Connecting to the Amazon Web Services (AWS) cloud provides customers with scale and elasticity, but creating a Virtual Private Connection (VPC) has its limitations. Fan Gu, technical marketing engineer at Cisco Systems, Inc., works with a team of engineers who are collaborating with AWS to provide one-click provisioning that will enable them to tap into the capabilities of the public cloud through private connections.

Gu joined Jeff Frick (@JeffFrick) and Lisa Martin (@Luccazara), cohosts of theCUBE, from the SiliconANGLE Media team, visited the AWS Summit – Santa Clara event in California to talk about the collaborative project and about Cisco’s vision of helping cloud adoption. This week, theCUBE features Fan Gu for our Women in Tech column.

The Cisco/AWS collaboration

Cisco is a well-known industry resource in networking, and Frick asked Gu why the company was at the AWS event. She explained he role in the partnership between the two companies.

“We are an enterprise routing business unit, and our team [supports] a product called Cloud Services Router CSR1000V [on public cloud], which has been [a part of] AWS for quite a while. It is a virtualized router and has all the capabilities of routing VPN and a zone-based firewall section.

“Right now we have seen a lot of requests from our customers, not just from a Cisco perspective but from AWS, and they would like to have some solution that can overcome the VPC peering limitation, which they can’t do the transfer of the VPC our team works on from the back end. The engineering team from Cisco and AWS architecture engineering, [has] a true collaboration here, and we work on this transit VPC solution now that it is available.”

Fitting the pieces together

Frick followed up by asking Gu the specifics of where the product “sits” and how the router works. Gu offered an example of how the product works.

“The VPC solution is a virtual private cloud, and the customers can deploy it as a virtual network. They can deploy as many as they want; but the second day, what they want is to have a connection between them. If it is just a three-way VPC, it’s probably easy. For example, [they say], ‘We have VPC-A and B and C and want to talk between A and B. We do a VPC peering through the [AWS], and we want to talk between B and C — we can do the same thing through VPC peering. But tomorrow I want to connect between A and C, but I want to go in through B, which is a typical transit routing requirements, and AWS right now has this limitation. They can’t do it because it is restricted.’

From our side, the CSR, which is bigger routing … we have the capability to deploy the CSR in B and then we can transit from A to C through B. This is a big deal, especially when the customer has tons of VPC systems to connect. If you have hundreds and you want to do a full mashup off VPC peering, it is time-consuming, and from a management perspective, it is very, very tedious. The transfer VPC — first, it’s simple to manage, it’s one click; second, you save money and time to deploy this whole thing; and, third, it’s very secure. We have the IP Stack encryption for the tunnels.”

This is a big deal

Picking up on Gu’s comment, “This is a big deal,” Frick asked her to differentiate the capabilities of the CSR1000V from other VPC peering products. Gu highlighted why it is different.

“The first thing is connection, making connections between all of the PCs and the on-prem network, that’s one thing. And the other big deal is that it is all automated. Initially, when we have the VPC peering, you do it manually. You go into the router or go into AWS to make this connection; however with the transit VPC, it’s one click, so it’s very simple. All the hard work is done on the back end. Through the cloud formation, which boosts the VPC infrastructure … and also have the [AWS] Lambda function for orchestration, which actually looks for which spoke hasn’t been connected yet. We will do the connection and push the configuration into the CSR.

“Another thing, the CSR itself is our router; the functionality it offers is not limited to the routing and the VPN termination, but it also has other features, such as zone-based firewall, which provides security. In addition to that we have a VPC capability, which is application visibility and control, which monitors and analyzes the traffic. You can have that capability and feedback for your management portal, and then you can control what traffic you want to go through and manage from there. So it’s a combination, not just a connection.”

Helping to grow public cloud adoption

Ending the interview, Frick asked about how Cisco is helping to move the enterprise to the cloud. Gu revealed how talking to customers is helping to understand what is necessary to get there.

“We talk to customers from time to time, and they are asking, ‘How do I proactively monitor our virtual machines on a public [cloud] and our instances?’ We are looking at a cloud watch as well. Another thing is about performance, when people put virtual machines in the cloud as their extension of the network, they ask: ‘How can I have more throughput on the network?’ We are looking at different image types which have network accelerators. Also we are looking at auto scaling, which you can think of CSR as a processor, as you need more they will expand and when you don’t they will shut down and shrink.”

Watch the full interview below, and be sure to check out more of SiliconANGLE and theCUBE’s coverage of the AWS Summit – Santa Clara.

Photo by SiliconANGLE