Real-time has the potential to be a large drain on system resources; particularly CPU load consumed by FLMService. As with many of the ADMS applications its usage and scope should be the result of careful consideration.

Obviously fast-cycle optimization will run the most frequently, potentially 4 times per hour using the default 15 minute cycle times. This analysis is performed on the real-time state of the network. This mode should probably be constrained to the areas of the network which are most sensitive to Volt/Var issues.

Slow-cycle optimization has the benefit of using the forecasted feeder loads provided by FLM to produce optimizations that achieve the best result over a number of hours. This means in practice that the best plan might not be used in the short term because it will either produce violations in the long term or will need to be backed off. The trade-off of course is that this mode is less stressful on the NMS system and more importantly, performs less field operations thus potentially extending field device life expectancy.