params {
            text(
                "performance.baselines",
                type.defaultBaselines,
                display = ParameterDisplay.PROMPT,
                allowEmpty = true,
                description = "The baselines you want to run performance tests against. Empty means default baseline."
            )

[main]
summary=Maximum server performance for MinIO

[vm]
transparent_hugepage=madvise

[sysfs]
/sys/kernel/mm/transparent_hugepage/defrag=defer+madvise
/sys/kernel/mm/transparent_hugepage/khugepaged/max_ptes_none=0

[cpu]
force_latency=1
governor=performance
energy_perf_bias=performance
min_perf_pct=100

[sysctl]
fs.xfs.xfssyncd_centisecs=72000
net.core.busy_read=50
net.core.busy_poll=50
kernel.numa_balancing=1

build --distinct_host_configuration=false

# Store performance profiling log in the mounted artifact directory.
# The profile can be viewed by visiting chrome://tracing in a Chrome browser.
# See https://docs.bazel.build/versions/main/skylark/performance.html#performance-profiling
build --profile=/tf/pkg/profile.json.gz

# Use the rebuilt gcc toolchain to compile for manylinux2014

Especificamente, observa-se Uvicorn, Starlette e FastAPI comparados juntos (entre muitas outras ferramentas).

Quanto mais simples o problema resolvido pela ferramenta, melhor a performance que ela terá. E a maioria dos _benchmarks_ não testam as características adicionais fornecidas pela ferramenta.

A hierarquia segue assim:

* **Uvicorn**: um servidor ASGI

  }

  /**
   * Returns an immutable array containing the given values, in order.
   *
   * <p><b>Performance note:</b> this method delegates to {@link #copyOf(Collection)} if {@code
   * values} is a {@link Collection}. Otherwise it creates a {@link #builder} and uses {@link
   * Builder#addAll(Iterable)}, with all the performance implications associated with that.
   */
  public static ImmutableLongArray copyOf(Iterable<Long> values) {

  }

  /**
   * Returns an immutable array containing the given values, in order.
   *
   * <p><b>Performance note:</b> this method delegates to {@link #copyOf(Collection)} if {@code
   * values} is a {@link Collection}. Otherwise it creates a {@link #builder} and uses {@link
   * Builder#addAll(Iterable)}, with all the performance implications associated with that.
   */
  public static ImmutableIntArray copyOf(Iterable<Integer> values) {

```

Create a file `testobject` with some test data that will be transformed
```
cat > testobject << EOF
MinIO is a High Performance Object Storage released under GNU Affero General Public License v3.0. It is API compatible with Amazon S3 cloud storage service. Use MinIO to build high performance infrastructure for machine learning, analytics and application data workloads.
EOF
```

Upload this object to the bucket via `mc cp`

Specifically, to see Uvicorn, Starlette and FastAPI compared together (among many other tools).

The simpler the problem solved by the tool, the better performance it will get. And most of the benchmarks don't test the additional features provided by the tool.

The hierarchy is like:

* **Uvicorn**: an ASGI server
    * **Starlette**: (uses Uvicorn) a web microframework

build --copt=-Wno-gnu-offsetof-extensions

# Store performance profiling log in the mounted artifact directory.
# The profile can be viewed by visiting chrome://tracing in a Chrome browser.
# See https://docs.bazel.build/versions/main/skylark/performance.html#performance-profiling
build --profile=/tf/pkg/profile.json.gz

# Use the rebuilt gcc toolchain to compile for manylinux2014

  /*
   * TODO(cpovirk): if we start to override indexOf/lastIndexOf under GWT, we'll want some way to
   * override subList to return an ImmutableSortedAsList for better performance. Right now, I'm not
   * sure there's any performance hit from our failure to override subListUnchecked under GWT
   */
  @Override
  ImmutableList<E> subListUnchecked(int fromIndex, int toIndex) {