With Oracle Openworld running this week it’s a good time to think about the performance of applications over the wide area network (WAN). With the changing business environment IT organizations are re-architecting how they deploy enterprise applications. Oracle applications automate business processes and increase productivity, but in this new network-centric world their performance is being impacted by longer WAN links and inefficient Internet protocols resulting in reduced performance and decreased user productivity.
Globalization and outsourcing has created a workforce that is spread across the world. At the same time, to reduce IT costs resulting from the spread of regional systems, and to take advantage of large scale virtualized systems, organizations are centralizing their application and IT infrastructure in the data center. The result of this trend is that the distance from the end user to the application server has increased dramatically and the connection is often over links with limited bandwidth.
Well, boys and girls, I have got a couple of cool things to report on the Cisco Unified Computing System front. First, and foremost, I am happy to present the latest iteration of the UCS complementary to the blade server : the C-Series rack- mount servers.
The Cisco UCS C210 M1 server is a general-purpose, two-socket, 2RU rack-mount server. Housing up to 16 internal disk drives for up to 8 TB of storage, the UCS C210 M1 is designed to balance performance, density, and efficiency for workloads requiring economical, high-capacity, reliable, internal storage
The Cisco UCS C200 M1 server is a two-socket, 1RU rack-mount server designed to balance simplicity, performance, and density for production-level virtualization, web infrastructure, and other mainstream data center workloads
You can follow each of the product links for details, but here is a quick snapshot of the differences between the models. The C200 and C210 will be available in November and the C250 well be here in time to tuck under the Christmas tree of that special server geek on your Christmas list.
On the UCS traction front, Gartner has released their Blade Server Magic Quadrant. While licensing restrictions preclude me from showing you the quadrant, I can discuss the results. Gartner has placed us in the Visionaries quadrant, which we believe is great place for us to debut (the positions in the Quadrant are based on completeness of vision and ability to execute). Since we are newcomers to this market, this is a fair assessment as we have to prove ourselves against the incumbent vendors. As we continue to deliver product innovation and increase customer momentum, we expect our position in the Magic Quadrant to change over time.
Finally, Gartner noted that they see an uptick in the number of conversations that they are having around UCS--that there is a high level of market interest. I was have having coffee with my old boss yesterday (he is now back in the field) and he mirrored this sentiment--every one of his customers--and these are large enterprise accounts--is interested in the UCS, regardless of what they already have in place.
OK so first of all, I’d like to thank all the folks here who took time to post extremely useful comments on my previous blog “ So What Exactly is a Nexus 4000?” . I’m glad to see that there is a lot of interest in this product. Also, glad to see that in terms of understanding and perception of the Nexus 4000 functionality, largely everyone got it right.
I’ve posted some details of Nexus 4000 along with some use cases below. I’ve also tried to respond to each of your comments individually towards the end of this blog. As we mentioned during the announcement, this is a general high level functionality introduction of Nexus 4000 Series. Our blade server partners will be making more details available as it relates to their blade chassis so stay tuned for more information. However, please do continue to post your comments & feedback around Nexus 4000 on this blog…
So what is a Nexus 4000?
The Cisco Nexus 4000 Series is a family of blade “switches” for scale-out x-86 blade servers ( non-Cisco). Nexus 4000 is NOT a Fabric Extender (aka FEX), Nexus 4000 is a “Switch”.
Nexus 4000 Series Blade Switches are the fourth generation of blade switches from Cisco. It will fit inside blade server I/O slots to provide network connectivity for blade servers. The first generation blade switch was the Cisco IGESM blade switch. The second generation was the Cisco Catalyst Blade Switch 3000 family , the third generation is the Cisco Catalyst Blade Switch 3100 family (aka VBS). And now the fourth generation is the Nexus 4000 Blade Switches.
So Cisco is not new to the blade switch market, in fact we are clear leader in blade switch market space both in terms of installed base of blade switches & market share even with our Catalyst blade switch product line.
This newest addition to the Cisco Nexus family of data center products, the Nexus 4000 Series will extend the benefits offered by the Nexus family to our partner blade servers. More specifically it provides the Unified Fabric functionality to our partner blade servers.
For starters, Nick Lippis posted a well thought out blog on his short list of cathedral-builders (Cisco, HP, and IBM). Yes, I know I have a bias, but these three companies would be my top 3 as well. I think all three companies have the correct blend of vision and completeness of footprint. If I were to add a fourth, it would probably be either Oracle or Dell--both companies are certainly working hard on carving out a broader role.
Over at Gigaom.com, Gary Orenstein pondered if networking would become the final area of differentiation in the data center. Certainly, I am not going to be upset if networking capability becomes the primary point of differentiation between Cisco, HP, and IBM :), but, sadly, I don’t think it will be that simple. I do, however, think he brings up an interesting dynamic for product vendors (the brick sellers in my original post). If networking becomes a key point of differentiation, I don’t think the current arms-dealer approach of selling their wares to everyone and anyone will fly any longer. If you are selling bricks with no compelling differentiation, then I think your business gets commoditized and the profit is driven out of it. On the other hand, if you do have compelling differentiation, then I think you get get snapped up by one of the cathedral-builders to give them a point of differentiation in the broader data center battle that Gary talks about.
All, of which, perhaps helps explain the reports that surfaced this week that Brocade is shopping itself around. This type of chatter usually makes customers nervous (often with good reason), but in the scheme of things, I can see why Brocade would want to take that risk and leverage its Fibre Channel expertise to find a home with one of the cathedral-builders in the market.
Anyway, I think it will be interesting to see what evolves over the next 12-18 months--how the cathedral builders flesh out their portfolio and what happens to the remaining brick sellers.