Something up Brocade’s Sleeve, and it looks Good

Brocade’s got some new tricks up their sleeve and they look good.  For far too long Brocade fought against convergence to protect its FC install base and catch up.  This bled over into their Ethernet messaging and hindered market growth and comfort levels there.  Overall they appeared as a company missing the next technology waves and clinging desperately to the remnants of a fading requirement: pure storage networks.  That has all changed, Brocade is embracing Ethernet and focusing on technology innovation that is relevant to today’s trends and business.

The Hardware:

Brocade’s VDX 8770 (http://www.brocade.com/downloads/documents/data_sheets/product_data_sheets/vdx-8770-ds.pdf) is their flagship modular switch for Brocade VCS fabrics.  While at first I scoffed at the idea of bigger chassis switches for fabrics, it turns out I was wrong (happens often.)  I forgot about scale.  These fabrics will typically be built in core/edge or spine leaf/designs, often using End of Row (EoR) rather than Top of Rack (ToR) designs to reduce infrastructure.  This leaves max scalability bound by a combination of port count and switch count dependent on several factors such as interconnect ports.  Switch count will typically be limited by fabric software limitations either real or due to testing and certification processes.  Having high density modular fabric-capable switches helps solve scalability issues.

Some of the more interesting features:

  • Line-rate 40GE
  • “Auto-trunking” ISLs (multiple links between switches will bond automatically.)
  • Multi-pathing at layers 1, 2 and 3
  • Dynamic port-profile configuration and migration for VM mobility
  • 100GE ready
  • 4us latency with 4TB switching capacity
  • Support for 384,000 MAC addresses per fabric for massive L2 scalability
  • Support for up to 8000 ports in a VCS fabric
  • 4 and 8 slot chassis options
  • Multiple default gateways for load-balancing routing

The Software:

The real magic is Brocade’s fabric software.  Brocade looks at the fabric as the base on which to build an intelligent network, SDN or otherwise.  As such the fabric should be: resilient, scalable and easy to manage.  In several conversations with people at Brocade it was pointed out that SDN actually adds a management layer.  No matter how you slice it the SDN software overlays a physical network that must be managed.  Minimizing configuration requirements at this level simplifies the network overall.  Additionally the fabric should provide multi-pathing without link blocking for maximum network throughput. 

Brocade executes on this with VCS fabric.  VCS provides an easy to set up and manage fabric model.  Operations like adding a link for bandwidth are done with minimal configuration through tools like “auto-trunking.’  Basically ports identified as fabric ports will be built into the network topology automatically.  They also provide impressive scalability numbers with support for 384,000 MACs, 352,000 IPv4 routes, 88,000 IPv6 routes, and 8000 ports.

One surprise to me was that Brocade is doing this using custom silicon.  With companies like Arista and Nicira (now part of VMware) touting commodity hardware as the future, why is Brocade spending money on silicon?  The answer is in latency.  If you want to do something at line-rate it must be implemented in hardware.  Merchant silicon is adept at keeping cutting edge at things like switching latency and buffering but is slow to implement new features.  This is due to addressable market.  Merchant silicon manufacturers want to ensure that the cost of hardware design and manufacturing will be recouped through bulk sale to multiple manufacturers.  This means features must have wide applicability and typically be standards driven before being implemented.

Brocade saw the ability to innovate with features while maintaining line-rate as an advantage worth the additional cost.  This allows Brocade to differentiate themselves, and their fabric, from vendors relying solely on merchant silicon.  Additionally they position they’re fabric as enough of an advantage to be worth the additional cost when implementing SDN for reasons listed above.

Summary:

Brocade is making some very smart moves and coming out from under the FC rock.  The technology is relevant and timely, but they will still have an uphill battle gaining the confidence of network teams.  They will have to rely on their FC data center heritage to build that confidence and expand their customer base.  The key now will be in execution, it will be an exciting ride.

GD Star Rating
loading...

Much Ado About Something: Brocade’s Tech Day

Yesterday I had the privilege of attending Brocade’s Tech Day for Analysts and Press.  Brocade announced the new VDX 8770, discussed some VMware announcements, as well as discussed strategy, vision and direction.  I’m going to dig in to a few of the topics that interested me, this is no way a complete recap.

First in regards to the event itself.  My kudos to the staff that put the event together it was excellent from both a pre-event coordination and event staff perspective.  The Brocade corporate campus is beautiful and the EBC building was extremely well suited to such an event.  The sessions went on smoothly, the food was excellent and overall it was a great experience.  I also want to thank Lisa Caywood (@thereallisac) for pointing out that my tweets during the event were more inflammatory then productive and outside the lines of ‘guest etiquette.’  She’s definitely correct and hopefully I can clear up some of my skepticism here in a format left open for debate, and avoid the same mistake in the future.  That being said I had thought I was quite clear going in on who I was and how I write.  To clear up any future confusion from anyone:  if you’re not interested in my unfiltered, typically cynical, honest opinion don’t invite me, I won’t take offense.  Even if you’re a vendor with products I like I’ve probably got a box full of cynicism for your other product lines.

During the opening sessions I observed several things that struck me negatively:

  • A theme (intended or not) that Brocade was being lead into new technologies by their customers.  Don’t get me wrong, listening to your customers and keeping your product in line with their needs is key to success.  That being said if your customers are leading you into new technology you’ve probably missed the boat.  In most cases they’re being lead there by someone else and dragging you along for the ride, that’s not sustainable.  IT vendors shouldn’t need to be dragged kicking and screaming into new technologies by customers.  This doesn’t mean chase every shiny object (squirrel!) but major trends should be investigated and invested in before you’re hearing enough customer buzz to warrant it.  Remember business isn’t just about maintaining current customers it’s about growing by adopting new ones.  Especially for public companies stagnant is as good as dead.
  • The term “ Ethernet Fabric” which is only used by Brocade, everyone else just calls it fabric.  This ties in closely with the next bullet.
  • A continued need to discuss commitment to pure Fibre Channel (FC) storage.  I don’t deny that FC will be around for quite some time and may even see some growth as customers with it embedded will expand.  That being said customers with no FC investment should be avoiding it like the plague and as vendors and consultants we should be pushing more intelligent options to those customers.  You can pick apart technical details about FC vs. anything all day long, enjoy that on your own, the fact is two fold: running two separate networks is expensive and complex, the differences in reliability, performance, etc. are fading if not gone.  Additionally applications are being written in more intelligent ways that don’t require the high availability, low latency silo’d architecture of yester year.  Rather than clinging to FC like a sinking ship vendors should be protecting customer investment while building and positioning the next evolution.  Quote of the day during a conversation in the hall: “Fibre channel is just a slightly slower melting ice cube then we expected.’
  • An insistence that Ethernet fabric was a required building block of SDN.  I’d argue that while it can be a component it is far from required, and as SDN progresses it will be irrelevant completely.  More on this to come.
  • A stance that the network will not be commoditized was common throughout the day.  I’d say that’s either A) naïve or B) posturing to protect core revenue.  I’d say we’ll see network commoditization occur en mass over the next five years.  I’m specifically talking about the data center and a move away from specialized custom built ASICS, not the core routers, and not  the campus.  Custom silicon is expensive and time-consuming to develop, but provides performance/latency benefits and arguable some security benefits.  As the processor and off the shelf chips continue to increase exponentially this differentiator becomes less and less important.  What becomes more important is rapid adaption to new needs.  SDN as a whole won’t rip and replace networking in the next five years but it’s growth and the concepts around it will drive commoditization.  It happened with servers, then storage while people made the same arguments.  Cheaper, faster to produce and ‘good-enough’ consistently wins out.

On the positive side Brocade has some vision that’s quite interesting as well as some areas where they are leading by filling gaps in industry offerings.

  • Brocade is embracing the concept of SDN and understands a concept I tweeted about recently: ‘Revolutions don’t sell.’  Customers want evolutionary steps to new technology.  Few if any customers will rip and replace current infrastructure to dive head first into SDN.  SDN is a complete departure from the way we network today, and will therefore require evolutionary steps to get there. This is shown in their support of ‘hybrid’ open flow implementations on some devices.  This means that OpenFlow implementations can run segregated alongside traditional network deployments.  This allows for test/dev or roll-out of new services without an impact on production traffic.  This is a great approach where other vendors are offering ‘either or’ options.
  • There was discussion of Brocade’s VXLAN gateway which was announced at VMworld.  To my knowledge this is the first offering in this much needed space.  Without a gateway VXLAN is limited to virtual only environments. This includes segregation from services provided by physical devices.  The Brocade VXLAN gateway will allow the virtual and physical networks to be bridged. (http://newsroom.brocade.com/press-releases/brocade-adx-series-to-unveil-vxlan-gateway-and-app-nasdaq-brcd-0923542) To dig deeper on why this is needed check out Ivan’s article: http://blog.ioshints.info/2011/10/vxlan-termination-on-physical-devices.html.
  • The new Brocade VDX 8770 is one bad ass mamma jamma.  With industry leading latency and MAC table capacity, along with TRILL based fabric functionality, it’s built for large scalable high-density fabrics.  I originally tweeted “The #BRCD #VDX8770 is a bigger badder chassis in a world with less need for big bad chassis.” After reading Ivan’s post on it I stand corrected (this happens frequently.)  For some great perspective and a look at specs take a read: http://blog.ioshints.info/2012/09/building-large-l3-fabrics-with-brocade.html.

On the financial side Brocade has been looking good and climbed over $6.00 a share.  There are plenty of conversations stating some of this may be due to upcoming shifts at the CEO level.  They’ve reported two great quarters and are applying some new focus towards federal government and other areas lacking in recent past. I didn’t dig further into this discussion.

During lunch I was introduced to one of the most interesting Brocade offerings I’d never heard of: ‘Brocade Network Subscription”: http://www.brocade.com/company/how-to-buy/capital-solutions/index.page.  Basically you can lease your on-prem network from Brocade Capitol.  This is a great idea for customers looking to shift CapEx to OpEx which can be extremely useful.  I also received a great explanation for the value of a fabric underneath an SDN network from Jason Nolet (VP of Data Center Networking Group.)  Jason’s position (summarized) is that implementing SDN adds a network management layer, rather than removing one.  With that in mind the more complexity we remove from the physical network the better off we are.  What we’ll want for our SDN networks is fast, plug-and-play functionality with max usable links and minimal management.  Brocade VCS fabric fits this nicely.  While I agree with that completely I ‘d also say it’s not the only way to skin that particular cat.  More to come on that.

For the last few years I’ve looked at Brocade as a company lacking innovation and direction.  They clung furiously to FC while the market began shifting to Ethernet, ignored cloud for quite a while, etc.  Meanwhile they burned down deals to purchase them and ended up where they’ve been.  The overall messaging, while nothing new, did have undertones of change as a whole and new direction.  That’s refreshing to hear.  Brocade is embracing virtualization and cloud architectures without tying their cart to a single hypervisor horse.  They are positioning well for SDN and the network market shifts.  Most impressively they are identifying gaps in the spaces they operate and executing on them both from a business and technology perspective.  Examples of this are Brocade Network Subscription and the VXLAN gateway functionality respectively.

Things are looking up and there is definitely something good happening at Brocade.  That being said they aren’t out of the woods yet.  For them, as a company, purchase is far fetched as the vendors that would buy them already have networking plays and would lose half of Brocade’s value by burning OEM relationships with the purchase.  The only real option from a sale perspective is for investors looking to carve them up and sell off pieces individually.  A scenario like this wouldn’t bode well for customers.  Brocade has some work to do but they’ve got a solid set of products and great direction.  We’ll see how it pans out.  Execution is paramount for them at this point.

Final Note:  This blog was intended to stop there but this morning I received an angry accusatory email from Brocade’s head of corporate communications who was unhappy with my tweets.  I thought about posting the email in full, but have decided against it for the sake of professionalism.  Overall his email was an attack based on my tweets.  As stated my tweets were not professional, but this type of email from someone in charge of corporate communications is well over the top in response.  I forwarded the email to several analyst and blogger colleagues, a handful of whom had similar issues with this individual.  One common theme in social media is that lashing out at bad press never does any good, a senior director in this position should know such, but instead continues to slander and attack.  His team and colleagues seem to understand social media use as they’ve engaged in healthy debate with me in regards to my tweets, it’s a shame they are not lead from the front.

GD Star Rating
loading...

The Brocade FCoE Proposition

I recently realized that I, like a lot of the data center industry, have completely forgotten about Brocade recently.  There has been little talked about on their FCoE front, Fibre Channel Front, or CNAs.  Cisco and HP have been dominating social media with blade and FCoE battles, but I haven’t seen much coming from Brocade.  I thought it was time to take a good look.

The Brocade Portfolio:

Brocade 1010 and 1020 CNAs The Brocade 1010 (single port) and Brocade 1020 (dual port) Converged Network Adapters (CNAs) integrate 10 Gbps Ethernet Network Interface Card (NIC) functionality with Fibre Channel technology—enabling transport over a 10 Gigabit Ethernet (GbE) connection through the new Data Center Bridging (DCB) and Fibre Channel over Ethernet (FCoE) protocols, providing best-in-class LAN connectivity and I/O consolidation to help reduce cost and complexity in next-generation data center environments.
Brocade 8000 Switch The Brocade 8000 is a top-of-rack link layer (Layer 2) CEE/FCoE switch with 24 10 Gigabit Ethernet (GbE) ports for LAN connections and eight Fibre Channel ports (with up to 8 Gbps speed) for Fibre Channel SAN connections. This reliable, high-performance switch provides advanced Fibre Channel services, supports Ethernet and CEE capabilities, and is managed by Brocade DCFM.
Brocade FCOE10-24 Blade The Brocade FCOE10-24 Blade is a Layer 2 blade with cut-though non-blocking architecture designed for use with Brocade DCX and DCX-4S Backbones. It features 24 10 Gbps CEE ports and extends CEE/FCoE capabilities to Brocade DCX Backbones, enabling end-of-row CEE/FCoE deployment. By providing first-hop connectivity for access layer servers, the Brocade FCOE10-24 also enables server I/O consolidation for servers with Tier 3 and some Tier 2 applications.

Source: http://www.brocade.com/products-solutions/products/index.page?dropType=Connectivity&name=FCOE

The breadth of Brocade’s FCoE portfolio is impressive when compared to the other major players: Emulex and Qlogic with CNAs, HP with FlexFabric for C-Class and H3C S5820X-28C Series ToR, and only Cisco providing a wider portfolio with an FCoE and virtualization aware I/O card (VIC/Palo), blade switches (Nexus 4000), ToR/MoR switches (Nexus 5000), and an FCoE Blade for the Nexus 7000.  This shows a strong commitment to the FCoE protocol on Brocade’s part, as does there participation on the standards body.

Brocade also provides a unique ability to standardize on one vendor from the server I/O card, through the FCoE network to the Fibre Channel (FC) core switching.  Additionally using the 10-24 blade customers can collapse the FCoE edge into their FC core providing a single hop collapsed core mixed FCoE/FC SAN.  That’s a solid proposition for a data center with a heavy investment in FC and a port count low enough to stay within a single chassis per fabric.

But What Does the Future Hold?

Before we take a look at where Brocade’s product line is headed, let’s look at the purpose of FCoE.  FCoE is designed as another tool in the data center arsenal for network consolidation.  We’re moving away from the cost, complexity and waste of separate networks and placing our storage and traditional LAN data on the same infrastructure.  This is similar to what we’ve done in the past in several areas, on mainframes we went from ESCON to FICON to leverage FC, our telephones went from separate infrastructures to IP based, we’re just repeating the same success story with storage.  The end goal is everything on Ethernet.  That end goal may be sooner for some than others, it all depends on comfort level, refresh cycle, and individual environment.

If FCoE is a tool for I/O consolidation and Ethernet is the end-goal of that, then where is Brocade heading?

This has been my question since I started researching and working with FCoE about three years ago.  As FCoE began hitting the mainstream media Cisco was out front pushing the benefits and announcing products, they were the first on the market with an FCoE switch, the Nexus 5000.  Meanwhile Brocade and others were releasing statements attempting to put the brakes on.  They were not saying FCoE was bad, just working to hold it off.

This makes a lot of sense from both perspectives, the core of Cisco’s business is routing and switching therefore FCoE is a great business proposition.  They’re also one of the only two options for FC switching in the enterprise (Brocade and Cisco) so they have the FC knowledge.  Lastly they had a series of products already in development. 

From Brocade’s and others perspectives they didn’t have products ready to ship, and they didn’t have the breath and depth in Ethernet so they needed time.  The marketing releases tended to become more and more positive towards FCoE as their products launched.

This also shows in Brocade’s product offering, two of the three products listed above are designed to maintain the tie to FC.

Brocade 8000:

This switch has 24x 10GE ports and 8x 8Gbps FC ports.  These ports are static onboard which means that this switch is not for you if:

  • You just need 10GE (iSCSI, NFS, RDMA, TCP, UDP, etc.)
  • You plan to fully migrate to FCoE (The FC ports then go unused.)
  • You only need FCoE, small deployment using FCoE based storage which is available today.

In comparison the competing product is the Nexus 5000 which has a modular design allowing customers to use all Ethernet/DCB or several combinations of Ethernet and FC at 1/2/4/8 Gbps.

Brocade FCoE 10/24 Blade:

This is an Ethernet blade for the DCX Fibre Channel director.  This ties Brocade’s FCoE director capabilities to an FC switch rather than Ethernet switch.  Additionally this switch only supports directly connected FCoE devices which will limit overall scalability.

In comparison the Cisco FCoE blade for the nexus 7000 is a DCB capable line card with FCoE capability by years end.  This merges FCoE onto the network backbone where it’s intended to go.

Summary:

If your purpose in assessing FCoE is to provide a consolidated edge topology for server connectivity tying it back to a traditional FC SAN then Brocade has a strong product suite for you.  If you’re end goal is consolidating the network as a whole then it’s important to seriously consider the purchase of FC based FCoE products.  That’s not to say don’t buy them, just understand what you’re getting, and why you’re getting it.  For instance if you need to tie to a Fibre Channel core now and don’t intend to replace that for 3-5 years then the Brocade 8000 may work for you because it can be refreshed at the same time.

Several options exist for FCoE today and most if not all of them have a good fit.  Assess first what your trying to accomplish and when, then look at the available products and decide what fits best.

GD Star Rating
loading...