New Portability Options for Cisco’s Data Center Networking Software and Hardware
As our customers look for more flexibility and portability across their data center network, Cisco continues to deliver open APIs to our networking devices, integration with popular automation frameworks like Ansible, Puppet, or Chef, and programming via any set of convenient programming languages, including Python. While these APIs obviously play a critical role for our service provider and web-scale customers, openness also means having flexibility to choose the hardware and software combinations that best meet their needs.
Cisco now offers numerous portability options for our Nexus switches and our Nexus Operating System (NX-OS) – including the Cisco Cloud Scale Switch Abstraction Interface (SAI). SAI allows customers the freedom to run the network operating system of their choice on our SAI-ready Nexus platforms.
Microsoft and other web-scale customers are now running their Sonic operating system on these Nexus 9200/9300 platforms. As Yousef Khalidi, CVP of Microsoft Azure Networking noted:
“Cisco support for the SAI abstraction layer helps fulfill our vision for SAI to enable rapid innovation in silicon, CPU, power, port density, optics, and speed across multiple platforms while enabling Microsoft and cloud operators to leverage the same software stack across a variety of switch hardware platforms.”
Opening up our hardware platforms is only one dimension of portability that we’ve enabled. We’ve also made it possible to run NX-OS on third-party hardware platforms – independent from our Nexus switches. Now, customers can leverage our industry-leading software innovations and gain the flexibility to adapt them to hardware platforms that best suit their needs.
To help customers evaluate how they can best use this portability, we now offer a virtual NX-OS version. Customers can simulate anything from new features and the transitional fabric state during upgrades through the software upgrade impact on existing tooling environment for their actual large-scale topologies.