*
   * <p>If {@code expectedSize} is exactly the number of entries added to the builder before {@link
   * Builder#build} is called, the builder is likely to perform better than an unsized {@link
   * #builder()} would have.
   *
   * <p>It is not specified if any performance benefits apply if {@code expectedSize} is close to,
   * but not exactly, the number of entries added to the builder.

  ASSERT_TRUE(GetFunctionDef(func_, &fdef));
  VerifyFDefInputs(fdef, M({{"feed1"}, {"feed2"}}));
  VerifyFDefOutputs(fdef, M({{"test_loop_exit"}}));
  VerifyFDefEdges(fdef,
                  {{"feed1", "test_loop/Enter:0"},
                   {"test_loop/Enter:output:0", "test_loop/Merge:0"},
                   {"test_loop/Merge:output:0", "test_loop/Switch:0"},
                   {"test_loop/Switch:output_false:0", "test_loop/Exit:0"},

   *
   * <p>If {@code expectedSize} is exactly the number of entries added to the builder before {@link
   * Builder#build} is called, the builder is likely to perform better than an unsized {@link
   * #builder()} would have.
   *
   * <p>It is not specified if any performance benefits apply if {@code expectedSize} is close to,
   * but not exactly, the number of entries added to the builder.

# errors.
# There is ongoing work on Bazel team's side to provide support for transitive
# shared libraries. As part of migrating to transitive shared libraries, we
# hope to provide a better mechanism for control over symbol exporting, and
# then tackle this issue again.
#
# TODO: Remove the following two lines once TF doesn't depend on Bazel wrapping
# all library archives in -whole_archive -no_whole_archive.

- EndpointSlice controller now handles terminating pods correctly and is better at preventing race conditions. ([#89118](https://github.com/kubernetes/kubernetes/pull/89118), [@robscott](https://github.com/robscott)) [SIG Apps and Network]

 * two separate characters. {@link #countIn} counts each supplementary character as 2 {@code char}s.
 *
 * <p>For up-to-date Unicode character properties (digit, letter, etc.) and support for
 * supplementary code points, use ICU4J UCharacter and UnicodeSet (freeze() after building). For
 * basic text processing based on UnicodeSet use the ICU4J UnicodeSetSpanner.
 *
 * <p>Example usages:
 *

		}
		if s[i+1] == c1 && Equal(s[i:i+n], sep) {
			return i
		}
		i++
		fails++
		if fails >= 4+i>>4 && i < t {
			// Give up on IndexByte, it isn't skipping ahead
			// far enough to be better than Rabin-Karp.
			// Experiments (using IndexPeriodic) suggest
			// the cutover is about 16 byte skips.
			// TODO: if large prefixes of sep are matching
			// we should cutover at even larger average skips,

さらに、並列性と非同期性を同時に活用できるため、テストされた多くの NodeJS フレームワークより高い性能を発揮し、C に近いコンパイル言語である Go と同等の性能になります [(すべて Starlette のおかげです)](https://www.techempower.com/benchmarks/#section=data-r17&hw=ph&test=query&l=zijmkf-1)。

### 並行処理は並列処理より優れている？ { #is-concurrency-better-than-parallelism }

いいえ！それがこの話の教訓ではありません。

並行処理は並列処理とは異なります。そして多くの待ち時間を伴う**特定の**シナリオでは優れています。そのため、一般に Web アプリ開発では並列処理よりはるかに適しています。しかし、すべてに対して最良というわけではありません。

バランスを取るために、次の短い物語を想像してください。

                slf4jLogger.debug(buff.toString());
            }
        }
    }

    // Needed to make this method package visible to make writing a unit test possible
    // Maybe it's better to move some of those methods to separate class (SoC).
    void properties(CliRequest cliRequest) throws Exception {
        Properties paths = new Properties();
        if (cliRequest.topDirectory != null) {

  /*
   * If n <= millerRabinBases[i][0], then testing n against bases millerRabinBases[i][1..] suffices
   * to prove its primality. Values from miller-rabin.appspot.com.
   *
   * NOTE: We could get slightly better bases that would be treated as unsigned, but benchmarks
   * showed negligible performance improvements.
   */
  private static final long[][] millerRabinBaseSets = {
    {291830, 126401071349994536L},