Reliability and Scalability

Everything fails.  There is only hope in numbers.  In traditional client/server setups, this means load balancing and fail-over -- racks of redundant servers replacing mainframes.

Service delivery is vulnerable to the same network outages and hardware failures as client/server setups, but it turns the whole concept of load balancing on its head.  Instead of fielding an unpredictable series of requests from clients, there is simply a given number of tasks that must be accomplished by a set number of servers.

Because the requests are managed within the Service Delivery Network, dividing those tasks amongst the available servers can be done intelligently.  And because Lopez is a truly distributed application, a node failure is a minor event, not a major catastrophe.

Intelligent Delegation

You have a task to do, you know it will take a lot of processor power, or a lot of time, or a lot of network.  What servers are best enabled -- at this very moment -- to handle it?  What happens if a server goes down?

This is Lopez's gig.  Throw a task into a Lopez cluster, it comes back completed.  You don't have to worry about where it gets done, but you do know that it will be the best place.

Lopez is new technology, so it's still being rigorously tested in our own Service Delivery Network, but soon it will be unleashed on the world. Watch this space for details.