Fatpipe Warp??

pham0329

Does anyone have any experience with these? From what I understand, it's essentially a load balancer, is that right? Where does it sit on the network (between ISP and edge router, or between edge router and firewall, etc), and how does it work?

I'm going to be working with these in a couple weeks so I'm trying to find as much info as I can.

dtlokee

Typicall deployed between the firewall and the outside router because AFAIK they only support ethernet interfaces, if you're using an ethernet handoff then I would say you don't need the router in front of it. They basically perform load balancing and traffic shaping based on the link speeds (they can be different) and it will perform policy NAT and stateful connection tracking based on the connections so the same session remains on the same upstream pipe. Inbound connectivity is also possible with one to one nat and overloaded nat for publishing servers. In most cases the Fatpipe will end up being your nat device and the firewalls can also nat (but may be a private IP range between them and the fatpipe) or can just route.

cisco_trooper

dtlokee, have you had throughput issues with the WARP appliances? When I was evaluating these my throughput actually suffered by about 30% on both a 100M Metro-E and a 150M MIS.

pham0329

dtlokee wrote: »

Typicall deployed between the firewall and the outside router because AFAIK they only support ethernet interfaces, if you're using an ethernet handoff then I would say you don't need the router in front of it. They basically perform load balancing and traffic shaping based on the link speeds (they can be different) and it will perform policy NAT and stateful connection tracking based on the connections so the same session remains on the same upstream pipe. Inbound connectivity is also possible with one to one nat and overloaded nat for publishing servers. In most cases the Fatpipe will end up being your nat device and the firewalls can also nat (but may be a private IP range between them and the fatpipe) or can just route.

So if theres a WARP sitting between the router, and an ASA, the ASA would have a default route pointing to the WARP? The WARP would then handle all the NAT translation, and would have a default route pointing to the router?

Are these managed by a web interface or they CLI based?

cisco_trooper

The WARP will handle NAT translation between your public IP space - we'll call this subnet A - and a single subnet that you will deploy between the inside interface of the WARP and the outside interface of the ASA - we'll call this subnet B. Then you can perform whatever NAT you need to perform on the ASA for internal and DMZ resources to get translated to an IP address in subnet B. The ASA(s) default route will point to your WARP(s), and the WARP(s) default route will point to the upstream router(s). This is a web interface. I don't remember there being a CLI.