assertEquals(0, Queues.drain(q, ImmutableList.of(), 1, 10, MILLISECONDS));

      Producer producer = new Producer(q, 1);
      // producing one, will ask for two
      Future<?> producerThread = threadPool.submit(producer);
      producer.beganProducing.await();

      // make sure we time out
      Stopwatch timer = Stopwatch.createStarted();

     */
    long deadline = System.nanoTime() + unit.toNanos(timeout);
    int added = 0;
    while (added < numElements) {
      // we could rely solely on #poll, but #drainTo might be more efficient when there are multiple
      // elements already available (e.g. LinkedBlockingQueue#drainTo locks only once)
      added += q.drainTo(buffer, numElements - added);

     */
    long deadline = System.nanoTime() + unit.toNanos(timeout);
    int added = 0;
    while (added < numElements) {
      // we could rely solely on #poll, but #drainTo might be more efficient when there are multiple
      // elements already available (e.g. LinkedBlockingQueue#drainTo locks only once)
      added += q.drainTo(buffer, numElements - added);

     * list's ordering. It is drained as a batch operation when either the DRAIN_THRESHOLD is
     * crossed or a write occurs on the segment.
     */
    final Queue<ReferenceEntry<K, V>> recencyQueue;

    /**
     * A counter of the number of reads since the last write, used to drain queues on a small
     * fraction of read operations.
     */

   *
   * This implementation uses a per-segment queue to record a memento of the additions, removals,
   * and accesses that were performed on the map. The queue is drained on writes and when it exceeds
   * its capacity threshold.
   *
   * The Least Recently Used page replacement algorithm was chosen due to its simplicity, high hit

- Supported fine-grained supplemental groups policy (KEP-3619), which enabled
  fine-grained control for supplementary groups in the first container processes.
  This allows you to choose whether to include groups defined in the container image (/etc/groups)

- Introduce a new server run option 'shutdown-send-retry-after'. If true the HTTP Server will continue listening until all non longrunning request(s) in flight have been drained, during this window all incoming requests will be rejected with a status code `429` and a 'Retry-After' response header. ([#101257](https://github.com/kubernetes/kubernetes/pull/101257), [@tkashem](https://github.com/tkashem))

For the 1.8 release, SIG Scalability focused on automating large cluster
scalability testing in a continuous integration (CI) environment. The SIG
defined a concrete process for scalability testing, created
documentation for the current scalability thresholds, and defined a new set of
Service Level Indicators (SLIs) and Service Level Objectives (SLOs) for the system.
Here's the release [scalability validation report].

- Improved scheduler performance when no scoring plugins were defined.
   ([#122058](https://github.com/kubernetes/kubernetes/pull/122058), [@aleksandra-malinowska](https://github.com/aleksandra-malinowska))
- Improved scheduler performance when no scoring plugins were defined.

  /**
   * Constructs a new {@code CacheStats} instance.
   *
   * <p>Five parameters of the same type in a row is a bad thing, but this class is not constructed
   * by end users and is too fine-grained for a builder.
   */
  @SuppressWarnings("GoodTime") // should accept a java.time.Duration
  public CacheStats(
      long hitCount,
      long missCount,
      long loadSuccessCount,
      long loadExceptionCount,