may not want to migrate to a new application, particularly one that is relatively immature in terms of third-party plug-ins and extensions. Traditional development tools can be offered as hosted services via desktop virtualization. For the time being, Cloud9 IDE is useful only to JavaScript developers. Other competitive browser-based IDEs, such as Kodingen, PHPanywhere and CodeRun Studio, are ahead in some functional areas, although none is open source. Only CodeRun supports debugging. Eclipse Orion, a proposed browser-based IDE from the Eclipse Foundation, is a potentially serious competitor to any open-source software offering in this category.

Who Should Care: Developers interested in building client or server JavaScript applications entirely in the browser, particularly for projects using Node.js, should care. If you are a Ruby, Python or PHP developer, keep an eye on Cloud9 IDE later this year.

Silver Spring, Maryland (www.sonatype.com)

Analysis by Brian Prentice and Thomas Murphy

Why Cool: Almost everyone knows that collaborative development is one of the most important dynamics of the open-source phenomena, but few enterprises outside the veteran open-source community and vendor open-source development teams have paid much attention to how one actually manages this process. Specific governance structures, as well as tools, need to be in place to coordinate efforts among collaborators, including version control, build management, defects management and discussion. Builds are becoming increasingly complex, and the drive toward continuous integration and delivery will only emphasize the need for coordinated technologies.

Sonatype was founded by the creator of Apache Maven (a project-management-oriented build, reporting and documentation tool). Sonatype owns and operates Maven Central, an open repository containing over 200,000 Java components. More than 40,000 organizations call the repository nearly 300 million times each month. Maven Central includes virtually all major open-source software environments, including Eclipse, Apache, Codehaus, FuseSource, Alfresco, Spring, eXo and XWiki, along with over 1,300 independent open-source software projects. Building on this cool foundation, Sonatype combines and contributes to other leading open-source projects, including Hudson and m2eclipse. The breadth of components included in Maven Central and its wide enterprise usage help organizations implement and enforce open-source governance policies, enabling them to realize economic and efficiency benefits of open source without taking unnecessary security, quality or intellectual property risks.

By pulling together and contributing to these disparate projects and using them as a core foundation, Sonatype exemplifies the emerging trend of collaborative solutions built on top of an open-source core. This trend stems from the open-source community in conjunction with the uptake of agile practices that drive a greater need for solutions that fit together. While piecing together independent projects may work, it is not an effective strategy for enterprise IT organizations. Consequently, there is a trend for open-source solutions that combine point solutions to create collaborative platforms.

Challenges: If combining a few popular open-source projects was the most important factor in building market share, then Sonatype would be in a good position. Unfortunately, there's no guarantee that the combining of a few popular open-source projects will happen. The market is rapidly growing with application life cycle management and DevOps solutions, although most of these are built on proprietary foundations. In addition, one of the key elements of Sonatype is the Hudson Project, which has seen instability (including a project fork) because of the acquisition of Sun Microsystems (the original home of Hudson) by Oracle. Open-source projects fluctuate in popularity, often at a faster pace than commercial tools. Sonatype has committed to ongoing support and enhancement of the Hudson Project, but the community could rapidly evaporate and shift to Jenkins.

Who Should Care: Every IT organization should care. That's an assertion we make with no sense of overstatement. To date, IT organizations have sat back and waited for the vendor community to throw them open-source alternatives that they could consider on their RFP vendor shortlists. Without taking anything away from the open-source software that's on the market, the broader point behind the creation of these solutions has been largely missed by enterprise IT organizations. If collaborative, interorganization development teams can yield highly functional and robust software, then why isn't this development strategy being applied to in-house development work? That notion is beginning to resonate with many enterprise IT organizations; and, as it does, the demand for solutions that Sonatype and others bring to market will increase.

San Antonio, Texas (www.openstack.org)

Analysis by Mark Driver and Lydia Leong

Why Cool: OpenStack is a collaborative community-driven project to develop and promote an open-source "cloud stack" — software for cloud enablement and cloud management. OpenStack can be used to build the core of a public or private cloud integration-as-a-service (IaaS) offering. The project focuses on OpenStack Compute as its loud computing fabric controller, and OpenStack ObjectStore for scalable object storage. In an industry segment dominated by a growing list of proprietary vendor efforts, OpenStack's objective is to provide an "open," hypervisor-neutral, cloud stack to drive consistency and economies of scale for cloud technology investments across the broader IT community.

OpenStack was founded by Rackspace, a Web hoster and cloud IaaS provider, and NASA. More than 50 companies and organizations support OpenStack. However, Rackspace still plays a particularly significant part in the community:

• A Rackspace employee leads the OpenStack project.

• Rackspace appoints one-third of the seats on the OpenStack governance board.

• Rackspace employees are the primary committers.

• Rackspace recently purchased Anso Labs, which offers professional services for OpenStack.

• Rackspace has committed to using OpenStack as the underlying infrastructure platform for its cloud IaaS business, and thus represents the strongest direct commercial interest in its success.

Challenges: Cloud-enablement software is still in its infancy, and OpenStack is no exception. VMware is the most significant competitor; it has an increasingly mature management stack on top of its hypervisor, including vCloud Director, which enables hybrid public-private clouds, as long as they are all VMware-based. There are many other commercial products in this space, including Abiquo, Cloud.com, Eucalyptus and Nimbula. Although some of these vendors may embrace open source, they are also highly motivated to focus on their own proprietary investments to gain first-mover advantage over the competition.

To succeed, OpenStack needs to build a large enough community of committers to match the speed of innovation and code quality of the commercial competition. Moreover, OpenStack is unusual for a large-scale, open-source project, because its primary commercial sponsorship comes from a service provider, not from a software company following a commercial open-source model.

Who Should Care: Those who should care are potential public cloud infrastructure and software-as-a-service (SaaS) providers (even private cloud) that seek an open and shared investment model to reduce R&D costs related to core cloud IaaS offerings, and that do not want to use commercial solutions; and enterprises that are looking for an open-source solution for cloud storage. At the current state of OpenStack's maturity, these organizations will be early adopters, and should be comfortable with a community, project and documentation that are all in their infancy.

Billerica, Massachusetts (www.voltdb.com)

Analysis by Donald Feinberg

Why Cool: VoltDB is a spin-off of Vertica Systems, which was recently acquired by HP and the latest startup by Michael Stonebreaker, noted Massachusetts Institute of Technology (MIT) professor and a specialist in database research. During the past five years, there has been a tenfold increase in in-memory database management system (IMDBMS) engines in response to the demand for the high-speed processing that comes with in-memory models (although many are open-source projects with no corporate backing or formal support beyond community support). Although there are many different DBMS models, VoltDB follows a more-traditional model of the transactional relational DBMS row store, with full atomicity, consistency, isolation, durability (ACID) compliance, while achieving more than 20 times the performance of the traditional DBMSs.

VoltDB has a series of different product offerings. There is VoltDB Community Edition, available since May 2010 and covered under a General Public License (GPL). There are also three enterprise variants that have been available since September 2010: Enterprise Developer, Enterprise Standard and Enterprise Premium. Each of the Enterprise products adds functionality atop VoltDB Community Edition. Support service levels, warranties and indemnification differentiate the Enterprise products. This enables users to freely download the community edition for use in test and development, as well as production, if the additional functionality is not desired.

Three primary distinguishing characteristics make VoltDB cool:

• K-safety — allowing transactions to simultaneously use multiple servers with synchronous commit across the servers for high-availability (HA) and scale-out for reads with increased performance, because they can be done in parallel on the HA nodes.

• Memory snapshot to disk for persistence at user-specified intervals — currently to the Hadoop Distributed File System (HDFS), Vertica and to a normal file (for loading into other DBMS engines).

• Partitioning of data across servers with automatic transaction routing for scale-out, not only allowing large database sizes, but also spreading transactions across multiple servers for writes.

For even greater scalability and performance, VoltDB allows a partition per core, as opposed to one partition per chip or per server. K-safety and partitioning scaling of reads add the potential of online analytical processing (OLAP) and analytics to the IMDBMS, a primary motivator for some users who are performing near-real-time analytics along with online transaction processing (OLTP) in the same database. The primary difference from a streaming DBMS (such as IBM solidDB, Oracle TimesTen and StreamBase) is that the data can remain in the DBMS, giving "state" to the data, and allowing for analytics and reporting on the data.

Challenges: VoltDB is in the early stages of startup, with only a few customers, although it is a full company (no longer an incubated project of Vertica) and has acquired formal financing. It needs to begin marketing more widely, gaining customers for references in different business sectors, proving multiple use cases for a wider variety of applications. This leads to the biggest challenge for VoltDB: to gain recognition in the market among the traditional market leaders, the open-source DBMS products and the other in-memory DBMS products.

As with all young innovative vendors, VoltDB is an acquisition target, which could cause customer disruption. However, we believe that VoltDB would be acquired for its technology, and its customers would, therefore, ultimately benefit as a result.

The final challenge for VoltDB is to increase revenue, which is a challenge for all open-source vendors. Most VoltDB revenue is from support for the Community version; however, as the functional differentiation increases in the VoltDB Enterprise Edition, revenue will begin to increase, close to the levels of non-open-source vendors.

Who Should Care: Data architects, application architects and organizations looking for the high performance of an IMDBMS for OLTP that is scalable with HA and persistence (both with minimal overhead) should look at VoltDB. VoltDB can handle data entering the system with high velocity and perform transactions with this data, followed by real-time analytics. Examples of applications possible with VoltDB are fraud detection over longer time periods, real-time repricing in retail, telco call data records management, traditional capital market data feeds and trading systems, and digital advertising applications requiring low latency with high-transaction throughput.