451 Take on Red Hat-CentOS

There was a somewhat quiet, cost-free acquisition of sorts in the Linux world earlier this year when Red Hat announced it was joining forces with Red Hat Enterprise Linux community clone CentOS. The move, which effectively brings organization, governance, backing and technology of CentOS under Red Hat’s brim, is interesting for a few reasons.

First, it illustrates the continued presence and power of unpaid community Linux distributions like CentOS. Second, it’s part of the changing Linux market, which is being driven by cloud computing and new types of uses on the rise. Third, it also may be a sign that open source software users and customers are exerting more influence than ever before.

Read the full article at LinuxInsider.

A new SUSE Linux, separate from Novell

I’m currently researching and writing about the changing Linux and operating system landscape, which is being impacted by cloud computing, additional competitors and devops, as well as the ongoing impact of unpaid, community Linux distributions such as CentOS and Ubuntu.

Another one of the changes underway is what’s happening with SUSE Linux, largely considered the second-leading enterprise distribution behind Red Hat Enterprise Linux, which has undergone some uncertainty thanks to the Attachmate deal to purchase the distribution and separate it from Novell.

To continue the strong SLES development and address this uncertainty, it was good to see recently extended support for SLES to provide greater assurance to the community and customers. We’ve also mentioned before the credit due to Attachmate for communicating its intentions on OpenSUSE –important since both RHEL and SLES are bolstered, particularly in cloud computing in my opinion, by their unpaid, community cousins: Fedora and OpenSUSE. We would note we also continue to see strong support for Fedora community, development and openness from Red Hat.

Given we have wondered in the past about whether the SUSE Linux technology and business (including OpenSUSE) would have more opportunity separate from Novell’s other business in networking, identity management and collaboration, it will be interesting to see what happens. Will a dedicated SLES focus lead to ore market traction? Will it accelerate the promotion and use of SLES and/or OpenSUSE in cloud computing, where open source continues to be significant? Will the contribution to the Linux kernel from SUSE and OpenSUSE continue? Will there be another sale of SUSE Linux? Perhaps, but it looks more and more like we’ll see an independent SUSE Linux operating out of its home town of Nuremberg, Germany. And while I continue to see and sense some degree of uncertainty around SUSE Linux, I am also getting reports that there is still good loyalty to the SUSE Linux brand, technology and community.

Given the changes afoot in the Linux and operating system space, the change in ownership of SUSE is a relatively minor disruption. How well SUSE Linux, OpenSUSE and their supporters are able to tap into the driving trends of cloud computing and devops will be much more significant to the users and future of SUSE Linux, as well as any other version of the open source OS.

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.

451 CAOS Links 2010.12.07

The future of Java. Google releases Gingerbread. And more.

Follow 451 CAOS Links live @caostheory on Twitter and Identi.ca, and daily at Paper.li/caostheory
“Tracking the open source news wires, so you don’t have to.”

# The Register reported, and Oracle later confirmed, that Oracle’s proposed roadmap for Java 7 and 8 has been accepted.

# Stephen Colebourne provided more details on the results of the vote on Java 7 and Java 8.

# Google introduced the Nexus-S with Android Gingerbread.

# Nuxeo launched new modules for semantic linking and auto-categorization.

# Red Hat released JBoss Enterprise BRMS 5.1.

# Minkels released Varicontrol 1.0, an open source-based data centre monitoring system.

# Oracle is planning to port its Enterprise Linux distribution to its Sparc processor.

# Groklaw offered its perspective on what Novell’s OIN membership means for those 882 patents.

# Mentor Graphics acquired certain assets of CodeSourcery.

# SAP leveraged SUSE Linux Enterprise Server for SAP StreamWork, enterprise edition.

# Henrik Ingo continued the discussion on the state of MySQL forks/ecosystem.

CAOS Theory Podcast 2010.11.12

Topics for this podcast:

*Our latest CAOS Special Report – Control and Community
*Red Hat releases RHEL 6
*Symbian and Oracle highlight community challenges
*The latest on government adoption of OSS from GOSCON
*Open core issue continues, now with Linux and evil twins

iTunes or direct download (31:02, 8.5MB)

Linux thunderstorm in the clouds

There’s a thunderous battle brewing in the enterprise IT market and it’s all about Linux. The battle brings new names to the fight as well, from the largest of cloud providers Oracle, Amazon and VMWare to smaller upstarts tuning Linux specifically for cloud computing and its users, such as the hoster and service provider-oriented CloudLinux.

As for Oracle and its introduction of Oracle Unbreakable Enterprise Kernel, this has been causing a stir in enterprise Linux circles. However, I have to agree with Red Hat CEO Jim Whitehurst that this latest Oracle move is quite reminiscent of when Oracle Unbreakable Linux, later known as Oracle Enterprise Linux, hit the market in 2006-2007. It was supposed to wipe out Red Hat, but it didn’t. I expect with plenty of Oracle shops in which to grow, Oracle’s Unbreakable Enterprise Kernel will manage to move ahead without much disruption to Red Hat’s growth, which based on its latest earnings call is healthy.

Oracle’s Solaris, on the other hand, continues to face an ongoing challenge of defection to Linux, whether Red Hat, Ubuntu, CentOS, SUSE or other OS including BSD. As we’ve covered on the CAOS blog, Oracle’s decision to kill off OpenSolaris may impact customers wary of the company’s growing presence throughout the software stack and now in hardware, too, leading many to finally make a decision on moving ahead with Solaris or moving to Linux. There’s also another option on the table now that we’ve seen the launch of OpenIndiana, a fork of OpenSolaris supported by the Illumos Foundation.

There’s another computing giant making noise around Linux: Amazon, which, similar to Oracle, is seeking to supplant use of and payment for Red Hat’s or other Linux distributions by making available its own Linux Amazon Machine Image (AMI) optimized for Amazon Web Services. There’s no question the presence and continued, increasing involvement from cloud players such as Amazon will impact what other vendors are offering and what customers are consuming in the cloud. This, and the announcement of OpenStack from Rackspace, are just the kinds of thing we can expect to see more of, as covered in our latest CAOS special report ‘Seeding the Clouds.’

It’s not just the largest of vendors that are getting in on the cloud computing action and traction for Linux. Another example is CloudLinux, a smaller vendor that aims its Linux, based on CentOS with added control panel and other features, squarely at service providers interested in cloud computing. Furthering its own efforts, CloudLinux recently announced continuing growth and an interesting deal with Ksplice to provide hosting service providers CloudLinux kernel updates without rebooting servers.

CloudLinux, as well as Canonical and its Ubuntu Linux, highlight the head start Red Hat, and to some extent its enterprise Linux counterpart Novell, have given other companies ready and willing to serve up Linux in the cloud. These vendors, many of which base their own offerings on CentOS, also highlight the ongoing presence of community Linux in cloud computing, a topic we’ve covered as well.

Just when it seemed things couldn’t get any crazier or more competitive, we haven’t even touched on the rumored, potential acquisition of Novell’s SUSE Linux by virtualization giant and cloud contender VMware, which has significant implications for the market. Sure VMware has been talking for years about how its software obfuscates the need for an OS, but hey, wasn’t it only a year ago when Oracle acted as if cloud computing didn’t exist? Turns out the clouds are for real, and it turns out Linux is more relevant than it ever has been. The real interesting thing is that in many ways, the competition is just getting started.

Oracle updates Linux with Unbreakable Enterprise kernel launch

Ever since Oracle introduced its Unbreakable Linux support program in late 2006 there have been doubts about whether the company could continue to maintain compatibility with Red Hat Enterprise Linux.

Doubts have increased recently following Red Hat’s decision to cut support for Xen in favour of the KVM virtualization technologies it acquired along with Qumranet, while Oracle remains committed to Xen.

While Oracle’s CEO Larry Ellison made no mention of this issue he effectively delivered Oracle’s answer to it on Sunday night as he announced the launch of Oracle’s Unbreakable Enterprise Kernel for Oracle Linux (formerly Oracle Enterprise Linux).

The Unbreakable Enterprise Kernel has been developed by Oracle in relation to its Exadata and Exalogic integrated systems and is being positioned by Oracle as the recommended choice for all Oracle software.

Ellison maintained that Oracle will continue to support Red Hat compatibility, and Oracle Linux will continue to include Oracle’s “Red Hat Compatible Kernel” should customers require compatibility.

The Unbreakable Enterprise Kernel is optimised for flash cache reads, Infiniband support, solid state disks and OLTP transactions, according to Ellison, who added that a new kernel was required not just to add Oracle’s additional functionality, but also because, he claimed, Red Hat is “slow to pick up community enhancements.”

Ellison also noted that Oracle’s products have not been tested or certified on Red Hat Enterprise Linux for some time, but on the Red Hat-compatible kernel.

Ellison also noted that Oracle now has more than 5,000 customers for Oracle Linux. To put that in perspective, that is approximately 1.4% of Oracle’s 370,000 total customers. It is also less than half the number of new customers Red Hat adds every quarter.

CAOS theory: The launch of Unbreakable Enterprise Kernel appears to be the inevitable result of development work Oracle has been doing to optimise Linux on its Exadata and Exalogic machines, as well as Red Hat and Oracle’s divergent virtualization strategies. While Oracle maintains that it will continue to support Red Hat compatibility it is clear that Oracle Linux with the Unbreakable Enterprise Kernel is the company’s Linux variant of choice, especially for its integrated systems and that overtime the need to maintain compatibility with Red Hat Enterprise Linux will diminish. The big barrier to cutting off ties with Red Hat remains ISV certification.

Red Hat’s organic growth opportunities

We reported recently on Red Hat’s revenue growth and deferred revenue. One of the things I have been looking at recently is the slowdown in Red Hat’s growth in recent years, and the opportunities that the company has to improve that growth.

For some perspective it is worth noting that while Red Hat’s revenue has been growing steadily:

The rate of growth has been in decline for some time:

We have also noted (451 Group clients only) that the company will in all likelihood have to invest in inorganic growth if it is to meet its ambitious targets (such as 50% of server operating system market share by 2015, and growing to $1bn in revenue over three years – from February 2008).

Unfortunately for Red Hat its opportunities for inorganic growth in its core Linux market are limited since its dominance of the enterprise Linux market means that very few vendors would help it gain serious Linux market share. While there are multiple opportunities for the company to expand into new markets one problem that the company has is that would-be acquisition targets (MySQL, Hyperic) keep getting snapped up by its rivals.

(This isn’t a post about inorganic growth opportunities, but given our suggestion that open source can serve as an on-ramp to the cloud I would suggest that Red Hat could do a lot worse than look at Eucalyptus Systems as a long-term growth opportunity).

Fortunately for Red Hat the two major acquisitions that it has made in recent years (JBoss and Qumranet) both provide the company with opportunities to drive organic growth. Indeed, looking at the company’s business it is interesting to note quite how many opportunities for organic growth are at its disposal:

* JBoss – Red Hat’s middleware business continues to grow faster than the Linux business, albeit still not as a reportable segment of the company’s revenue. The company noted that 30% of its largest deals involved a middleware is fiscal 2009, there is still a lot of opportunity for greater cross-selling. The acquisition of systems integration and consulting firm Amentra was designed to help it deliver better value to JBoss customers. That and the JBoss MASS migration tools should start to deliver.

* The channel – 61% of deals came from the channel in Q1, up from 56% Q4. Red Hat more than doubled channel partners to 4,500 in 2009. Advanced Partners – VARs/SIs – grew from about 100 to about 350 in 09. Additionally the company has noted that while its renewal rates for its biggest accounts are close to 100% (“I think the only one that didn’t in the last couple of years was Oracle itself”, noted Jim Whitehurst in June) renewal rates from channel deals tend to be lower. It has put a program in place to rectify that.

* Increased penetration into existing accounts – Red Hat had 40,000 new customers in FY09. As the Eclipse Foundation’s Donald Smith noted, that means the company has a low revenue per customer. However, it also suggests huge opportunity for increased penetration into existing customers.

* Up-selling to Advanced Platform – traditionally, 70% of Red Hat’s Linux users have been on what was Red Hat Enterprise Linux ES. However, 50% of renewals in FY09 upgraded to the higher price Advanced Platform, rather than going for the standard Enterprise Linux Server.

* Virtualization – One of the drivers for AP is that it includes the advanced virtualization capabilities. Interest in virtualization is not only generating demand for the higher-priced RHEL variant but also helps Red Hat to avoid spending freezes on new hardware by de-coupling RHEL adoption from new hardware spending.

* Free to fee – Matt Asay noted recently that “nonpaid usage of Red Hat’s software that may well pose a bigger risk” to Red Hat than its chief rival Novell. That is something that the company is aware of, and it has been auditing customers to ensure that the amount of RHEL systems they have running is suitable for their subscription level. The company also sees a significant opportunity in converting users of community Linux distributions – such as CentOS or Ubuntu – to RHEL subscribers, and in 4Q09 landed a multi-year, multi-million dollar free-to-paid deal.

Will those be enough to help the company achieve its ambitious goals? Possibly not, and we do believe that Red Hat needs to expand its addressable market, but it is clear that even without acquisitions Red Hat has multiple opportunities for growing revenue in the next couple of years.

Red Hat cloud program needs partners, punctuation

Red Hat announced a new cloud certification and partner program. I’ve been expecting to see and hear more from Red Hat on cloud computing given its prominence in Linux and open source and the prominence of Linux and open source in the cloud. However, I’m left asking a few questions.

First of all, where are all of the partners in this new partner program? Other than Red Hat, the only other vendor mentioned is Amazon. Okay, Amazon is a good one to have on the cloud announcement, and is a logical starting point for what will hopefully attract more partners. However, as Red Hat describes a ‘triangulation’ of forces — customers, ISVs and cloud providers — that are driving the climb to the cloud, I wonder why we don’t have representation from any other points of the triangle or any other cloud providers or ISVs. Upon seeing the announcement, I was expecting to see a long list of vendors ready to put RHEL and JBoss to work in the cloud, similar to what we’ve seen from Red Hat in its announcement of the Red Hat Exchange or the more recent Open Source Channel Alliance. In response, Red Hat says it is too early in cloud computing and in its partner program to name additional vendors, and to be fair, we agree the industry is still early in the process. Still, given the number of open source cloud computing vendors crowding to market, we might expect more than Amazon to kick things off.

I also wonder how much of Red Hat’s cloud program may be a reaction to the use of Ubuntu Linux in the construction, deployment and management of enterprise clouds, particularly internal clouds. RightScale, for example, recently switched from using, for its cloud management software, Red Hat clone CentOS to Ubuntu based on customer demand. Ubuntu distributor Canonical also includes a preview of its Ubuntu Enterprise Cloud in its latest release and the company is focused on leveraging its significant opportunity in cloud computing. With its leadership on the desktop and, with Ubuntu Netbook Remix, in netbooks, Canonical’s Ubuntu also stands to benefit from its client capabilities, which may be key to connecting with the cloud. For its part, Red Hat says it does not see Canonical and Ubuntu as competition here, but rather as a benefit in promoting the advantages of Linux and open source for enterprise cloud computing.

Of course Red Hat is competing with other Linux distributions beyond Ubuntu in the cloud. Novell’s SUSE Linux is the most direct, enterprise Linux competition, but consistent with our report, CAOS Eight-The Rise of Community Linux, and with statements from Red Hat CEO Jim Whitehurst, there is also competitive pressure from ‘community’ or unpaid Linux. This includes community versions from the leading Linux vendors — Fedora and OpenSUSE — as well as other non-commercial Linux distributions that are popular in enterprise datacenters and, from what we’re hearing, in cloud computing as well. This would include Red Hat community clone CentOS and Debian, upon which Ubuntu is based.

One final question from Red Hat’s cloud partner program announcement: where’s the open source (or at least, where’s the noise about open source)? I’ve been hearing from vendors and customers and saying to reporters and clients that open source software is popular and critical for cloud computing mainly for its flexibility and freedom from vendor lock-in. Customers that have been burned by the end-of-life of whole enterprise operating systems are more confident moving to Linux when they know their staff skills and connections will continue to be useful beyond the life or success of any particular vendor. In looking at Red Hat’s announcement and partner program, and following the vendor closely as I do, I believe it could do more to set open source and itself apart from others when it comes to cloud computing. I don’t doubt Red Hat’s dedication to open source and true openness, but I think it is time to take advantage of that dedication and its community and leave no doubt that Red Hat and its partners and Linux and open source will play a prominent role, if not the lead, in cloud computing.

Sun full of open source and skepticism

Sun continues to take a performance pounding, and the rumors of replacements, layoffs and revamps are beyond swirling and now perpetuating skepticism of the company. It strikes me as odd that Sun, which has embraced open source and is also the defacto leading corporate open source software contributor, is continually dogged by doubts about its transitions and tenures despite well-respected technology and participation in open source. Part of this lies in the company’s continuing dichotomy in strategy — a reference to tepid support for Linux and continued preference for and focus on Solaris. This is a large part of Sun’s ‘handicap,’ IMHO when it comes to Linux and open source. Sun has its own OS, and therefore is in the same category as the dreaded Microsoft for many.

However, Sun has a longstanding, solid history with open source. OpenOffice, OpenSolaris, OpenSparc, Java, etc. While the company has generally benefited from its move to make Java open source under the GPL, its OpenSolaris and Solaris OS under the CDDL have been a somewhat different story. Nevertheless, Sun knows how to do open source right and continues to participate effectively in a variety of open source software communities, projects and enterprise products.

Let’s also not forget that it was Sun that started off this year with a billion dollar bet on open source, MySQL and its database software and business. When Jonathan Schwartz and co. were on the conference call for the acquisition in January, there were many references to Sun’s belief in the LAMP stack (along with the expected reference to the possibility of a SAMP stack that includes Solaris). And therein lies the dichotomy again.

Does Sun want to support and see success from Linux? Or does it want to see success from Solaris (and OpenSolaris)? The company may want to have it both ways and while it’s certainly possible and practical to support multiple operating systems in this day and age, Sun needs to make it clear whether it wants to fan the flames or fight the fire that is Linux. Let’s consider Novell. Is it putting much investment or roadmapping into Netware? No, the company is focused on Linux and integration of NetWare and Linux in Open Enterprise Server since it acquired SUSE in 2003. While an acquisition spurred the Linux embrace in Novell’s case, Sun does not necessarily need to buy a Linux vendor (there are fewer of those, too with Xandros’ recent purchase of Linspire).

The bottom line is that many if not most enterprise Linux wins come at the expense of Solaris and other Unix software. Sun would be wise to recognize this and it could go a long way toward clarifying its achievements and objectives with open source and getting its house in order.