The Neil Gehrels Swift Observatory

Swift Data Center (SDC) Implements Changes to Process Burst Data More Quickly

July 12, 2005

The SDC has recently put in place several processing enhancements that result in faster processing and availability of the first Swift data from GRBs. This note describes the changes you can expect to see in data on the SDC Quicklook site.

The pipeline now uses 99x segment numbers to quickly process XRT and UVOT data in parallel. Beginning July 11, 2005, the first data to arrive on the ground from an autonomous slew snapshot is being split into separate segment numbers as follows: 

TDRSS messages (990: to be completed)
XRT science (991)
UVOT science (992)

Segment number 000 will continue to be processed as previously, and will contain all hk and auxiliary data, as well as BAT, XRT and UVOT science data. At the same time, during the first 24 hours, as soon as new XRT science data is received, then segment 991 will rerun, with only XRT processing, and will complete in significantly less time that the entire pipeline (segment 000). Similarly, during the first 24 hours, when new UVOT data is received, segment 992 will rerun with only UVOT processing. BAT data will continue to come out quickly in segment 000; these modifications will not change that.

The 99x segments are temporary, and are meant only to be used in the very early analysis. They will not go to the permanent archives as they contain no unique data that is not in segment.

TDRSS messages will eventually go into segment 990. Currently they are not being handled properly but this is in the process of being corrected.

Example:

Here is how the three sequence numbers work for a new GRB. For target ID 145581, the first data would be available on the Quicklook site in the following sequences: 

00145581000 Full autonomous sequence (BAT, XRT and UVOT)

00145581991 just XRT data from autonomous sequence

00145581992 just UVOT data from autonomous sequence

00145581001 First observation after autonomous sequence (BAT, XRT and UVOT)

Each sequence also contains all the necessary auxiliary data and log files for complete analysis of the data contained within.

While this substantial change has already resulted in data being available more quickly to the community, we expect that it will take some time to find and fix any remaining bugs in the system. Please bear with us while we track these down. If you encounter problems please report them to us via the Swift Feedback form.