Help: rrds v.1


The Research Ready Datasets (RRDS) service is a tool for timeseries data selection based on quality control criteria selected by the user.

Design and use philosophy

This interface is designed to support client-side access to data that meets with specific quality controls as defined by metrics associated with that data. In this model, user software submits a request containing limits on various metrics of interest. The data that pass this set of search filters are returned in a form suitable for submission to other web services.

The definition of quality is thus up to the user. By combining different metrics, and adjusting the search filters on those metrics, the user can design a flexible and reusable interface for obtaining research-ready data sets.

Output formats

All parameters listed under the “Service-options”, “Channel constraints”, and “Time constraints” are applied as search filters into the metadata records. The “Metric constraints” offer access to the qualitative properties (metrics) associated with those records.

The default request format (…&format=request)

The request format shows the channel data that passed all of the user’s metric constraints. It is a simplified “request format” ready to be submitted to another web service.

The report format (…&format=report)

The report output provides additional information below the request section. Channel data that failed a particular metric constraint are listed in a simple line-by-line format suitable for subsequent script input or data table.

The verbosereport format (…&format=verbosereport)

The verbosereport output, like the “report” output, shows the metric failures but is organized in a more human-readable display.

Metrics: Using comparators

RRDS data are queried using a string combination of a metric plus a comparator suffix. There are six comparators:

  • _eq – Equal (the default if omitted)
  • _ne – Not equal
  • _gt – Greater than
  • _ge – Greater than or equal
  • _lt – Less than
  • _le – Less than or equal
You can specify a value range for a particular metric, for example, say we want all the data with a sample_rms metric between 1,000 and 5,000 (inclusive). We would achieve that with: sample_rms_ge=1000&sample_rms_le=5000

Here are some other examples of the metric & comparator combination:

example meaning notes
percent_availability_eq=100 The percent_availability metric is equal to 100
percent_availability=100 The percent_availability metric is equal to 100 Omitted a comparator, therefore the parameter defaults to ‘percent_availability_eq’
percent_availability_ne=0 The percent_availability metric is not equal to 0
num_gaps_le=1 The num_gaps metric is less than or equal to 1
timing_quality_gt=95 The timing_quality metric is greater than 95
transfer_function-gain_ratio_gt=1 The gain_ratio submetric of the transfer_function metric is greater than 1 Use a hyphen (-) between a metric and its submetric
noise_mode_diff_lt=hnm-15,0.001,100.0 Deviations from the HNM are lower than -15 dB from the HNM in the frequency range 0.001 to 100.0 Hz The range values default to Hz (frequency range)
noise_mode_diff_ge=lnm+15,1000s,0.01s Deviations from the LNM exceed or are at +15 dB from the LNM in the period range 1000 to 0.01 seconds Use the ‘s’ character to indicate seconds (period range)