The technical pathways of Brocade in cloud storage adoption

Print Friendly, PDF & Email

Brocade isn’t always very forthcoming about what they are working on. Obviously a fair chunk of development and engineering efforts are spent on cloud integration  and enablement of their software and hardware stack into this computing methodology. Acquisitions like Foundry, Vyatta and now Connectem show that the horizon has broadened the views of Brocade. To keep up with the ever increasing demands for network features and functions it makes sense to review the current product lines they have and when you read between the lines you may be able to spot some interesting observations.

Cloud Storage

Obviously the sands are shifting for all companies that are involved in IT development. This goes across the board from both hardware and software vendors. Arguably initiated by the enormous influx of server virtualization and the accompanying swarm of management changes the way computing has been will never return. Large scale mainframe architectures and client/server computing will decrease ever more. I’m not saying they will disappear but they will be declining.

All this virtualization and cloud-enablement also sees an increase is changes how data is stored and managed. The limits of traditional storage in SAN based environment become very clear in the multi-petabyte scale. Even with all the smarts and grunt that have been put into these machines you will hit a brick wall at some stage when your Openstack, Hyper-V or VMware clusters expand.

The way data is stored and retrieved in these kind of environments is ever so more relying on two-dimensional scaling. Horizontal  scaling provides a near infinite access method where processing power and capacity is not containerized in individual machines (arrays, servers) but is abstracted by a highly resilient and flexible software layer. The vertical scale provides the ability to obtain a tiered architecture in that same storage environment but is also handled by the software layer.

All vendors have jumped on this bandwagon and especially the networking people seem to be extremely hyper-active on this. Every homepage of every vendor mentions the word “Cloud” and “SDN -Software Defined Networking”. Given the fact the competition is fierce in this area the marketing departments are doing a fair bit of overtime.

Brocade

So how does Brocade seem to handle all this? Being a storage networking vendor by nature the acquisition of Foundry gave them a very strong foothold in the networking space. This left them with the problem that overall management of these two distinctly different technologies needed to be addressed. To a certain extent they have by incorporating IP management capabilities in their Brocade Network Advisor (BNA) suite but the majority of the Foundry equipment still has its own management tool. The VDX platform, which is their core Ethernet Fabric product, is well equipped for data-centre operations from an Ethernet perspective plus it allows FC connectivity in some of these switches.




From a hardware engineering perspective you can almost predict when they will have a single platform for both Fibre-Channel and Ethernet. By looking at the development stages of both FOS and NOS and correlating the capabilities of the Condor, HAWK and Wolverine ASIC’s it is just a matter of time when Brocade will have a “hybride” ASIC where the required functionality is either determined by the attached physical media (FC-SFP or ETH-SFP) or, when even that becomes a common ground, a software-based configuration setting. The back-office of the Brocade VDX Ethernet product line is already handled by the Condor 3 Fibre-Channel ASIC. All internal chassis and external  fabric traffic is already handled by this chip for both FC and Ethernet. (I joked once about the fact that Brocade had flipped FCoE to EoFC but that is actually somewhat a reality.)

It makes sense for Brocade to work towards this common ASIC as it will have huge market value and will provide great investment protection for customers. To be able to migrate almost transparently between FC and Ethernet without the need of forklift upgrades is very appealing to any CIO and CFO. It will also provide Brocade a more cohesive engineering incentive where features like RAS (Reliability, Availability and Serviceability) can be significantly enhanced as these engineers can focus on a single platform. The overall management complexity will be reduced as well which also benefits customers from an OPEX perspective and, again, speeds up development of BNA (or whatever it will be called then) concurrently with a huge cost reduction of development resources. I just hope Brocade does not use this as an excuse to slash engineering jobs as there is still a lot do do in this arena.

Cisco, although still having the MDS platform, have a common chassis (Nexus) which allows Ethernet and FC. There is no hybrid ASIC available as far as I know as their . Also Juniper and Arista, although not playing in the FC world, have a fairly broad portfolio which supports the new software defined storage paradigm.

Conclusion

The simplification of networking, storage and compute in the private, public and hybrid cloud environments is paramount in order for adoption ratios to increase. Large scale technology driven companies like Paypal, E-Bay, Microsoft and Yahoo etc. have no problem being at the start of the hype-cycle given the sheer amount of technical resources and PhD grade brain capacity to their disposal. To be able to have less fortunate companies adopting these same technologies the underlying platforms need to be abstracted with an umbrella layer for customer consumption. Initiatives to consolidate and simplify the complex technologies and therefore be able to re-align development resources to create these customer grade tools and platforms will determine which vendor will be at the forefront of the pack.

There is no winner or loser yet in this space. Object based storage solutions driven by large scale software defined storage and networking infrastructures are still in their infancy. Many software storage solutions like Gluster, Ceph, Swift and Cinder in addition to flexible filesystems like Btrfs still have to prove their wits when it comes to assuring customers they are bulletproof and don’t give them any surprises. (where surprise == “data-loss”).

It’ll be interesting to see how it develops. I’ll keep a keen eye on Brocade and Cisco to see where things are heading.

 

Cheers

Erwin

About Erwin van Londen

Master Technical Analyst at Hitachi Data Systems
Brocade, Storage Networking , , , , ,

Comments are closed.