The Cisco Unified Computing System (UCS) is an (x86) architecture data center server platform composed of computing hardware, virtualization support, switching fabric, and management software introduced in 2009.Just-In-Time deployment of resources and 1:N redundancy can be configured with UCS systems.
Computing
The computing component of the UCS is available in two versions: the B-Series (a powered chassis and full and/or half slot blade servers), and the C-series for 19-inch racks (that can be used with fabric interconnects).The servers are marketed with converged network adapter and port virtualization.
Virtualization
Cisco UCS supports several hypervisors including VMware ESX, ESXi, Microsoft Hyper-V, Citrix XenServer and others.
Networking
The Cisco 6100 or 6200 Series switch (called a "Fabric Interconnect") provides network connectivity for the chassis, blade servers and rack servers connected to it through 10 Gigabit and Fiber Channel over Ethernet (FCoE). The Fabric Interconnects are derived from the Nexus 5000 and run NX-OS as well as the UCS Manager software. The FCoE component is necessary for connection to SAN storage, since the UCS system blade servers have very little local storage capacity.
I am more interested in this Fabric Interconnect , because i am currently exploring these for my work. so, I will provide more information in another post.
Management
Management of the system devices is handled by the Cisco UCS Manager software embedded into the 6100/6200 series Fabric Interconnect, which is accessed by the administrator through a common browser such as Internet Explorer or Firefox, or a Command line interface like Windows PowerShell or programmatically through an API.
Stateless Computing
A key benefit is the concept of Stateless Computing. Each compute node has no set configuration. MAC addresses, UUIDs, firmware and BIOS settings for example, are all configured on the UCS manager in a Service Profile and applied to the servers. This allows for consistent configuration and ease of re-purposing. A new profile can be applied within a matter of minutes.
Now That's UCS. Lets move on to "Fabric Interconnect"
The UCS fabric interconnect is part of a range of products provided by Cisco that are used to uniformly connect servers to networks and storage networks. These devices are usually installed as head units at the top of server racks. All the server components are attached to the fabric interconnect, which acts as a switch to provide access to the core network and storage networks of the data center.
The high-end model is the UCS 6296UP 96-port fabric interconnect, which is touted to promote flexibility, scalability and convergence. It has the following features:
a) Bandwidth of up to 1920 Gbps
b) High port density of 96 ports
c) High performance and low-latency capability, lossless 1/10 Gigabit Ethernet and Fiber Channel over Ethernet
d) Reduced port-to-port latency to only 2 ms
e) Centralized management under the Cisco UCS Manager
f) Efficient cooling and serviceability
g) Virtual machine-optimized services through the VM-FEX technology, which enables a consistent operational model and visibility between the virtual and the physical environments
In above diagram, The rack-mount servers are shown connected to Nexus 2232s which are nothing more than remote line-cards of the fabric interconnects known as Fabric Extenders. Fabric Extenders provide a localized connectivity point (10GE/FCoE in this case) without expanding management points by adding a switch.
UCS Logical Connectivity
In the last diagram we see several important things to note about UCS Ethernet networking:
UCS is a Layer 2 system meaning only Ethernet switching is provided within UCS. This means that any routing (L3 decisions) must occur upstream.
All switching occurs at the Fabric Interconnect level. This means that all frame forwarding decisions are made on the Fabric Interconnect and no intra-chassis switching occurs.
The only connectivity between Fabric Interconnects is the cluster links. Both Interconnects are active from a switching perspective but the management system known as UCS Manger (UCSM) is an Active/Standby clustered application. This clustering occurs across these links. These links do not carry data traffic which means that there is no inter-fabric communication within the UCS system and A to B traffic must be handled upstream.
The Fabric Interconnects themselves operate at approximately 3.2us (micro seconds), and the Fabric Extenders operate at about 1.5us. This means total roundtrip time blade to blade is approximately 6.2us right inline or lower than most Access Layer solutions.
For other questions like , The question then becomes how is traffic between fabrics handled? - Please refer to link xxx in references section.
References :
1) http://en.wikipedia.org/wiki/Cisco_Unified_Computing_System
2) https://interestingevan.wordpress.com/tag/fabric-interconnect/
3) http://www.techopedia.com/definition/30473/ucs-fabric-interconnect
4) http://www.definethecloud.net/inter-fabric-traffic-in-ucs/
0 comments :
Post a Comment