|
|
| |
|
| CASE STUDY |
|
| HIGH AVALIBILITY |
|
| SAN's-CLUSTERING-GEOGRAPHICAL & LOCAL LOAD
BALANCING |
| WIRELESS.COM |
|
| |
|
| The Challenge |
|
|
There is one thing in common with all .COMs and their
computing environments. This common item is UPTIME. In this type of scenario
99.997% up time is a cause for concern. Each minute that they are down,
they are losing a potential customer to a competitor that might never
come back to their site. The end-user in this case are all the cellular
phone users across the USA who are using their WAP browsers to access
this customer's website. So the problem for them is how to build an infrastructure
that can withstand hardware failures, hackers etc. and still be up all
of the time.
|
 |
| |
|
| The Solution |
|
|
This virtual farm is created with a pair of F5's BigIp
Load Balancers. To the customer, they browse the virtual IP address of
the farm and the Load Balancer distributes the load among the pool of
web servers. Each Server will have two redundant power supplies. Multiple
Network Interfaces are bound together in case of hardware failure. This
allows their customer traffic to be balanced among multiple facilities
and protects against the failure of any one facility.
|
 |
|
|
|