* #offer} and a call to {@link #topK}, with O(k) memory. In comparison, quickselect has the same
 * asymptotics but requires O(n) memory, and a {@code PriorityQueue} implementation takes O(n log
 * k). In benchmarks, this implementation performs at least as well as either implementation, and
 * degrades more gracefully for worst-case input.
 *
 * <p>The implementation does not necessarily use a <i>stable</i> sorting algorithm; when multiple

独立した TechEmpower のベンチマークでは、Uvicorn で動作する**FastAPI**アプリケーションが、<a href="https://www.techempower.com/benchmarks/#section=test&runid=7464e520-0dc2-473d-bd34-dbdfd7e85911&hw=ph&test=query&l=zijzen-7" class="external-link" target="_blank">Python フレームワークの中で最も高速なものの 1 つ</a>であり、Starlette と Uvicorn（FastAPI で内部的に使用されています）にのみ下回っていると示されています。

詳細は<a href="https://fastapi.tiangolo.com/benchmarks/" class="internal-link" target="_blank">ベンチマーク</a>セクションをご覧ください。

## オプションの依存関係

## 效能

來自獨立機構 TechEmpower 的測試結果，顯示在 Uvicorn 執行下的 **FastAPI** 是 <a href="https://www.techempower.com/benchmarks/#section=test&runid=7464e520-0dc2-473d-bd34-dbdfd7e85911&hw=ph&test=query&l=zijzen-7" class="external-link" target="_blank">最快的 Python 框架之一</a>， 僅次於 Starlette 和 Uvicorn 本身 (兩者是 FastAPI 的底層)。 (\*)

想了解更多訊息，可以參考 <a href="https://fastapi.tiangolo.com/benchmarks/" class="internal-link" target="_blank">測試結果</a>。

## 可選的依賴套件

用於 Pydantic：

rodando sob Uvicorn como <a href="https://www.techempower.com/benchmarks/#section=test&runid=7464e520-0dc2-473d-bd34-dbdfd7e85911&hw=ph&test=query&l=zijzen-7" class="external-link" target="_blank">um dos _frameworks_ Python mais rápidos disponíveis</a>, somente atrás de Starlette e Uvicorn (utilizados internamente pelo FastAPI). (*)

Para entender mais sobre performance, veja a seção <a href="https://fastapi.tiangolo.com/pt/benchmarks/" class="internal-link" target="_blank">Comparações</a>....

   * than this method, and this method in turn generally performs better than copying the list and
   * calling {@link Collections#sort(List)}.
   */
  // TODO(kevinb): rerun benchmarks including new options
  public <E extends T> List<E> sortedCopy(Iterable<E> elements) {
    @SuppressWarnings("unchecked") // does not escape, and contains only E's
    E[] array = (E[]) Iterables.toArray(elements);

		}
		got := buf.Bytes()
		if !bytes.Equal(expected, got) {
			t.Fatalf("read data is different from what was expected, offset=%d length=%d", offset, readLen)
		}
		buf.Reset()
	}
}

// Benchmarks

func benchmarkErasureDecode(data, parity, dataDown, parityDown int, size int64, b *testing.B) {
	setup, err := newErasureTestSetup(b, data, parity, blockSizeV2)
	if err != nil {

	DeleteObjectTags(context.Context, string, string, ObjectOptions) (ObjectInfo, error)
}

// GetObject - TODO(aead): This function just acts as an adapter for GetObject tests and benchmarks
// since the GetObject method of the ObjectLayer interface has been removed. Once, the
// tests are adjusted to use GetObjectNInfo this function can be removed.

### Eşzamanlılık paralellikten daha mı iyi?

Hayır!  Hikayenin ahlakı bu değil.

### Конкурентность лучше параллелизма? { #is-concurrency-better-than-parallelism }

///

## Karşılaştırma ve Hız