What is Geo-Targeting

A sudden increase in traffic from China? No problem, the load balancer creates 10 new instances in the Asian Cloud and point all Chinese traffic to those instances.

Geotargeting, or geographical targeting, is a way for servers and applications to automatically adjust for a geographical location. Load Balancing tied directly to geo-targeting allows you to target system resource more specifically to the region requiring the most demand. Each cloud could handle various "storm fronts" and allows customers to intellegently & elastically adapt their environment based on real world conditions. This approach allows for an adaptive scaling for only the regions that require it. By providing operating system level geotargeting, administrators can more intellegently create an elastic computing server environment.

When using edge computing, fragments of the intended applications are replicated across a vast distributed network of web servers. This type of technology is not completely new. Many large organizations have had to deploy Web server farms with clustering to ensure acceptable performance.

Edge computing now makes this technology available to all businesses small, medium and large. The technology does however impose certain limitations to the choice of technology platform and even to the implementations in that the applications need to be specifically developed for edge computing.

Edge computing is based on a charged for network services model. The end user is any Internet user making use of a commercial Internet service. The customer for edge servers is the organizations wanting to linearly scale their business application performance to the growth of is subscriber base.

Edge computing and Grid computing are related. Whereas Grid computing would be hardcoded into a specific application to distribute its complex and resource intensive computational needs across a global grid of cheap networked machines, Edge computing provides a generic template facility for any type of application to spread its execution across a dedicated grid of prepared expensive machines.

We think the minimum specs will be as follows

• at least 5 dedicated servers
• with at least 2 gigs of ram,
• 100gb or more storage (benchmark maybe required)
• decent internet connection (depends on the region), benchmark may be required)
• Xen virtualization software (Possibly VMware & Virtuozzo later).

Pricing coming soon. We're thinking about a variable pricing matrix depending on the cluster size. A credit system will likely be setup where users can exchange capacity in a trading environment.This allows GeoElastic Alliance members to barter excess capacity for later usage.