Large scale VoIP Platforms

Deploying a carrier grade VoIP service capable to handle a huge amount of users and simultaneous calls can quickly become a very costly operation or a disaster.
With over 10 years’ experience in VoIP field we can help you to build cost optimized big installations, fulfilling the most rigorous quality and availability requirements.

Our services includes:

planning: platform/network architecture most suitable for your needs. Disaster recovery.
customization for your needs including branding and covering all the possible routing/billing scenarios, number portability, integration with third party solutions if necessary (API/billing engine/radius/LDAP/Web/others)
deploying: cutting edge VoIP solutions covering all the needs of a VoIP business including our customized VoIP clients, load balancer, softswitch and database backend with continuous backups
high-availability configuration and deployment, including auto failover/failback, load sharing, load balancing and database HA
support: long term support and maintenance and debugging/assuring protocol compatibility with all peers

The Mizu VoIP platform is purely software based running on cheap legacy x86/x64 hardware (common Intel/AMD servers) providing the same quality as hardware based implementations at a fraction of costs.

There are multiple ways to implement a highly available fault tolerant VoIP system.
A common architecture is illustrated here:

Large VoIP

A few important parameters to know about our system:

Load balancer: capably to handle millions of simultaneous calls
A single SIP server instance (app server node) is capable to handle up to 8000 simultaneous calls on a common single CPU Xeon server (fully business to business or retail, including PBX features and media routing with sophisticated routing and billing)
Call routing delay on high load servers are below one second
Media routing time is 2 milliseconds in average with only less the 0.01% of packets above 20 milliseconds (same as with dedicated hardware)
Fulfills legal requirements such as lawful interception and emergency calls
You can comfortably manage all the system from a single MManage (admin client) graphical user interface covering all your application nodes
Scalability and redundancy for all elements of the network (in the smart way, so you will not need unutilized spare servers)
Failover at all levels: in-call, route, server, database
Maximum compatibility with all third-party VoIP implementation including carriers, gateways, softphones and other equipment’s