The hosting and management of the booking and management system is undertaken internally.

The system has been architected to minimise the risk of service disruption from single point of failure problems.

Examples of this can be found in areas such as power where all critical services are backed up by both multiple UPS systems (Uninterruptible Power Supplies) and generator. The UPS backup systems are capable of running for a minimum of 90 minutes and the generator is capable of running all essential services and recharging the UPS systems. Taxibank also has multiple telephone systems that are able to take calls on the service numbers.

Taxibank has documented all aspects of its system and full emergency response procedures. All documentation is constantly updated and regular reviews and simulated exercises are undertaken to ensure that staff training in emergency procedures remains current.

Highlights of the Taxibank booking and management system capacity include:
* Support for up to 100 concurrent bookers
* 10,000 bookings per 24 hour period (average in excess of 400 bookings per hour)
* Mean time between failure (MTBF) six months for minor failures including disk replacement

Due to its tiered architecture and splitting of database functions from business logic functions the system can be scaled up by the addition of either more business logic servers for booking, administration functions etc or improved database response more database servers may be added into the cluster.

Given the system architecture it is possible to further increase system capacity by the splitting of core functionality within the business logic tier for function such as bookings, reporting and administration onto separate servers. This architecture makes the system scaleable to in excess of 10,000 transactions per hour for booking and administration purposes.