For most VLBI observations, the data are recorded at the stations and played back later at the correlator. Management of the recording process imposed significant constraints on the observing schedule in the days when the recordings were made on tape. The newer disk-based systems have reduced the complexity and severity of the constraints, but have not eliminated them entirely, especially where the constraints were really imposed by the operation of the correlator. Future systems, currently (Feb. 2008) under development will reduce the constraints yet further. The current concerns are discussed here. Many were also discussed in the Observing Strategy and Scan Times sections earlier in this manual.
The most important constraint imposed by the use of data recordings is on the total amount of data recorded in an observation. The total supply of recording media, combined with the cycle time through the system, sets the average sustained bit rate that can be recorded. Many projects are allowed to exceed that average rate, but the consequence is that the instrument has to sit idle, or do non-recording observations, for enough time to bring the average rate down to the sustainable level. When a project is allocated time, it is also allocated an average bit rate that it is allowed to use depending on it's needs and the resources available. Any schedule made for the project should not exceed the allowed bit rate.
The disk-based systems, such as the various flavors of Mark5, are not yet totally transparent to the user. Attention still needs to be paid to allowing for module changes, for readback tests and maximum reasonable recording scan lengths, and to not causing problems for synchronization of the recordings on playback. All of these topics are discussed in the Observing Strategy and Scan Times section and will not be discussed in detail here.