This is the first in a series of blogs I am writing to cover technical topics as the industry gears up to build next-generation data center architectures. These blogs are intended to provide a basic overview of the innovations that Brocade and the industry at large are developing.

In today's introductory blog, we take a step back with an overview of a technology that we are very familiar with in the Ethernet world—Spanning Tree Protocol (STP). If you know Ethernet networking, you know that STP has been the primary protocol in Layer 2 LANs since the very beginnings of Ethernet. Its primary goal is to create a loop-free topology that provides a single active path between any two network endpoints, by shutting down all but one path between any two switches. While this approach has served the development of Ethernet well, STP does have bandwidth, performance, and, ultimate cost and management limitations in light of the advent of converged data center architectures.

Various networking vendors have developed mostly proprietary enhancements to STP to address these shortcomings, but these are all stopgap measures to mask network-level inefficiencies. As we move forward to the virtual data center and cloud computing architectures, these vendor-specific solutions will take too much time to reconverge when a link or switch goes down or an application migrates from one server to another. With STP, data can take a path through several switches because a shorter path is disabled. During the time it takes for links to reconfigure and reconverge, the entire network is unavailable. This is unacceptable for mission-critical applications, hence the move toward "lossless" Ethernet technologies—namely Converged Enhanced Ethernet (CEE) or Data Center Bridging (DCB). (More on this in a later post.)

What is needed is to re-architect the way traffic should flow, so that it makes better use of all the available paths. This requires the building of larger, physical/logical networks (think flatter) to account for applications such as Virtual Machine (VM) mobility across interconnected hosts within a single network. Brocade is working with the Internet Engineering Task Force (IETF) on a standard called Transparent Interconnection of Lots of Links (TRILL), which provides multiple paths via load splitting. TRILL will allow us to reclaim network bandwidth and improve utilization by establishing the shortest path through Layer 2 networks and spreading traffic more evenly. The net effect is that the network can respond faster to failures. By addressing the limitations that STP poses, the data center network will be able to scale to meet the demands of virtualized and cloud computing environments with on-demand Layer 2 network capacity.

In my next blog, I will explore the benefits of flattening your Layer 2 networks.

You hear a lot from vendors these days about the importance of being “open” in their approaches to enabling customers to get the best network and application functions to enable the customers’ core businesses. What this usually means in practice is that those vendors are “open” to others aligning with their solutions and architectures to facilitate work they’ve already done. This usually means less choice for the customer when deploying networks and apps, lessened investment protection, and faster upgrade cycles – all of which benefit only the “open” vendor.   

If you’ve seen Mike’s video from the sales conference this week, you know that, and hopefully this is no news, Brocade is VERY focused on creating partner-friendly, open-standards-based solutions for networking on both sides of the server. As I have said many times before, we don’t choose to actively compete with our partners; instead we focus on our core competency – providing exceptional end-to-end networking gear…and we embrace an open partnership ecosystem relationship with the rest.

A couple of announcements we made this week show that we’re not only talking a good game there, but playing one too, and not just with other network providers (as you have seen in the past) but with application developers as well.

First, Brocade was selected as a provider of Fibre Channel connectivity for the Microsoft SQL Server Fast Track Data Warehouse and the project code-named “Madison”— reference architectures for data warehousing based on Microsoft SQL Server 2008 Enterprise. These solutions will help customers reduce costs, save time, and reduce risk with reliable, pretested hardware configurations and best practices for data warehousing applications.

We also announced that we had unveiled a set of solutions developed specifically for use with Oracle database and application suites to enable delivery of mission-critical business systems. These are focused on data warehousing, business applications and virtualization deployments with inclusion in Oracle reference configurations.

We have a long history of working with Oracle to help enable optimization of Oracle applications with our high-performance network connectivity solutions. We are continuing to build on that foundation by unveiling key interoperability and tight integration of our Brocade storage networking solutions through the Oracle VM Storage Connect framework announced a few weeks ago at Oracle OpenWorld in San Francisco.

And as we did here, we will do with others in both applications and networking. Because the things that make a company like Brocade a good partner for solving customer problems cross technology and segment-specific barriers.

But you don’t have to take my word for it – check out what Wim Coekaerts, Oracle’s VP of Linux Engineering had to say about what makes a good technology partner, and why Brocade stands out in that regard:

There’s one on every block.

Draped in stiff, weather-resistant canvas, its iconic wheel-centered emblems just barely sheltered from the sun and rain, a classic car of yesteryear. The apple of its owner’s eye, still as shapely as ever, permanently on display in the driveway, it’s no longer fit for the road but sure to be worth big money with just a little work. If you ask the owners why they never take the stunners out anymore, the answer is almost always a variation on the same theme. “The master cylinder went out in ‘98 and I never found a replacement,” one will say. “The cost of parts for these things, especially this model, just through the roof,” says another. “Ah, I keep meaning to fix her up, but I have my (inset hybrid name here), and times are tough – hard to justify it, at least for now.”

And, inevitably, there sits the hybrid, the mini-van, the ¾ wagon or the even smaller passenger sedan – slightly less stylish but steadfastly dependable, the parts readily available, the service cost reasonable, the model name practically unknown just a few years ago.

And every day the owner throws his briefcase in, fires it up, backs it out and hits the road to work. The oil gets changed every 3-5,000 miles, the tires less often, its general operation and faithful service a non-issue in its owners mind.

But pride, promise, hope and past glory keep the aging lux-mobile in the driveway, and every once in a while the owner uncovers it, washes it, buffs it and puts the cover back, waiting for that day he finds that missing part or that financial windfall that finally allows him to move forward with that frame-up restoration.

He joins clubs and associations, buys caps and shirts, and every once in a while the maker even sends him a little badge for the dash, noting the historical significance and increasing cultural cache of his chrome-covered prize.

As with cars, so with networks. That almost every business in the world today has a network unsuited to current application, traffic and user demands is a fact. And that the supplier of those networks is tone-deaf to those businesses’ current and future needs is equally true.

That those networks were state-of-the-art and ahead of their time when their components were first developed and deployed in the mid to late 1990s is indisputable. But no matter how many chrome upgrades, transmission flushes, new tires and re-upholsteries they’ve been through, at heart, and in light of current needs, they are still based on outdated principles, goals and ideas.

And that’s to be expected, because as a company becomes a certain size, its culture, focus and allegiances begin to shift. In this case they have shifted away from the core business, here, enterprise networks – routing and switching – into a wide variety of other areas. Thirty to 50 new businesses each year, by its own account, to enable continuance of high margins and overall revenue growth. This is how companies run, and that its primary allegiance is to its shareholders and its focus is on top line growth is neither shameful nor surprising.

But it is unfortunate for its customers, those who bought the basic gear more than a decade ago and who have accepted the incremental upgrades at monumental cost with the promise of a coming revolution that would build on, not replace, their investment. Because along with the loss of focus on the core business that made it the tech bellwether it is, this company has also forgotten those promises.

A revolution is here, indeed, but like many one-time-folk-heroes who find their way to power and grow fat, lazy and disconnected as a result, the would-be revolutionary finds itself in the crosshairs of a new generation with different needs, principles and economic realities.

To mix my metaphors as I do, businesses today don’t need networks with heavy chrome bumpers, solid-steel bodies, high-displacement V8 engines and flashy hood ornaments protruding from their noses. They need new technology with an awareness of today’s customers’ needs, with higher efficiency, lower cost, and sharper focus. Companies like Brocade are meeting that need where others are not, and customers, prospects, partners, investors and employees are recognizing it.

So, that some will both continue to wax nostalgic for bygone days and long for the promised increase in value of their spring-sagging, oil-leaking, flat-tired hulks is inevitable.

Thankfully, other people just want to go to work.

From Wingspan Guest Blogger Masakazu Aoba, VP Japan Sales:

June is the month which IT professionals get together to celebrate all things IP in Japan.  It’s the month of Interop Tokyo 2009.  The biggest networking technology event in Japan has just ended and it was the first time for Brocade to participate in the Interop Tokyo event as an exhibitor.

Even in the midst o f a“once-a-100-years” recession, so many people attended the expo, especially at the Brocade booth. Some people asked, “Where is Foundry?”. I also heard someone said, “Who is Brocade?”. We definitely had to explain that Brocade acquired Foundry and as a company we have now stepped into the “End-to-End Networking” world.  It’s the beginning of our long term relationship.

At the Brocade booth, we exhibited about 40 products including newly launched ServerIron ADX, FastIron CX, and TurboIron 24X, as well as some products and solutions from partners. These  products drew much attention from the visitors, though, the greatest catch of the year was, no doubt, Brocade FCoE/CEE solution which was demonstrated live. We had a half-rack with Brocade 8000, CNA, Brocade 300, FastIron EdgeX, and a Dell storage mounted. All these devices were connected through CEE, FCoE, and FC and this was the only live FCoE/CEE solution introduced at the three-day event. In fact, Brocade 8000 won the Special Award of the Best of Show Award program of Interop Tokyo 2009 in the category of Networking Infrastructure Product for Large Engerprises. Hooray!

Cleary, one of the key objectives for us to participate in this event this year was to let everyone living in the networking world, especially the residents of IP side of the world, know about who we are and what benefits we can provide as a new combined networking solutions vendor. For those who dropped by our booth, thank you for coming and hope you enjoyed learning about Brocade and our solutions. For those who didn’t have a chance to visit our booth or maybe the event itself, you can learn about more about us at www.brocadejapan.com.

Brocade is known for a couple of things – building Extraordinary Networks, and developing and maintaining close partnerships with other companies in ways that benefit customers  beyond the benefits that they would receive if Brocade “made it all here.” One of the areas in which we partner closely with other vendors to deliver a complete solution is campus wireless access, or WiFi.  Our focus on cross-compatibility not only among our own products but with many other vendors makes our combined offering here stronger and more flexible than single-vendor solutions.   With the increasingly distributed enterprise, the benefits of a combined WiFi offering are absolutely clear.

The driver for the growth in enterprise wireless   to this point is primarily due to convenience; the increased needs of a mobile enterprise. Moving forward, one of the major trends is that  wireless  is becoming an integral part of a networking infrastructure as opposed to something that is really built for convenience and casual use. This is particularly important as the new wireless standard built in 802.11n begin to emerge, as opposed to the past standards – 802.11a, and 802.11b/g. – 802.11n can now carry traffic comparable to your wired local area network speed, which is 100 megabits per second (Mbps) and above while enhancing the robustness and range of the wireless signal. So potentially users can actually leverage wireless access to access mission critical applications and access data which normally would be too bulky or too large to be downloaded through the legacy wireless technology. With 802.11n some organizations and enterprise may actually choose to bypass the wiring of the building and go completely wireless.

As this trend increases, we will continue to see customers source and fund the deployment of wired and wireless networks at the same time, vs. doing what they have traditionally done and  treating them separately. In this context, it would seem better for them to get the entire solution from a single vendor, preferably with unique value-added functionalities – It is. It might also seem that it would be better to have the entire solution developed by that vendor –  It’s not.

Vendors who would offer you their, and only their wares from end to end in the development of a local area network are not necessarily selling you the best products for your needs, but, potentially, products that meet the lowest requirements of the stated need at the highest margins for themselves.  The Brocade approach, on the other hand, is to continuously survey the landscape for the best available technology, regardless of the source, establish its interoperability with all of your various network components, build strong relationships with the vendors who supply it and deploy it at the best possible cost/performance ratio. This strategy is enabled by the fact that we do not need to subsidize other areas of the company by charging you more for portions of an end-to-end solution than the market demands, under the guise and per the assertion that, “If we make it all it must be the best.” In fact, we have an increasing number of customers who did buy into that, and who were greatly disappointed later, and who are now turning to us for a complete replacement of parts of, if not all of their network infrastructure.

While we are happy to help in that regard, we’d rather work with them to get it right the first time – selecting from best-of-breed vendors, working to ensure interoperability and availability in IP networks with the same intensive processes we always have in the data center, and extending our partnership beyond our technology partners and to the end users themselves.

The global data center ecosystem is in the process of rapid and energetic realignment and where it will stop is anybody’s guess. With Cisco’s announcement of project California, IBM and Sun in discussions, HP coming out with a new set of servers, Dell signaling acquisition intentions in data center, and who knows what we might see tomorrow or next week, it truly is a fascinating time to be in this business.

The changes taking place in the data center permeate the entire structure, from the application all the way down to the virtual machines, the servers, the network infrastructure and the storage infrastructure. With a lot of money in play and a shrinking list of very qualified vendors in this space all trying to put the best solution forward, the stage is set for an increase in dynamism, and possible in confusion, prior to a settling of dust and a true surge in productivity.

Customers need to be careful. What they’re being promised is not always new, not always possible and not always in their best interests if it were. Both the engineering and the marketing machines of every company in the industry are being pushed to their limits with a single goal, survival through industry domination, in mind.

When Cisco came out with Project California, they put a line in the sand. Essentially, they’ve said, “We are going to deliver end to end solutions in the data center all the way from the server technology down to the networking and storage – but we’re going to do it our way, and you can do it our way, or not, but we’re not going to interoperate and we’re not going you give you choice.”

By drawing this line in the sand, they’ve just given notice to everybody in the industry – friend and foe – from large server companies to software companies to others in the networking industry that they’re going to go this one alone.

The big challenge for customers is, frankly, Cisco’s never been a server company, or an end-to-end data center provider. It has a minority, though large, share of the market in the data center network industry as it’s currently structured, and with California, it’s alienated most of the partners that got it even to that point.

So what is the next move for Brocade?

From a technology point of view we’re focusing on three areas. One, Brocade (via the Foundry acquisition) is continuing to bring out the best IP products in the market. Second is we’re going to continue to bring out market-leading fibre channel infrastructure products. Third, we will continue to focus on the converged network. Interestingly, we’re aligned here with Cisco in our view that the network is converging, servers are converging and they’re all converging onto fibre channel over Ethernet (FCoE) and converged enhanced Ethernet (CEE). Brocade is launching multiple products this year in these areas and will work closely with customers to test them and enable them to adopt them when it suits them best.

What we’re saying to our customers is that if you have an IP infrastructure there’s a clear and productive path to the future for you. If you have a fibre channel infrastructure there’s another equally clear and productive path. We also understand that the people making decisions on the fibre channel side and the people making decisions on the IP side are not the same people – but we know them both equally well and don’t favor either over the other. No customer should be forced into converged its network on any vendor’s timeframe, and the “right” time will vary widely.    

Those who would propose that you rip out everything that you have installed in your data center and replace it with a new architecture are tone deaf to the economy, and tone deaf to your needs. Customers don’t have the CAPEX budget or the people or the operations budget to implement a new architecture for it’s own sake. What customers do say is that they would like to take the infrastructure they currently have today and extend it, but still take advantage of virtualization and application mobility. Brocade’s message is, “You can do that.”