Measuring data freshness

freshness delay value by measuring the difference between log creation at the source (_TIME) and ingestion into Cortex XSIAM (_INSERT_TIME).

Metrics are collected and calculated per data source during five-minute aggregation periods and allocated into the following buckets. The recorded freshness delay value is the top value in the range of the bucket:

  • 0 to 30 seconds → 30 seconds
  • 30 to 60 seconds → 60 seconds
  • 60 seconds to 5 minutes → 300 seconds
  • 5 minutes to 1 hour → 3,600 seconds
  • 1 hour to 24 hours→ 86,400 seconds
  • 24 hours to week→ 604,800 seconds
Metric Description
data_freshness_max_delay <p>Maximum freshness delay value among all log entries in an aggregation period.</p><p>This reflects the worst case.</p>
data_freshness_median <p>Median freshness delay value among all log entries in an aggregation period.</p><p>50% of values are smaller than the median, and 50% of values are higher or equal to the median.</p>
data_freshness_ninetieth_percentile <p>Ninetieth percentile of delay values among all log entries in an aggregation period.</p><p>This delay value is 90% higher than other log entry differences. It reflects the worst case, but eliminates the spikes.</p>

The metrics are saved to the metrics_source dataset and are also available in the metrics_view preset.

  • The max_delay metric is taken from the maximum bucket value with a restricted limit; therefore, metrics show whole numbers.
  • The median and ninetieth_percentile metrics are statistical calculations that give an approximation of the real value; therefore, metrics show decimal numbers.
  • Time slots with a zero log count or zero byte count display records with zero values. Subsequently, the data freshness metrics will also have zero values.
  • Timezone differences between _TIME and _INSERT_TIME might cause time skews with negative differences. Negative differences are rounded to zero values.