Subscribe

Category Archives: Networking

The cost of Cisco UCS?

Lately I have been having a lot of discussions around Cisco UCS and the common theme of these meetings tends to be "My traditional servers cost much less." A lot of times there is a misconception of what's included with Cisco UCS compared to what's included when purchasing traditional servers. UCS is more about a server architecture and less about just your typical server blade, so you need to factor in more components when comparing it to a traditional servers. Below is a summary of components that you should consider when comparing the cost of Cisco UCS to traditional servers.  
Comments Off

What’s the “Big Deal” about Cisco UCS ver 2.0 Fabric?

What's the "big deal" about Cisco UCS ver 2.0 fabric?  Two words: bandwidth and flexibility. Cisco recently expanded the UCS product line to include an additional Fabric Interconnect (model 6248), an additional Chassis IO Module (model 2208), as well as software release 2.0 with many new features. The new 2.0 components support existing chassis and server blades, and are back-compatible to inter-operate with existing first-generation UCS hardware (in original modes of use.) Although the new models need not obsolete the original hardware, many of the recent Cisco UCS deployments by Ahead have opted to take advantage of the benefits of the new components. The new 6248 Fabric Interconnect packs 48 Unified Ports into a 1 RU form factor. Any of the base Unified Ports may be configured to ...

Read More

3 Comments

What’s the Forecast? Network…

Without the network - it really wouldn't matter what applications, servers, storage, or virtualization technologies are deployed.  The network is by definition, what brings everything together and ensures that systems and users can interact to run business.   I figured this would be a good place to start with my observations on what's coming... Network virtualization will continue to the point that traditional models of distinct location or geographic-based network topologies will in many cases no longer be sufficient to support applications.   In order to provide high availability, location independence, and application portability, many organizations will look at a massive flattening of the network infrastructure and the expansion of Layer 2 networking across datacenter locations. Traditional three tier designs and the approach of extending L3 as far as ...

Read More

Comments Off

OpenFlow Networking

Odds are that you've heard the buzz about OpenFlow. Lets take a quick look at what it is, how it works and a few benefits and challenges of the new technology. What It Is OpenFlow is a method to allow the traffic-forwarding behavior of an Ethernet switch to be controlled by an external software program. It was created by researchers at Stanford University (and others) to share physical networks between production use and experimental research use. The researchers needed a way to test new networking protocols at large scale without effecting normal user traffic. OpenFlow was developed to allow the production network to be carved into virtual segments - some segments for production use and other segments for experimental use. The production segments use traditional techniques to ...

Read More

1 Comment

Building Out Ahead’s Data Center Lab Series – Part 6 – Networking Double Vision

The only thing better than one Cisco Nexus 7010 is…. You guessed it – two Nexus 7010s. As I’d mentioned months ago in this space, the lab equipment is never going to be entirely static; we’re always try to build an environment that represents both what customers are doing in their environments today, but also represents what they might start thinking about for tomorrow. The initial round of diagrams I had put together indicated a sort of “OTV-in-a-box” (with an external L3 10GE switch to provide our L3 "cloud") approach to networking, where we were going to leverage the virtualization functionality of the Nexus 7000-series to partition 4 Virtual Device Contexts (VDCs) to parcel up the networking elements for both Data Center 1 and Data Center ...

Read More

Comments Off

Nexus Feature – FabricPath

Nexus Feature – Fabric Path April 6, 2011 In yet another of several papers discussing the attributes of Cisco’s Nexus product line, this paper will discuss the characteristics of FabricPath (FP). Most modern data centers require some form of L-2 accessibility throughout the location.  Partly this is due to some applications requiring L2 connectivity (think VMotion), and partially to ease network implementation and configuration.  If each rack within a data center was configured on a separate IP address the configuration of a server would have to be changed if they were to be moved around the data center.  In addition, the DNS/IP relationships would have to be proactively managed.  This would require the coordination of multiple teams (network, server, application, storage, etc.).  To counter this, ...

Read More

Comments Off

Nexus – 7000 Hardware

The Nexus 7000, 5000, 5500, 2000, and the 1000 devices comprise the current chassis types within the Nexus family.  Each contributes key aspects to the overall capabilities found within this cutting edge family of network hardware.  Below is a breakdown of the current hardware abilities of the Nexus 7000 chassis The 7000 currently comes in two forms.  The 7010, 10-slot chassis, and the 7018, 18-slot chassis. Chassis Types: 7010: The 7010 chassis is a cutting edge core chassis designed for high bandwidth, low-latency, next-generation network topologies.  Every attempt has been made to make this a future proof chassis for evolving technologies and physical layer exchange of data. 10-Slots: 1-4 and 7-10 are line card slots, 5-6 are supervisory ...

Read More

Comments Off

Nexus – vPC (virtual Port-Channel) Defined

Nexus Feature – vPC (virtual Port Channel) March 28, 2011 The concept of Port-Channels (PC), or Ether-Channel bundles, has been around for quite some time.  One of the major limits of a PC was that the connections were always point to point.  In a normal configuration two PC’s per access switch would be configured, one per distribution switch, to provide redundancy and fault tolerance.  In addition Rapid Spanning Tree (RSTP) was required to ensure a loop free topology between the access switch and the distribution switches.  With RSTP only one of the two PC uplinks would be active at any one time.  This limited the available bandwidth that would otherwise be available.  An example of this configuration is shown below.  Read More

Comments Off