So, lets take a closer at OTV and how it works. As a reminder, OTV is an NX-OS feature that allows us to extend Ethernet LANs between data centers. One of the nice things about OTV is that it is transport agnostic–the connectivity between data centers can be L2 based, L3 based, IP switched or label switch–pretty much anything that can transport IP.
OTV works by creating an OTV control plane through authenticated links between the Nexus 7000 switches at each of your data centers (called edge nodes in OTV parlance). You can then “route” your LAN traffic by encapsulating it and routing it through this IP infrastructure. Routing of the traffic is determined by associating a MAC address with a next-hop IP address. The process is fully dynamic, so there is no need to establish and manage tunnels and virtual wires. This approach certainly simplifies management and administration over existing approaches, but it also allows you to take full advantage of your IP core such as optimal routing and features such as load balancing, multicast traffic replication, and fast failover.
In their Q4 2009 WAVE report Forrester recognized Cisco WAAS as a market leader in WAN optimization in their comparison of the top 8 vendors, considering criteria such as the offering, strategy, market presence, and technology.
Forrester says that WAN optimization is increasing in importance to IT organizations and that business issues such as consolidating branch office servers and rolling out new collaborative applications are driving its adoption. This is because companies rely on the WAN for delivery of their business-critical services, but the WAN suffers from poor performance because of latency over distance.
Forrester advises that WAN optimization technology plays a critical role in improving application performance by using techniques such as caching, protocol optimization, compression, traffic management and quality of service (QoS), to increase effective throughput and mitigate latency, while providing visibility into the traffic mix.
Organizations are ready to promote greater value with virtualization, however, data center complexity and costs are skyrocketing. Virtual machines have emerged as the new “atomic unit,” moving fluidly throughout the data center, promising reduced costs and increased flexibility.
Hindered by decreasing budgets and accidental architectures of the past, IT managers are struggling to realize the full potential of virtualization.
IT leaders must navigate a complex maze; combining diverse solutions and manual processes in the quest to cost effectively refresh aging technology, adhere to industry standards, and plan a safe path to the future.
Strategic architecture decisions made today will largely determine an organization’s competitive position for the next economic cycle.
As a result, core assumptions about infrastructure and data center strategies are shifting radically.
The key to a more efficient, simplified, and manageable data center lies with Cisco Unified Computing. This integrated architectural approach is built upon the virtual machine as the core element, uniting virtualized compute, network, and storage resources to reduce costs and increase agility. Cisco’s innovations are designed to unify data processes, simplify data center complexity, and amplify business results.
One of the central tenets for Data Center 3.0 is the migration from GbE to 10GbE. Whether its to support unified fabric (FCoE or iSCSI) or to support the type of I/O consumption server virtualization is driving, we feel 10GbE is a fundamental building block of the next generation of data centers.
To that end, we have spent a good deal of effort providing our customers a granular and cost effective path from their current GbE infrastructure to 10GbE. We support fibre based connections across our switching portfolio with a wide variety of optics. With the advent of the Nexus 5000, we also added twin-ax to the mix with significantly lowered costs. A little while later, we introduced the Nexus 2000 fabric extenders at yet another option for customer migrating their data centers to 10GbE that both lowered costs and simplified management. We also recently added the Cisco Nexus 4000 blade switch for the IBM BladeCenter to the mix. The latest option we are offering is 10GBase-T (IEEE 802.3an-2006) support which allows customers to take advantage of their existing copper cabling as they navigate the transition to 10GbE. In keeping with the extend-your-investment theme, 10GBase-T will initially be available for the Cisco Catalyst family first, then the Cisco Nexus family.
With the addition of the 10GBase-T options, we continue to offer the broadest, most flexible portfolio of 10GbE options. Here are some more details:
One of the ongoing challenges for our customers is finding ways to easily interconnect their data centers. Traditional drivers for this have been business continuance and the desire to load balance and make better use of underutilized resources. While these continue to be important, because of the spread of server virtualization, we also see emerging drivers around supporting inter-data center workload mobility and cloud import/export of workloads.
At this point, you may be thinking “Omar, there are already ways to do this, some even offered by Cisco–some you have have blogged about!” Yes, dear reader, its true, but with the release of a new NX-OS feature called Overlay Transport Virtualization (OTV) on our Nexus 7000, we expect to make connection your data centers simpler while at the same time, making those connections more intelligent and better suited to the emerging demands on a data center interconnect solutions.
The cool, unique thing about our OTV solution is that it works with your existing transport. Essentially, OTV provides Ethernet LAN extension and “MAC routing” on top of the existing layer 3 (i.e. IP) infrastructure. Here is a quick (~ 3 min) video that goes over the basics of the OTV solution.