Tech Showcase: High-Fidelity Network Validation for Cloud Scale Networks

We propose a platform to validate networks with high-fidelity emulations. On a computing cluster such as Azure, we run the full stack (including real operating system, software, and chip emulators) of each network device in a virtualized sandbox (a virtual machine or a container). We interconnect the sandboxes of the network devices with virtual network links in the same way as in production and load real configurations onto them. The network operator can directly log into emulated devices and carry out tests, or we can extract the routing information and analyze it logically by using a logic solver such as Z3. The key advantage is high fidelity: it directly verifies the network status and behaviors with the same software and configurations as in production, so that it can produce much more trustworthy validation results to network developers. We have shown an early prototype to Azure network engineers, to enthusiastic response.

Speaker Bios
Haseeb Niaz
Yibo Zhu
Yibo Zhu is a PhD candidate at University of California, Santa Barbara. His research interests are in datacenter networks and wireless networks. He explores new network technologies, such as 60GHz wireless, Remote Direct Memory Access (RDMA) and programmable network switches. He designs multiple novel systems that are published in top conferences and are deployed in Microsoft Azure datacenters. He receives B.S. from Tsinghua University (2011), with distinction among all college students in Beijing. He is a recipient of Microsoft Research Fellowship (2015) and UCSB Holbrook Fellowship (2011).
Séries:
Microsoft Research Faculty Summit
Date:
Haut-parleurs:
Haseeb Niaz, Yibo Zhu
Affiliation:
University of California, Santa Barbara

Taille: Microsoft Research Faculty Summit