Today, we are announcing a number of new UCS products. But this is not just a product launch; it’s arguably the biggest product launch since the original UCS unveiling in 2009, both in terms of breadth and importance. One of the products that we are unleashing is the Cisco UCS M-Series Modular Server, a brand new product line within Cisco UCS.
So important is the M-Series that I thought it was worth a quick blog to help frame it and show both how unique it is, and how it’s changing the game and really defining a new class of server.
We named the product M-Series to highlight its modularity, but that is only part of the story. We have also taken to using the term ‘disaggregated server’ to help describe it. I think it is an apt term, but want to further elaborate, as that alone does not do the M-Series justice.
As cloud scale applications become more prevalent and evolve in the datacenter the infrastructure needs change too. Attributes like scale, performance and lifecycle require more flexibility than a traditional server can deliver efficiently. What M-Series does is separate the CPU and memory complex from the other subsystems that would make up a traditional server. Thus, when matching a particular application’s requirements to server infrastructure, M-Series allows for the independent scaling of all of the subsystems. No longer does the need for a few more cores and GB’s of memory dictate the need to duplicate the entire traditional server with more power supplies, more fans, more raid controllers, more network adapters, etc. And the same holds true for upgrading subsystems. The lifecycle of an Ethernet adapter is not the same as a RAID controller and not the same as a CPU. M-Series frees customers from the constraints.
Simply taking a server apart or disaggregating the subsystems alone is not what makes M-Series a game-changer, but creating a fabric that extends the benefits of UCS Manager all the way into the server is. What sets UCS M-Series apart is something we call UCS System Link Technology. This silicon-based technology gives M-Series the ability to connect these disaggregated subsystems via a UCS System Link fabric and create a truly composable infrastructure. With the policy-based provisioning, deployment and management of UCS, the application can be presented the exact amount of resources from each subsystem for truly a software-defined server.
Similar to how a hypervisor parses resources from a traditional server and presents them in a virtual machine, M-Series now can do that and more, but at a hardware level. A collection of these new composite machines defined from the M-Series infrastructure is the most efficient and agile way to power the emerging class of cloud scale applications. Cisco UCS M-Series with System Link Technology is defining a new class of server architecture to power cloud scale applications.
Please watch this space for more information on Cisco UCS M-Series, as we will dive deeper into this new and exciting product.