Blog

Ixia Can Help Assess Benefits Offered by SDN/NFV-Based Network Platforms

January 26, 2015 by Ixia Blog Team

Today’s networks are going through a major transformation with the introduction of SDN/NFV. This new transformation promises to offer benefits for network operators, such as service agility, enhanced visibility, greater flexibility, and technology/product independence. While the benefits are encouraging, it is important to measure the impact of this network transformation to the end user—and measure those benefits objectively.

For example, when application performance degrades due to increased user activities, the controller/orchestrator MUST observe this change and start adding more compute/network resources dynamically without user’s intervention. It is very important to test network infrastructure in terms of deployment delay, performance, jitter, and packet loss, which can play key roles in user applications. This benchmarking allows network operators and businesses to validate the true SLA for the user applications. The applicability of such performance may be seen in multi-tenant data centers.

Enterprise businesses and cloud service providers must be able to assess new platforms in a pre-production environment to ensure their business applications will run with optimal performance as well as also take an advantage of the benefits offered by SDN/NFV. Without proper assessment, deploying such environment in production network can be risky and might have adverse effect on business.

In order to validate network performance you need to simulate real-world traffic in a lab environment, where one can easily test large-scale virtualized infrastructure. To achieve this goal, one has to build real network consisting thousands of servers and millions of VMs. While it may seem like a viable choice but it may not be practical in terms of CAPEX and/or OPEX.

Faced with this issue, Ixia thought “why can’t we emulate racks and racks of servers with large numbers of Hypervisors and VMs running realistic application work-loads?” After lot of research and engineering efforts, we created the RackSim solution to address these needs. RackSim is a 2U appliance with 8 10GbE ports. It helps managing VM/Network resources, VM migration convergence time, as well as Network Overlay technology (VXLAN).

This solution can also help validating performance/scalability of SDN Controller/Orchestration function.

The following scenario describes how the Cisco APIC controller is integrated with VMware vCenter.

Cisco’s APIC architecture is designed to manage ACI fabric, policy enforcement, as well as network health monitoring for both physical and virtual environment. Not only does it has end-to-end visibility, but in the event of link/port failure, link saturation, policy change, or any other unexpected event, APIC can communicate with vCenter (via VMware published API) to take corrective actions on VMs.

RackSim can help validating this use case by supporting vCenter API on an Ixia-emulated VM manager.

cisco_apic

Source: Cisco.com

The following diagram shows RackSim emulating hundreds and thousands VMs and how APIC controller can talk to emulated VM manager via VMware API

RackSim

Check out for information at our RackSim webpage.

Additional Resources:

RackSim