Jeff Jaffe’s Blog

Two months ago, Novell quietly made a preview announcement of Open Enterprise Server 2. Actually, it wasn’t that quiet. It was picked up reasonably well by the technical press. It included details of the rich ecosystem of backup and antivirus support committed for OES 2 and specifics around OES training and support promotion. I call the announcement “quiet” because the press coverage that we received was dwarfed by the coverage that our Microsoft agreement had received earlier that month.

Over the next few months Novell is going to get very noisy about this product. Open Enterprise Server 2 brings together many elements of Novell’s mixed-source interoperability strategy [see my January 2, 2007 blog on Novell strategy]. These key elements include:

• Linux as the universal host – the host for numerous operating environments
• Xen as a key virtualization technology
• Interoperability
• A migration path for NetWare customers
• Linux/NetWare services equivalence
• A full NetWare ecosystem for OES
• Manageability
• Training

Thus it continues our drumbeat of a continuous set of offerings that realize the mixed source vision requested by customers.

Linux as the universal host

A key design point of OES 2 is that NetWare is fully virtualized on SUSE Linux Enterprise Server (SLES). This provides substantial customer benefit. All core operating system services are provided through SLES, so advances in hardware come to NetWare users on the accelerated time scale of Linux. This includes device drivers for new disks and printers, processor enhancements, multiprocessor capabilities, etc.

We see great synergy between fulfilling the needs of NetWare customers and Novell’s broader strategy of being the leader in interoperability between Linux and other platforms. Here we use core Linux technologies to advantage NetWare.

Xen as a key virtualization technology

This virtualization leverages Xen technology. In that sense, Novell’s Microsoft partnership is also synergistic with the OES design. Virtualizing both Windows and NetWare on SLES further hardens both SLES and the Xen virtualization support.

The OES 2 release demonstrates the maturity of Xen virtualization. We are substantially improving the support for Xen in SLES 10 Service Pack 1 (the basis for OES 2) by relying on Xen 3.0.5. This has the right capabilities to support the NetWare functionality.

Interoperability

OES 2 realizes greater interoperability between Microsoft environments and eDirectory – a key requirement especially for large customers. This is true whether OES 2 users deploy on Linux natively or on virtualized NetWare. Much of this greater interoperability will ship with OES 2; other pieces will come with future Novell investments in OES and the Novell Open Workgroup Suite.

Why is there this staging of capability? There are a wide variety of scenarios where Novell and Microsoft need to interoperate. There are workstations that need to get access to both Windows services through Active Directory (AD) as well as NetWare or Linux services through eDirectory. There are scenarios where AD and eDirectory need to interoperate directly – to share information. More demanding scenarios have AD and eDirectory behaving like a single virtual directory.

In early November, Microsoft and Novell made a broad announcement related to Linux and virtualization. Part of this announcement was a commitment to collaborate on directory technology. Since our OES 2 announcement came less than a month later, we did not have time to factor in how much directory interoperability could we get into OES 2. Instead, we will fill in this detail at Brainshare in March, with future enhancements coming later.

Migration path for NetWare customers

Several years ago, Novell announced its strategy to move NetWare customers over to Linux. This is not a journey that is completed in one step. Most NetWare customers today are purchasing OES and are developing the Linux skills to begin to leverage the Linux side. As soon as we move to OES 2 with the full virtualization support, there is a step function advance. Now, the value proposition to rely on Linux for operating systems services is completely realized.

Linux / NetWare services equivalence

NetWare has always been noted for its superior file, storage, and print server capability, with very tight administrative control. For customers that have built additional applications on top of NetWare, OES 2 provides a path for these customers to stay with NetWare. Our approach is to virtualize NetWare so that Linux is providing the operating systems services and the existing NetWare code base is providing the higher level services.

New customers are better off getting these same services native to Linux. Over time, many NetWare customers will also migrate to get these services native to Linux. OES 2 provides Linux / NetWare equivalence in that most NetWare services now have equivalent or better performance and capability on Linux.

The availability of both approaches within a single product provides customers with the choice to migrate at the pace they are comfortable with. If they prefer the NetWare bits – these bits exist on Linux and will do so for the foreseeable future. Over time, if customers want to migrate to a pure Linux approach – that is available too.

A full NetWare ecosystem for OES 2

We have heard from customers that they would like to see more NetWare partners join us in our announcements of new releases. Of particular interest are the vendors that provide backup and antivirus services. That is why in our November press release we were joined by Computer Associates, Commvault, McAfee, Symantec, Syncsort, and Trend Micro. With OES 2 we provide a breadth of support not previously available.

Manageability

Aside from the excellent manageability that has always been the hallmark of NetWare, additional tools are needed to help manage the virtualization capabilities for OES 2. We have also recently announced new ZENworks tools to manage the virtualized environment which significantly will support OES 2 with the other platforms. More about that in an upcoming posting on ZENworks.

Training

After the release of OES, we received significant input from customers that more training was needed. We have begun an extensive training program – in fact that was the highlight of our recent announcement. Within one month of the announcement, we already had 150% of the downloads of the training materials than we had expected for the entire length of time until “announce”. Clearly a great deal of interest. This training helps customers and partners develop skills in Linux, virtualization, and all the aspects of Open Enterprise Server. If you’re not familiar with the OES training and support promotion, please visit www.novell.com/oespromo. Partners can go directly to the PartnerNet portal to learn about the offer.

In my last posting, I defined Novell’s technical strategy as:

The world’s best and most interoperable Linux surrounded with management services that leverage our footprint to build a business solving customer problems in heterogeneous systems management.

Here I will describe steps that we took in 2006 to achieve this. There were four ingredients:

1. Create the most advanced Linux.
2. Recognize that virtualization was a key component in this most advanced Linux, and a key leverage point for interoperability.
3. Recognize that interoperability with Microsoft Windows was a key customer driver in the heterogenous world, and leverage our Linux and virtualization technologies to create a Linux strategy that would interoperate with Windows.
4. Build out our management services and workgroup products, with a focus on Linux, heterogeneity, and virtualization.

1. Creating the most advanced Linux

For Novell’s strategy to succeed we need – first and foremost – the best Linux in the industry. With SUSE Linux Enterprise (SLE) 10 – shipped in July – we became the unquestioned Linux leader.

Our campaign – Your Linux is Ready – emphasized that as an Enterprise Linux; as a good fit from Desktop to Data Center – we were the technology leader.

In terms of the product itself, we created numerous differentiations. Examples start with process innovation. A single code base to address both desktop and server. An Autobuild process to blend hundreds of open source projects into a single well tested distribution. 500 million lines of code shipped in SUSE Linux Enterprise.

Time to market leadership. We were the first Linux to market with XEN virtualization. Several months later we are still waiting for others to catch up. We were first to market with XGL graphics.

Server side Feature/function leadership. Xen virtualization. CIM (or common information model) monitoring. AppArmour security.

And on the desktop side. XGL graphics. Wobbly windows. 3 dimensional windows. Search. Media players. Plug and play. Interoperability. Almost a never ending list.

2. Virtualization

Recall the capability provided by virtualization. On a single computer, you can run multiple guest operating environments on a single host operating system. This allows many applications to share compute resources which both optimizes resource usage and also reduces management complexity.

We have been leading the XEN virtualization solution in the marketplace since July. As mentioned in an earlier blog, a key benefit of the XEN approach is that by using paravirtualization one gets much better performance than with full virtualization.

Here, however, I would like to emphasize a different aspect of XEN virtualization. That is the role it plays in interoperability.

There is a growing industry consensus that the XEN technology embedded in Linux is the universal hypervisor for server side computing. This means that XEN suddenly has a huge role in interoperability.

Today’s data centers are littered with different server platforms. Bringing them together with a single hypervisor technology is too important to be in the hands of one company. So the entire industry (including chip manufacturers, server vendors, and operating systems companies) united around an open source project – XEN. Once the XEN technology is embedded within a Linux distribution, such as SUSE Linux Enterprise Server, then Linux becomes the universal host. The ability to run disparate operating environments – virtualized – on this universal host gives an unprecedented level of interoperability to the end customer. Systems management simplified.

We have had a flurry of announcements to strengthen XEN and SUSE Linux Enterprise Server. In July, we announced SUSE Linux Enterprise Server 10.

In July, we announced SLES 10 as first to market with XEN. In September, we added support for SLES 9 and RHEL guests. In November, we announced our agreement with Microsoft to support Windows paravirtualized on SLES and SLES optimized on Windows. Both of these are important to the mixed source solution. Later in November, we announced ZENworks Virtual Machine Management and Orchestrator to manage these environments. Also, we provided our preview announcement of Open Enterprise Server 2, with NetWare paravirtualized on SUSE Linux Enterprise Server (using XEN). Clearly we are creating the mixed source environment, driving interoperability and management, and leveraging virtualization to be both a good host and a good guest.

3. The Microsoft agreement

We have already alluded to the fact that our virtualization and interoperability strategy is strengthened with our agreement with Microsoft.

It should be clear that, for the next twenty years, Windows and Linux are the two key operating environments for customers. Before this deal there were Linux focused companies and Windows focused companies like Microsoft. No company had legitimacy to assert itself as key to both environments.

That has irrevocably changed. Novell is the company which stands at the nexus of both of these key platforms. (In addition to bringing the NetWare base forward on SUSE Linux Enterprise Server as part of Open Enterprise Server.)

Why is that? Only Novell has Microsoft’s endorsement as its partner to drive Linux-Windows interoperability. So – of the Linux vendors – only we can speak authoritatively about Windows. With our virtualized solutions, we are linked even closer with the solution that customers need.

Of course, Microsoft is also at the base of Linux and Windows since we will work with them to optimize SUSE Linux Enterprise Server on Veridian. However, although they endorse SUSE at the Linux level, they do not aspire to manage heterogeneity at higher levels. Novell, on the other hand – has made multi-platform support a key requirement from day 1.

Other aspects of the Microsoft agreement

While we have focused on virtualization, there are several other significant pieces of our agreement with Microsoft. Our business collaboration (sales and marketing) is significant. Beyond that, we are collaborating technically in three additional areas related to interoperability and management.

One component is collaboration with Microsoft on standards for Web services management. Technically, there is a great deal of synergy between our two companies on how we see the proper architecture for the future. This collaboration will simplify our systems management products, since the management architectures for Linux and Windows will come together.

Another component is directories. Here, the open source world, Microsoft with Active directory, and traditional Novell products based on eDirectory require considerable joint activity. Customers regularly ask for greater interoperability. We have agreed to move this space forward.

Finally, in the area of document interoperability, we have already provided technology that allows interoperability between Open Office and Open Office XML, the basis for Vista’s Office products.

4. Building out the Management and workgroup services portfolio

We made progress in 2006 in building the management piece of our strategy. For identity and security management, 2006 was a key year to further our heterogeneous management strategy. We started the year with Identity Manager 3 – soon to be Novell’s fastest growing product. We ended the year with Access Manager 3, a breakthrough in integrating Web access control with enterprise access control. In the interim we acquired e-Security to ensure multi-platform compliance solutions. We have built out an industry leading portfolio.

Also, consider industry announcements that we led last year. Jointly with IBM and Harvard, we announced Project Higgins, a project that rationalizes Internet based identities with corporate identities. And we were the singular leader of the Bandit project – an Open Source Identity framework. Only Novell could bring Microsoft, Red Hat, IBM, Sun, and a half dozen others to the table to agree on a common approach.

In the broader systems management area, our most significant move was to move strongly into data center management. We recently unveiled three products that are focused on challenging data center issues: interoperability, virtualization, workload balancing, policy-based automation. These are ZENworks Orchestrator, ZENworks Virtual Machine Management, and ZENworks High Performance Computing. These are rather significant and I will elaborate on them in a subsequent posting.

Another key piece of this to to move our workgroup services to Linux. For Open Enterprise Server/NetWare, 2006 was the year when we started to make huge progress in transitioning the base from proprietary NetWare services to Linux based services in file, print, and storage management. In that way, we brought these significant workgroup related management services into our heterogeneous systems strategy. And we continue to strengthen GroupWise on Linux: all the while adding significant function in mobility, and broadening the value proposition with a suite approach to workgroup computing.

Summary

As I reflect on my first year at Novell it has been an exciting time. We have built a unique, durable strategy to carry us into the future. Our product teams have been executing and 2006 was a watershed year to instantiate this strategy.

Since we are beginning a new year and I have recently complete my first year at Novell, I would like to review Novell’s technical strategy. I will outline what it means to have a strategy, what Novell’s strategy is, and how our recent announcements have filled in that strategy.

What is a corporate technical strategy?

Often when one looks at a corporation, it is not clear what they are trying to achieve. Some companies are too large, and it is difficult to rationalize their entire product line together. Other companies are too small. They have interesting point products but nothing that is sustainable for the long term.

Other companies have a strategy which is not unique so they have low likelihood of success. This is often seen in start-up companies. An idea gets hot, lots of companies run after it, and few succeed. The area is not broad enough to support several competitors.

The importance of a strategy

Companies without a strategy are lost. They do not gain brand recognition. They do not build increasingly powerful solutions for customers – year after year. They have a hard time prioritizing because different constituencies cause them to invest differently. Most important, they don’t have a rudder to guide them for the long term. After all, the marketplace is constantly changing. A company with a direction can withstand small perturbations as it stays focused on a larger goal. A company without a clear direction will always be at the mercy of the winds of change.

A clear strategy therefore leads to a winning business model. A business model is fundamental because it represents a way to approach the marketplace unlike what any other company is doing. Jim Collins, in his bestseller “Good to Great,” notes that the critical success factor is to create what he calls the hedgehog concept – the concept of what is the essence of the business, the concept that no one else can figure out. There are several ingredients to long-term success. Figuring out the hedgehog concept is one of the first things a company must do.

Novell’s technical strategy

In developing Novell’s technical strategy, there were several things that were clear. The strategy must fulfill a significant area of customer need. It must be an emerging area because we must address unmet market needs. It must be durable, so we solve customer needs and establish a franchise for Novell that will succeed in 2007 and beyond. It must be a need that we can fulfill given our competencies and the value promised by our brand.

Novell is very fortunate. As opposed to many other companies we have a very clear strategy. It is based on meeting customer needs that they have today and will continue to have for the next 20 years. They are needs that fit perfectly with Novell, that we can execute against. And it is unique to Novell.

Here is the strategy. It starts with Linux and Open Source. Open Source is transforming the computing landscape. It is changing economic models and creating new winners and losers. And we are on the ground level with SUSE Linux Enterprise.

Companies like Red Hat start with Open Source and end with Open Source. That misses the market. IDC estimates that customers will spend $500 million on Open Source products in 2008. Over the last 20 years, IT organizations have spent trillions of dollars on proprietary software. Not millions. Not billions. Trillions!! That’s four orders of magnitude larger than the Open Source number.

So our strategy is a mixed source strategy. We are unique in the industry. We are focused on managing the introduction of Open Source into the existing enterprise proprietary world. We are focused on important customer problems related to interoperability.

Said differently, Linux ensures that Novell is relevant to customers. But it is not all we do. We also build management services around Linux. These are services that manage the heterogeneous Linux, Windows and UNIX worlds. These services deal with resource management, identity management, file and storage management, and the like.

Net-net: the simple statement of the strategy? The world’s best and most interoperable Linux surrounded with management services that leverage our footprint to build a business solving customer problems in heterogeneous systems management.

Assessment

This mixed source model is durable. It is hard to copy. It is hard to form that DNA. You can’t go in a lab and reverse engineer it. Who else has a world-class Linux distribution, and provides the mixed source “surround” required for success in the enterprise? So many of the themes from Novell over the last year are expressions of our hedgehog concept. We have the best enterprise Linux. Your Linux is Ready. Desktop to data center. We surround this enterprise Linux with mixed services which connect the Open Source and proprietary worlds.

And we also have a well-developed roadmap which guides when it is more appropriate to go Open Source and when it is more appropriate to go proprietary. Check out this post for a review of that strategy.

The strategy is a good fit for Novell

In history, every successful operating system company from MVS to the UNIX companies to Microsoft has used the operating system as the insertion point. You get relevance from the operating system, but you build an ecosystem of services around the OS.

There are many other companies that want to build an ecosystem for Linux. Without the operating system as the anchor point, they will fail. And unless they embrace mixed source to connect to the enterprise, they will fail. Building services around operating systems is not unique. But building mixed solutions around Linux is unique, we are the only ones doing it, and is exactly what the industry needs today.

This strategy is a good fit for many other reasons. We have the right assets in Linux and management, and we have market position. More important, it is a good fit for our brand, our culture and our history.

It is a good fit for our brand because the Novell brand has stood for enterprise quality computing – for as long as anyone can remember.

It is a good fit for our culture. We are the fusion of multiple cultures. We have the cultures of SUSE and Ximian – Open Source passion. We have the culture of Novell – solving business problems.

It is a good fit for our history because our history is to find the critical gaps and lead the industry in addressing them. In the 80’s, there was a gap in network computing and NetWare won the day. In the 90’s, there was a gap in directory-based computing, and our ZENworks and identity management solutions went on to lead the industry. More recently, there was a gap in the introduction of Linux and we are addressing it.

What’s next?

It’s fine to have a technical strategy. Much more difficult is to take steps that significantly move the ball forward in the implementation of that strategy. This posting has reviewed our going-forward strategy. In my next posting, I will reflect on steps taken in 2006 to instantiate this strategy.