November 2010


Cisco stock has plummeted nearly 20 percent since announcing fiscal 1Q2011 results. While Cisco’s revenue and growth was solid, their forecast left investors and analysts stunned. One open source vendor thinks that open source is encouraging competition in Cisco’s arena. If so, what should Cisco do, and how should IT decision makers prepare.

Cisco’s revenue slows down
Cisco projected revenue growth of 3 to 5 percent for fiscal 2q11 and 9 to 12 percent for fiscal 2011 in total. Both rangers were below the 13 percent Wall St. expected for fiscal 2q11 and fiscal 2011.

On the earnings call, Cisco CEO John Chambers explained that a slow down of sales in the government sector was to blame:

However, we did see challenges, for example, in the U.S., say, government business, down in Q1 approximately 25% year-over-year from an orders perspective.

To give you an idea of how rapid a change this was, the state government business was down 48% from Q4 to Q1 from an orders perspective.

Vyatta CEO Kelly Herrell thinks there’s more to Cisco’s slowing revenue growth than its prospects in the government sector alone. Herrell writes in response to Cisco’s results:

Routing was a central focus of discussion by analysts, because it’s always been assumed that it was a guaranteed cash cow for Cisco. That’s no longer the case; the routing market is changing fast…

Routers are the bulk of the Layer 3 networking market. Layer 3 plays the most crucial role for Internet plumbing. This complex set of protocols enables the connectivity between devices over a distance, and between virtual machines in an integrated or cloud solution. That explains why the market for Layer 3 products is more than 2X the size of the entire market for Layer 4-7.

But the Layer 3 Overhaul is upon us now… But the new wave of L3 Infrastructure clearly isn’t coming from Cisco.

Open source enabling Cisco alternatives
In case you hadn’t guessed, Herrell’s company competes with Cisco’s router business. Herrell believes that open source software routers, such as Vyatta’s offerings, are threatening Cisco’s cash cow router business. Herrell claims: “Vendors and partners adjacent to Cisco’s router business used to recommend or resell Cisco. They no longer have to. Vyatta has removed that barrier and enabled them to add complete L3 functionality as software.”

Vyatta’s products, which range from an open source and unsupported Vyatta Core, to commercial Vyatta Subscription Edition and Vyatta PLUS, have been adopted by companies such as CBS Corp., Toyota and Rackspace. Vyatta serves up nearly 25,000 downloads of their software-based routing technology per month and is closing in on 750,000 downloads in total.

Vyatta customers typically opt to use software-based routing technology on commodity x86 hardware. Herrell claims that the resulting price-performance is far and above the price-performance of Cisco’s hardware routers in the vast majority of cases.

Give customers open source and commercial choice
Given’s Cisco’s internal experience with open source, and the lessons learned by traditional software vendors faced with open source entrants, the most effective response from Cisco would be to promote a Cisco open source offering.

Cisco’s contributions to Linux and other open source projects are well documented and marketed. As such, it’s not as if Cisco doesn’t have the internal skills or executive support to promote an open source offering to customers.

Cisco can learn from IBM and Oracle, amongst other large IT vendors, that have been able to balance a portfolio of open source and commercial products.

Cisco’s first inclination may be to lower price points or introduce a lower featured version of its router product line to compete with open source offerings.

This initial response to open source alternatives is natural. Vendors don’t typically want to open source an existing closed-source product without significant competitive pressure to do so. And sometimes even not them. Instead, vendors will introduce a new open source offering, either through a third party acquisition or in the form of a new internally developed projects meant to be open sourced from the beginning.

Consider how Oracle and Microsoft initially responded to the threat of MySQL – by offering closed source “Express” or “community” versions of their commercial products for low or no charge. Next, consider the unique and complete story that Oracle now has with both MySQL and Oracle DB in its product portfolio. Few could argue that Oracle’s database revenue has fallen off a cliff since acquiring MySQL. Next, consider the application server market, where the two revenue share leaders, IBM and Oracle, both offer open source and commercial products.

Cisco needs to offer customers the choice of open source or commercial router offerings, likely differentiated based on stand alone software versus an integrated hardware and software stack.  Being able to address the needs of customers, big and small, with varying degrees of affiliation to open source or commercial products, demonstrates an understanding that customers differ and want choice.

IT decision makers – plan for an acquisition
IT decision makers searching for options to reduce networking infrastructure costs should consider open source products such as those from Vyatta.

Buyers choosing an open source network routing provider should plan for a scenario in which the vendor is acquired by the likes of Cisco. Considering Cisco’s history of acquisitions, it’s prudent to expect Cisco to address open source competition in a core product area like their router business.  Don’t be surprised if Cisco does so in a similar fashion as other large IT vendors have successfully done – through an acquisition.

Follow me on Twitter at SavioRodrigues. I should state: “The postings on this site are my own and don’t necessarily represent IBM’s positions, strategies, or opinions.”

Red Hat released Red Hat Enterprise Linux (RHEL) 6 earlier this week. The new operating system, filled with technical innovations, performance enhancements and customer-requested improvements, has met with positive press, customer and partner response. However, how it’s being marketed could be much more important – to customers and to open source vendors in general.

Leading with lower cost
According to IDC market share figures, Red Hat is the undisputed leader in the enterprise Linux market. Red Hat achieved this position, with the help of partners such as Dell, IBM and HP, largely on a simple message of “lower cost”.

In the past, RHEL was not as feature rich as Unix variants or even Microsoft’s Windows NT, and Windows Server product line. Yet, the “lower cost” message was powerful enough to encourage customers to adopt RHEL in areas where its features, or lack thereof, didn’t impact the operating system selection process.

In October 2008, Red Hat’s CEO, Jim Whitehurst told Computerworld:

I’ve had a couple of conversations with CIOs who said “we’re a Microsoft shop and we don’t use any open source whatsoever, but we’re already getting pressure to reduce our operating costs and we need you to help put together a plan for us to help us use open source to reduce our costs.

The low cost glass ceiling
In response to this quote, I wrote that the focus on low cost risked unnecessarily limiting the growth of open source businesses.

I also wrote about Red Hat’s revenue glass ceiling as being directly attributed to its primary focus on “lower cost”. In the post, I wrote:

Yet, Red Hat’s business model, which focuses on low cost, has trained its customer base to fixate more on the price of Red Hat products than the value these products deliver. It encourages customers to trade RHEL for cheaper options.

A renewed focus on high value
It seems Red Hat has learned that “lower cost” is important, but “higher value” is much more important to customers. I’d like to say, “I told you so”, but that wouldn’t be very Canadian of me.

Look at RHEL’s marketing today, the title reads: “More reliable. More open. More comprehensive.” A search for the term “cost” returns zero hits on the RHEL for Servers marketing page.

Look at how Red Hat compares itself versus Microsoft, Oracle and VMware: “More reliable than Microsoft.” More open than Oracle” and “More comprehensive than VMware”. Not as single mention of cost.

Red Hat has also been much more vocal and upfront about technical benefits in RHEL 6. For instance, virtualized guests can achieve 85 percent to 90 percent of the performance of running on native hardware. Additionally, comparing the maximum memory or logical CPUs supported by RHEL 6 versus, for instance, Microsoft Windows Server 2008 R2, shows a wide gap in Red Hat’s favor. Specifically, 2 TB to 64 TB versus 2 TB and 32 to 4096 CPUs versus 4 to 64 CPUs, when comparing RHEL 6 versus Windows Server 2008R2 respectively.

Counter to Whitehurst’s prior suggestion that open source products do not face feature bloat risk, RHEL 6 includes 2,058 packages, and 85 percent increase over RHEL 5. This represents a significant increase in features that RHEL customers can beneit from.

What IT decision makers can expect with high value OSS
The shift from an emphasis on low cost to high value will surely become a growing trend in open source vendor marketing. This will be a positive, and important shift in the ongoing evolution of open source software vendor business models.

IT decision makers are likely to face a moderately more complex task when comparing offerings and features from competing vendors. Higher value is much more subjective than deciding between two products primarily on costs.

This shift could also result in open source vendors requiring additional professional sales resources, beyond today’s inside sales teams. IT decision makers can expect more direct sales interactions going forward than in the past.

Last, but not least, IT decision makers should plan for potential subscription cost increases as open source products begin to focus on value delivered, and price for it accordingly.

Follow me on Twitter at SavioRodrigues. I should state: “The postings on this site are my own and don’t necessarily represent IBM’s positions, strategies, or opinions.”

Faster page loads and web application performance has a significant positive impact on business results. Google knows this, as do IT decision makers. However, improving page load times is far from a simple task.

A Google-led open source project and a related commercial CDN offering from Cotendo are attempting to simplify the task of improving web application performance.

The partnership between Google and Cotendo also presents another proof point to Stephen O’Grady’s claims that companies that wouldn’t traditionally meet the “software company” definition could very well disrupt traditional software vendors.

Speeding Apache 2 based websites
The newly introduced mod_pagespeed for Apache 2 module is an open source project from Google under its Page Speed family of open source projects on Google Code.

Google describes mod_pagespeed as:

an open-source Apache module that automatically optimizes web pages and resources on them. It does this by rewriting the resources using filters that implement web performance best practices. Webmasters and web developers can use mod_pagespeed to improve the performance of their web pages when serving content with the Apache HTTP Server.

mod_pagespeed includes several filters that optimize JavaScript, HTML and CSS stylesheets. It also includes filters for optimizing JPEG and PNG images. The filters are based on a set of best practices known to enhance web page performance.

Google provides additional documentation on the filter categories, the filters themselves and associated considerations before implementing the module in an Apache HTTP Server 2.2 environment.

As with most Google open source efforts, users and prospective customers are unable to purchase support or maintenance subscriptions directly from Google.

Lack of purchasable vendor support can present a barrier to adoption in enterprises without the skills or spare cycles to maintain and support mod_pagespeed themselves.

Securing a vendor contact for mod_pagespeed
This is where CDN provider Cotendo comes in.

Cotendo worked with Google on the mod_pagespeed project and jointly presented it at ApacheCon 2010 earlier this week.

Cotendo is building mod_pagespeed into Cotendo’s commercial Dynamic Site Acceleration services. As a result, enterprises gain the benefits of mod_pagespeed and also have a vendor responsible for its maintenance and support.

Cotendo claims that the use of this open source component will offer significant benefit for its customers:

This new Cotendo service will optimize caching, reduce the number of data requests, and reduce the payload size of pages, among other features. In aggregate, these code modifications, when rendered within Cotendo’s network, can reduce image size by 20% to 30% and page load time by as much as 50% on top of the acceleration already achieved by Cotendo’s existing site acceleration services including its Dynamic Site Acceleration Service.

Adopting open source projects based on productization likelihood
It goes without saying that the performance claims above likely provided significant incentive for Cotendo to leverage mod_pagespeed versus building an alternative in house.

The more important trend for IT decision makers to recognize is the fact that third-party vendors such as Cotendo are growing their product capabilities by partnering with and leveraging work from industry titans such as Google, Facebook or Yahoo – vendors that really don’t sell software or support subscriptions.

While Google, Facebook or Yahoo typically garner significant press coverage when they open source a previously internal project, it can be a risky venture for enterprises to adopt the project in question. That is, unless a third party vendor decides to productize the open source project in a fashion that can deliver and support the code for enterprise usage.

Cotendo is doing just this for Google’s mod_pagespeed project. Not all open source projects contributed by non-traditional software vendors are as lucky.

IT decision makers are encouraged to evaluate open source project announcements not just on the merit of the technology, but the likelihood that it will be productized. Doing so will help minimize risk of using an open source project that your team must support itself, or a project that struggles to attract a large user community.  It will also encourage vendors to think about partnerships and third-party productization before open sourcing a project.

Follow me on Twitter at SavioRodrigues. I should state: “The postings on this site are my own and don’t necessarily represent IBM’s positions, strategies, or opinions.”