CIPHER CONNECT is a scalable, high performance, carrier-grade runtime environment designed to run on standard commercial-off-the-shelf computing hardware. Specifically, it connects existing networks with evolving IP networks, enabling evolution in a cost-effective manner. It uses a programmable protocol converter with an integrated orchestration engine.

Through COMPOSE (Eclipse - Software/Service Development Tool Kit) a user can import protocols and define protocol specific state machines and protocol conversion state machines. All state machines and orchestration scripts are then deployed through one common management and monitoring system. The use of a high speed, low latency framework (Common Framework) then provides necessary internal messaging and failover capabilities for highly available systems.

The CONNECT Common Framework simplifies development of redundant and clustered systems which require automatic fault recoveries with minimal maintenance and administration.

CONNECT is designed to support in excess of 50,000,000 Busy Hour Call Attempts and 15,000 simultaneous sessions per second (SPS).

The core components of the Common Framework are the following:

Messaging bus

The messaging software uses TCP/IP and UDP protocols to deliver messages across the cluster which can be a geographically distributed system.

Object Repository

A subsystem used for replication of session state data from active server to a backup server. The Object Repository supports active-active, active-backup and N+1 replication modes.

Facilities

Subsystems of the Common Framework, including: Timers, Database Access, Directory, Logging & Tracing, Statistics & Configuration, and OAM and Configuration Management Facilities.

Network Element Manager (NEM)

A set of processes which automatically manage a cluster from startup to shutdown. The major role of the NEM is to ensure minimal levels of maintenance and administration required to run the system on a day-to-day basis. The NEM provides a GUI for administration and operation tasks which require human intervention as well as for system monitoring purposes.

In-memory database

A data cache which keeps critical application data in-memory to minimize performance losses caused by database access.

The framework is optimized for handling a large number of parallel sessions. The sessions are considered to be state machines switched from one state to another by incoming network events or timer events.

Connect Functional Overview