How do we do it, you ask? How do we use 100% solar power and maintain a 99.9% uptime? It’s simple. Well, not really. It’s redundant.
Our servers are part of a redundant solar powered network that has three separate Internet backbones. The first backbone goes through AT&T and is a wired link. The second backbone goes through Verizon and is a carrier grade wireless FCC Licensed link capable of scaling up to 400Mbps. The third backbone goes through Time Warner and is a wireless link as well. Each wireless dish uses a different mount peak to provide redundancy in case of any failures. All of these Internet backbones are running BGP (Border Gateway Protocol), which routes all traffic coming in and out of the network on the shortest possible path. BGP also allows complete redundancy, if one or two Internet backbone links goes down, the other will still handle the traffic. The three separate Internet connections then connect to two redundant Cisco 7200 VXR series routers, which use HSRP (Hot Spare Router Protocol) allowing one to take over the other during a failure. From there, each one goes to separate trunked switches, which allows one to go down and the other will take over. From there, two separate Cisco ASA 5500 series firewalls block all but needed ports and monitor each other, so if one goes down the other takes over. Out from the firewalls the traffic goes to another set of separate trunked switches, which can failover from one to the other if one fails. At this point the servers connect to these switches in the following fashion. Each server has two dual port NIC cards, one dual port NIC card in each PCI-133 slot. Slot 1′s NIC card, first ethernet port is teamed for failover with the first ethernet port on Slot 2′s NIC card. And Slot 1′s NIC card, second ethernet port is teamed for failover with the second ethernet port on Slot 2′s NIC card. Then Slot 1′s first ethernet port and Slot 2′s second ethernet port is connected to Switch A-M, and Slot 1′s second ethernet port and Slot 2′s first ethernet port is connected to Switch B-M. Each of these two port separate NIC card teams are then teamed together for complete redundancy. Read more…
