before deleting them. This improves performance by not requiring it to
  check for stale rules on every sync. (In smaller clusters, it will still
  remove unused rules immediately once they are no longer used.)

  (The threshold for "large" used here is currently "1000 endpoints" but
  this is subject to change.) ([#110334](https://github.com/kubernetes/kubernetes/pull/110334), [@danwinship](https://github.com/danwinship))

* Deprecate outofdisk-transition-frequency and low-diskspace-threshold-mb flags ([#41941](https://github.com/kubernetes/kubernetes/pull/41941), [@dashpole](https://github.com/dashpole))

        // Test very large values close to EB boundary
        assertEquals("1.024EB", MemoryUtil.byteCountToDisplaySize(FileUtils.ONE_EB));

        // Test exact thresholds
        assertEquals("1.024KB", MemoryUtil.byteCountToDisplaySize(1024L));
        assertEquals("1.024MB", MemoryUtil.byteCountToDisplaySize(1024L * 1024L));

,"asHours","asDays","asWeeks","asMonths","asQuarters","asYears","makeGetter","thresholds","relativeTime$1","posNegDuration","abs$1","toISOString$1","totalSign","ymSign","daysSign","hmsSign","total","toFixed","proto$2","monthsFromDays","argWithSuffix","argThresholds","withSuffix","th","assign","toIsoString","version","updateLocale","tmpLocale","relativeTimeRounding","roundingFunction","relativeTimeThreshold","threshold","limit","myMoment","HTML5_FMT","DATETIME_LOCAL","DATETIME_LOCAL_SECONDS","DAT...

Here I'm also highlighting contributions from sponsors.

switch to square-and-double once the exponent is large // enough to justify the cost of the multiplications. // The threshold is selected experimentally as a linear function of n. threshold := n / 4 // We calculate how many of the most-significant bits of the exponent we can // compute before crossing the threshold, and we do it with doublings. i := bits.UintSize for logR>>i <= threshold { i-- } for k := uint(0); k < logR>>i; k++ { rr.Add(rr, m) } // Then we process the remaining bits of the exponent...

     * Custom Metrics for Autoscaling (HorizontalPodAutoscaler API in the Autoscaling
API group): The Horizontal Pod Autoscaling feature now supports custom metrics
(Alpha), allowing you to specify application-level metrics and thresholds to
trigger scaling up and down the number of pods in your application.
  * <strong>New GUI</strong> (dashboard) allows you to get started quickly and enables the same

* Fix detection of docker cgroup on RHEL ([#25907](https://github.com/kubernetes/kubernetes/pull/25907), [@ncdc](https://github.com/ncdc))
* Kubelet evicts pods when available memory falls below configured eviction thresholds ([#25772](https://github.com/kubernetes/kubernetes/pull/25772), [@derekwaynecarr](https://github.com/derekwaynecarr))

* Eviction thresholds set to 0% or 100% will turn off eviction. ([#59681](https://github.com/kubernetes/kubernetes/pull/59681), [@mtaufen](https://github.com/mtaufen))

- Users will now be able to configure all supported values for AWS NLB health check interval and thresholds for new resources. ([#96312](https://github.com/kubernetes/kubernetes/pull/96312), [@kishorj](https://github.com/kishorj)) [SIG Cloud Provider]