CAOS Theory Podcast 2012.08.17

Topics for this podcast:

*Red Hat puts enterprise cred and bet on OpenStack
*LexisNexis touts open source benefits of Hadoop alternative
*Who doesn’t love Hadoop?
*Proprietary vendors siding with open source
*PostgreSQL and its cloud, commercial opportunity
*Our Hosting and Cloud Transormation Summit NA event

iTunes or direct download (32:24, 5.8MB)

Ubuntu on the move more than in decline

Ubuntu has been taking some criticism and heat for its falling Distrowatch rankings. I don’t doubt that after years of popularity, we’re finally seeing a bit of a return to the desktop Linux world of old when a new distribution shot up every week or month, then faded, then re-appeared … and so on. However, when I consider where Canonical and Ubuntu are heading, I question the significance of desktop OS standing and Distrowach rankings.

First off, I must say that Ubuntu’s slip off the ‘king of the hill’ game on Distrowatch came at the expense of Linux Mint, another polished, user-friendly Linux. It wouldn’t surprise me if some Ubuntu users may be migrating to Mint or other distributions largely out of frustration or dislike of the new Unity interface over the previous primary interface, Gnome. However, I think the move will be worth it in the long run to Ubuntu, as I’ll explain further.

If considering desktop OS, the most important aspect to me as an enterprise software analyst is enterprise desktop, and Ubuntu does well there. I’m sure there are plenty of shops running other flavors of Linux, including Mint, Gentoo, Fedora, OpenSUSE, Debian and many, many others, but for corporate desktop, the list quickly thins. Nevertheless, this is where Canonical has had some big victories, including the French police. In terms of consumer and user desktop PCs, the category itself is disappearing into converged and touch-capable devices, further distancing us from the ‘distro wars’ of the past.

Still, the server is where the real action and revenue from Linux exist. Here, Ubuntu still faces a role-reversal from most Linux distributions, using desktop and developer popularity to fuel its use as a server OS, which is also helped by free availability and cloud computing. Ubuntu continues to benefit from its early move to cloud computing and its popularity among developers, but also still faces a huge challenge in monetizing use. Significantly, the latest version, Ubuntu 11.10, incorporates support for OpenStack (or Eucalyptus) and VMware Cloud Foundry PaaS. This could be significant given what we’ve seen from this type of integration and bundling in the past. In addition, Ubuntu benefits from being among the select few Linux distribution that exist in both free, community and paid, commercial form. As reported in our special report, ‘The Changing Linux Landscape,’ the existence of an unpaid community cousin can help drive commercial growth for paid, subscription Linux, as we’ve seen happen with free Ubuntu and paid Ubuntu, as well as Fedora and RHEL and OpenSUSE and SLES.

Finally, the explosion of smartphones, tablets and converged devices — many of them running embedded Linux — makes clear there is more opportunity in these newer devices than in the desktop PCs of old. Ubuntu got a good start in netbooks and continues to be among the most advanced netbook operating systems. This should help its move to smartphones, tablets, other mobile devices, TVs and more and this is where the payoff of Unity occurs. Canonical with Ubuntu may have a real advantage as a user-friendly, mobile Linux OS that can be used by OEMs and carriers without the intellectual property stress that has marked Android, which has nonetheless laid the groundwork for mobile Linux in the industry. In the end, the pain of leaving Gnome has been significant, but the promise of where Ubuntu is headed seems worth that pain.

Our view on the Changing Linux Landscape is out

We are pleased to present our latest CAOS special report, ‘The Changing Linux Landscape.’ This latest in our series of long-format reports takes a more in depth look at the Linux server market and how cloud computing, competition and the confluence of application development and IT operations known as devops are all affecting it.

Basically, we still see commercial vendors Red Hat Enterprise Linux (RHEL) and SUSE Linux Enterprise Server (SLES) leading the market, but there are significant changes afoot, ushered in by cloud computing, wide use of other distributions such as Ubuntu, and continued use of unpaid community Linux such as CentOS and Debian. In addition, other distributions such Oracle Enterprise Linux continue to evolve and grow, as do the providers of Linux support, which now includes Microsoft. These additional competitors and choices, along with the new way of developing and deploying enterprise applications known as ‘devops,’ are all driving and disrupting the Linux server market.

This means challenges and opportunities – particularly in PaaS, which embodies devops practices – for both vendors and users. The report focuses on market dynamics with competitive analysis of leading Linux distributions, analysis of adoption drivers and hurdles, and customer case studies highlighting how Linux is put to work in today’s cloud computing environments.

Got open source cloud storage? Red Hat buys Gluster

Red Hat’s $136m acquisition of open source storage vendor Gluster marks Red Hat’s biggest buy since JBoss and starts the fourth quarter with a very intersting deal. The acquisition is definitely good for Red Hat since it bolsters its Cloud Forms IaaS and OpenShift PaaS technology and strategy with storage, which is often the starting point for enterprise and service provider cloud computing deployments. The acquisition also gives Red Hat another weapon in its fight against VMware, Microsoft and others, including OpenStack, of which Gluster is a member (more on that further down). The deal is also good for Gluster given the sizeable price Red Hat is paying for the provider of open source, software-based, scale-out storage for unstructured data and also as validation of both open source and software in today’s IT and cloud computing storage.

This is exactly the kind of disruption we’ve been seeing and expecting as Linux vendors compete with new rivals in virtualization, cloud computing and different layers of the stack, including storage (VMware, Microsoft, OpenStack, Oracle, Amazon and others), as covered in our recent special report, The Changing Linux Landscape.

While the deal makes perfect sense for both Red Hat and for Gluster, it also has implications for the white hot open source cloud computing project OpenStack. There was no mention of OpenStack in Red Hat’s FAQ on the deal, but there was a reference to ongoing support for Gluster partners, of which there are many fellow OpenStack members. OpenStack was also highlighted among Gluster’s key open standards participation along with the Linux Foundation and Red Hat-led Open Virtualization Alliance oriented around KVM. Sources at both Gluster and Red Hat, which point to OpenStack support being bundled into Red Hat’s coming Fedora 16, also reiterated to me Red Hat is indeed planning to continue involvement with OpenStack around the Gluster technologies. I suspect Red Hat is looking to leverage Gluster more for its own purposes than for OpenStack’s, but I must also acknowledge Red Hat’s understanding of the value of openness, community and compatibility. Taking that idea a step further, Gluster may represent a way that Red Hat can integrate with and tap into the OpenStack community by blending it with its own community around Fedora, RHEL, JBoss, RHEV and Cloud Forms and OpenShift.

The deal also leads many to wonder whether or what may be next for Red Hat in terms of acquisition. We’ve long thought database and data management technologies were areas where we might see Red Hat building out. This was also the subject of renewed rumors recently, and we believe it might still be an attractive piece for Red Hat given the open source opportunities and targets around NoSQL technologies such as Apache Hadoop distributed data management framework and Cassandra distributed database management software. We’ve also believed systems management to be a potential place for Red Hat to further expand. Given its need to largely stay within open source, we would expect targets in this area to include GroundWork Open Source, which joins Linux and Windows systmes in its monitorig and management, and Zenoss, which works with Cisco and Red Hat rival VMware in monitoring and managing systems with its open source software. Another potential target that would increase Red Hat’s depth in open source virtualization and cloud computing is Convirture, which might also be an avenue for Red Hat to reach out to midmarket and SMB customers and channel players. Red Hat was among the non-OpenStack members we listed as potential acquirers when considering the M&A possibilities (451 subscribers) out of OpenStack.

Given its recent quarterly earnings report and topping the $1 billion annual revenue mark, Red Hat seems again to be bucking the bad economy. We’ve written before in 2008 and more recently how bad economic conditions can be good for open source software. Red Hat is atop the list of open source vendors that suffer as traditional, enterprise IT customers such as banks freeze spending or worse, fail. However, the company’s deal for Gluster is yet another sign it is thriving and expanding despite economic difficulty and uncertainty.

You don’t have to just look at Red Hat’s earnings or take our word for it. On Jim Cramer’s ‘Mad Money’ this week, we heard Red Hat CEO Jim Whitehurst praised for Red Hat performance and traction where most companies and many economists are throwing the blame: financial services, government and Europe. Cramer credited Red Hat for a ‘spectacular quarter’ and allowed Whitehurst to tout the benefits of the Gluster technology and acquisition, particularly Gluster’s software-based storage technology that matches cloud computing. It was quite a contrast to the news out of Oracle Open World, where hardware was a focal point.

Canonical, Ubuntu broadening cloud coverage

Whether it’s been our discussion of unpaid, community Linux, the changing Linux landscape or cloud operating systems, we’ve always seen Canonical’s Ubuntu Linux as a major factor in the emerging cloud computing software market.

Canonical was the first Linux provider to so aggressively and prominently target cloud computing by its support and incorporation of the open source Eucalyptus cloud framework more than two years ago.

More recently, Canonical signaled a move with its next version of Ubuntu Server 11.10 will support a different cloud stack, the open source OpenStack software, as its default cloud platform. Eucalyptus will still be included in the Ubuntu distribution and will remain an option, which is key as we see the desire for multiple technologies and choices emerging as increasingly important to customers (the same thing seems to be happening with open source hypervisors Xen and KVM).

Given our coverage of the significance of open source in cloud computing and the importance of openness to customers moving into cloud computing, it is critical for vendors such as Canonical and technologies such as Ubuntu to be flexible in the other technologies and players with which they integrate.

That’s why it was even more impressive to see Canonical strike a deal with VMware. The two announced recently that Ubuntu 11.10 will also feature integration of and support for VMware’s Cloud Foundry platform-as-a-service (PaaS). This is yet another indicator of increased competition between VMware and Red Hat, which has its own version of PaaS in OpenShift. Regardless of the impact to its fellow Linux provider Red Hat, Canonical’s support for CloudFoundry is wise and positions Ubuntu as among the most flexible Linux distributions for cloud computing.

Canonical still faces significant challenges, primarily the monetization of developer, pilot and unpaid Ubuntu use and also its lack of pre-installation on server hardware from major OEMs. Nevertheless, the company manages to set itself apart from all other Linux providers in its continued focus on mobile and converged devices, as well. HP’s abandonment of the space and the idea of synergy between back end servers and mobile devices running the same OS is not much of a validation. However, it could also be an opportunity for Canonical, which is not burdened by the hardware business that became so painful for HP.

CAOS Theory Podcast 2011.07.08

Topics for this podcast:

*Numerous vendors pursuing PaaS
*Update on MongoDB vendor 10gen
*Structure Conference and Future of Cloud Survey
*LexisNexis open sources its platform amid Hadoop
*Decline of differentiation from open source
*Open source shows staying power in cloud, mobile

iTunes or direct download (30:01, 5.1MB)

The rise, fall and reality of commercial open source

We’ve been writing ourselves about the move toward more permissive licensing in commercial open source, as well as a lessening of the use of ‘open source’ as an identifier or differentiator. We’ve also seen others comment on a perceived loss of significance and importance of free and open source software and open standards. Combine this all with some typical observation on the lack of contribution back to open source software projects, and it might appear that open source software is a once-mighty empire in the midst of decline. However, from my perspective it seems despite all of this, open source software has never before been as pervasive, disruptive and innovative as it is right now. While we have yet to reach open nirvana, open source software is playing a pivotal role in the two most significant software markets currently: cloud computing and mobile computing.

Much of the gloom and doom in open source software the last couple of years has centered on the evil that is ‘open core,’ yet I have been among those contending that open core and the mixing of open source and proprietary models is often something that customers want. In addition, rather than just a matter of converting much or all that open source community goodness to cold hard cash, I believe all of these trends and perspectives support the idea that open source software is actually gaining in significance. Whether it is viewed as an effective marketing mechanism may be another thing, but the fact that open source is prevalent in the two hottest categories of IT today: cloud computing and mobile devices.

We’ve written extensively about open source software’s prevelance in cloud computing. We’ve also covered how the many, critical open source pieces of cloud computing stacks, whether SaaS, IaaS or PaaS, are also having an impact on openness and discussions of it, something we also see when considering recent partnerships and a changing landscape for Linux and open source software.

We’ve also covered the significance and prevalence of open source software in mobile computing. At the same time, we recognized that while open source software was a key ingredient to most if not all mobile software platforms and application ecosystems, there was a lack of open source software reaching end products and users.

In both cases, there are reasons and incentives for ‘going closed,’ so to speak, but it is the true open source efforts that elicit true community benefits: collaboration, transparency, speed, flexibility, security and more. So while open source as a term or identifier may not be what matters most to vendors or customers, there is no question open source is key to the business and future of many, if not most vendors in cloud and mobile computing. Ask Puppet Labs or Chef sponsor Opscode whether open source matters to their customers and their business. Ask Google whether openness is something they consider as they move forward on Android and Chrome. Ask Rackspace whether open source is critical in its open source cloud computing stack, OpenStack. Ask HP whether it is meaningful that WebOS is open source. I have. It is. So the next time we hear about the surrender, retreat, fade or decline of open source software or its importance in today’s computing landscape, just remember that today’s key markets tell a different story.

CAOS Theory Podcast 2011.05.27

Topics for this podcast:

*Smaller PaaS players unite with DotCloud-Duostack deal
*Typesafe taps Scala, Akka, open source and devops
*Changes continue at Continuent
*Future of Open Source Survey debrief
*Changes afoot at the OSI
*Microsoft moves to support CentOS Linux

iTunes or direct download (32:15, 5.5MB)

CAOS Theory Podcast 2011.05.13

Topics for this podcast:

*Watching for possible devops deals
*New technology, offerings highlight Hadoop
*Oracle proposes Hudson as Eclipse project
*Red Hat’s latest IaaS and PaaS
*Defining open source
*Big changes in the Linux and open source landscape
*451 Group at OSBC 2011 in San Francisco

iTunes or direct download (36:17, 6.2MB)

DevOps and PaaS, yes, but now No-Ops?

We continue to closely watch the devops trend, with some new offerings and new nomenclature, but also validation of our contentions this would begin washing over more mainstream enterprise IT.

Some of the most recent discussion of devops is coming in context of VMware’s Cloud Foundry announcement and offering, an open source PaaS that gives developers another option for building, testing and deploying cloud applications and services. While I do believe Cloud Foundry and VMware’s decision to opt for an open path in PaaS is further evidence that cloud computing may be opening up.

Based on some of the initial Twitterverse reaction to Cloud Foundry, it is also further evidence that devops is contending with another term that has emerged in the discussion of deploying applications in and among today’s cloud computing resources and environments: ‘no-ops.’ The idea is that infrastructure – servers, storage and network — as well as its configuration and maintenance are so automated, there is really no need for the ‘ops’ or system administration part of devops. However, in the larger picture and in the long run, particularly at greater scale, there is undoubtedly need for system administrators. One of the bottom line findings of my research on devops is that the trend is very much about a dramatically changed purpose and role for system administrators, who are typically freed up of mundane OS maintenance and other tasks, but who must also embrace openness and transparency in their operations and scripts, which can be very foreign. While no-ops may be one way to respond to developers cries of ‘give us root,’ I believe that devops with the ops is required for a successful approach. That ops part may indeed be handed off to someone else, and the options and ability to do so have never been greater — again thanks mostly to readily-available cloud resources and infrastructure. Another perspective on devops is that it is bringing some of the agile and automated practices and procedures of software development into the datacenter and operations team, which have previously been focused on their own scripts and stability above all else.

So when I’m asked does devops mean devs doing more ops? Is it ops doing more dev? I say this: devops is the confluence of roles and duties among both software developers and IT operations professionals — many of whom are increasingly working in both jobs at various points or together in their careers. No-ops may emerge as a preferred option as organizations use and grow confidence in various PaaS offerings, as well as more openness in the clouds in general, perhaps. Still, I think that the ops folks still have a tremendous role to play, and I wonder about the PaaS innovation that will be possible when we see the same style of collaboration and communication in operations that we have had on the development side, in large part because of open source, an example being Facebook’s recent move to open up on its datacenters.

Is cloud computing opening up?

We’ve already identified the significance of open source software to cloud computing, based on the cloud stacks from large IaaS, PaaS and other providers, on the most popular projects used for public, private and hybrid clouds and on the traction of key open source pieces such as Linux, Xen, KVM, Apache Tomcat, Hadoop, PHP, Ruby and many others. We’ve also discussed how open source is playing a role not only in the technology, but in the discussions, debates and overall evolution of cloud computing. While we believe the continued use and growth of critical open source pieces in cloud computing will contribute to a more open cloud ecosystem and market, we actually saw some evidence of this recently with word that the next Ubuntu Linux from Canonical will support not only the Eucalyptus cloud framework, but also the ever-popular OpenStack technology, project and community.

We wondered recently about the impact of a cloud partnership between Red Hat and Eucalyptus Systems, which also works closely with Canonical for its Ubuntu Enterprise Cloud. In a recent discussion, Marten Mickos told me Eucalyptus Systems fully expects and supports Canonical’s moves toward another cloud framework, OpenStack. While Canonical’s strategy probably has as much to do with customer demand, particularly for cloud flexibility, as it does with responding to rivals’ moves and deals, I believe that both the Red Hat partnership with Eucalyptus Systems and Canonical’s support for multiple, open source cloud computing frameworks signal a more open cloud computing market that is evolving. Customers are prioritizing flexibility and portability, and open source represents both perceived and real mechanisms for providing it. We’ve seen similar support from rival vendors on the operating system and hypervisor, most notably with Red Hat and Microsoft on virtualization, and I expect we’ll see this repeated with other vendors and technologies in cloud computing.

Sure there is still the question of how open is open enough, but the latest activity is truly good news for users of open source software and customers of open source vendors, who will benefit from this cross-project, cross-cloud platform support, collaboration and perhaps, community.

Red Hat reaches for clouds with interesting Eucalyptus deal

It was interesting to see a partnership announcement between Red Hat and Eucalyptus Systems, commercial supporter of the Eucalyptus cloud computing framework. Why so interesting? Mainly because there are several, significant competitive, partnership and technology implications from the deal.

First off, we’re been waiting for a more pronounced response from Eucalyptus Systems to OpenStack, the open source cloud computing stack backed initially by Rackspace and NASA and supported by a growing cast of both open source and non-open source players. We’ve also wondered about the competitive implications of OpenStack for Red Hat, which despite previously being sharply focused on the enterprise server market, is now working hard to extend into cloud computing, service providers and Paas, as evidenced also by its recent acquisition of Makara, which was a Eucalyptus Systems partner. Thus, the Red Hat-Eucalyptus partnership may be in part a response to OpenStack on the part of either or both partners.

The pairing is also interesting since Eucalyptus Systems has long worked closely with Ubuntu Linux distributor Canonical, which relies on the open source Eucalyptus software for its Ubuntu Enterprise Cloud. While Canonical has not presented much of a threat to Red Hat and its RHEL in the overall enterprise server market, the same cannot be said for cloud computing, where Canonical was first to address cloud computing users and also benefits from the popularity of Ubuntu in both public and private clouds and among developers. So it will be intersting to see whether Eucalyptus Systems’ partnership with Red Hat has any impact on Canonical’s own partnership with Eucalyptus Systems or use of Eucalyptus software.

There are also other significant Eucalyptus Systems partners that come into play, particularly the recently-launched NRE Alliance, a coalition of newScale, rPath and Euaclyptus Systems for self-service private and hybrid clouds, as covered in a 451 Group report. Additional Eucalyptus partners that make its partnership with Red Hat and its progress even more interesting: Convirture, Dell, GroundWork Open Source, HP, Membase, Novell, Puppet Labs, Rightscale, rPath, Terracotta, Vmware and Zmanda.

Finally, it’s interesting yet again to note that we may be seeing some response to the fact that much of the Linux that is used in cloud computing is unpaid, community Linux, such as CentOS, Debian, Fedora and Ubuntu, which is frequently a leader among Linux distributions, whether public or private clouds. The good news for Red Hat is that we are seeing Fedora among these community editions among Linux in the clouds. The use of unpaid, community Linux by many cloud users — particularly those testing and building privte and hybrid clouds — also highlights the monetization and commercialization challenges faced by Linux vendors here, particularly cloud leader Canonical.

Back to the Red Hat-Eucalyptus deal, we have no question we will continue to see interesting partnerships form among Linux and other open source software vendors and others as they all come together in the clouds.

CAOS Theory Podcast 2010.12.10

Topics for this podcast:

*Oracle, Java, the Apache Software Foundation and open source
*An update on some open source database and data management players
*CorraTech grows with support for open source application alternatives
*Red Hat-Makara acquisition analysis and impact
*Linux kernel report shows strong support, but what now for Novell?

iTunes or direct download (29:31, 5.1MB)

Linux kernel solid, but what will become of Novell contribution?

The Linux Foundation has just released its annual Kernel Development Report, tracking the amount of changes and contributions to the Linux OS kernel. Similar to last year, there are no real surprises on this year’s report, which provides a good check on who is developing the core of the OS and what they are doing. However, I do have one big question for the kernel and Linux contributions after reading the report: what will become of the Linux kernel work from Novell now that it has been acquired, along with the SUSE Linux technology and business?

We do see consistency again this time on the top kernel contributors, particularly Red Hat, Novell, IBM and Intel. While Red Hat continues toward its goal of a billion dollars in annual revenue (helped along by its cloud computing and PaaS play), the fate and future of Novell and SUSE Linux is less certain. While we do believe the OS itself will live on thanks largely to big enterprise users and developers that are part of the SUSE Linux and OpenSUSE communities, and Novell acquirer Attachmate has signaled continued support, the kernel contribution from Novell developers is really another matter. We will be watching closely to see whether the kernel support continues, but we may see Novell fade on the top kernel contributor list as others, such as Oracle, SGI, Fujitsu, Parallels, Nokia, Google and others may rise.

It is also interesting in the kernel report to see yet more mobile and embedded contributors — including AMD, Freescale, Cavium Networks’ MontaVista, Renesas Technology and others — something we started seeing with last year’s kernel report and expect will continue to grow.

In conclusion, Linux fans should be encouraged by the quality, diversity and new directions of the Linux kernel development community. While there is cause for some concern regarding Novell’s contribution, overall, Linux development seems to be charging ahead.

Is it asking too much to expect corporate contributions from paying customers?

Are vendors that ask their paying customers to also contribute code changes trying to have their cake and eat it too?

The issue of increasing corporate contributions to open source has been bubbling under ever since Jim Whitehurst highlighted the lack of enterprise participation during a speech at the 2008 OSBC conference.

Whitehurst also mentioned the topic during 2009’s OSBC but in the intervening year there has been little in the way of visible progress. Perhaps this is because it is asking too much.

If it is true that there are two types of people, those who will spend money to save time, and those who will spend time to save money then is it not somewhat unfair to expect those that have already spent money to save time to also spend time on open source contributions?

Most open source vendors expect enterprises to be the paying customers rather than the community users but if the software in question is only really of interest to enterprises then it stands to reason that the vendor will struggle to encourage community contributions.

A radical solution to this problem is to not ask enterprises to pay for your software.

This is the approach taken by Funambol, for which enterprises are the community users of its push email and mobile sync software and service providers, carriers and device markers are the paying customers.

The company separated its user base in this way because, as CEO Fabrizio Capobianco explained at OSBC, “in my opinion the community edition should be targeted at the enterprise, as they are more likely to contribute code back”.

Clearly this approach is not going to work for all software segments but the rise of cloud computing, SaaS/PaaS and hosting providers does point to a new target market for software vendors: ask the service providers to pay for the right to build or offer a service based on your software and sell that to enterprises, but allow enterprises to use it for free in-house.

This approach relies on the use of the AGPL license, which ensures that third parties are not able to offer the software as a service without being compelled to contribute back.

Even then there are no guarantees that enterprises would be more inclined to contribute back to the community. There is no way of compelling corporate contributions unless enterprise distributes the modified software beyond the firewall.

A lot of vendors would find it hard to change the habit of a lifetime and not expect enterprises to be their direct paying customers. Fabrizio’s other piece of advice, which is “do not sell anything to your community, do not even sell them support” would be particularly hard to stick to.

But if a vendor is asking enterprises to spend time and contribute code, then a start would be not asking them to also spend money.