This can be very useful for setting up **resources** that you need to use for the whole app, and that are **shared** among requests, and/or that you need to **clean up** afterwards. For example, a database connection pool, or loading a shared machine learning model.

## Use Case

Let's start with an example **use case** and then see how to solve it with this.

Let's imagine that you have some **machine learning models** that you want to use to handle requests. 🤖

#### JVM platform

This is a platform that builds on the core and software platforms to add support for developing software that runs on the JVM.
This includes software that is implemented using Java, Kotlin or some other JVM language.

This platform provides specific support for Java, Groovy and Scala, and includes the foojay toolchain plugin. 

#### Extensibility platform

import java.lang.annotation.Target;

/**
 * Signifies that a test should not be run under Android. This annotation is respected only by our
 * Google-internal Android suite generators. Note that those generators also suppress any test
 * annotated with LargeTest.
 *
 * <p>Why use a custom annotation instead of {@code android.test.suitebuilder.annotation.Suppress}?
 * I'm not completely sure that this is the right choice, but it has various advantages:
 *

  - type: textarea
    id: current-behavior
    attributes:
      label: Current Behavior
      description: Tell us what happens
    validations:
      required: true
  - type: textarea
    id: expected-behavior
    attributes:
      label: Expected Behavior
      description: Tell us what should happen
    validations:
      required: true
  - type: textarea
    id: context
    attributes:

    // Verify the peer received what was expected.
    assertThat(peer.takeFrame().type).isEqualTo(Http2.TYPE_HEADERS)
    assertThat(peer.takeFrame().type).isEqualTo(Http2.TYPE_RST_STREAM)
  }

  /**
   * Confirm that the client times out if the server stalls after 3 bytes. After the timeout the
   * connection is still considered healthy while we await the degraded pong. When that doesn't

  - type: textarea
    id: context
    attributes:
      label: Context
      description: |
        How has this issue affected you? What are you trying to accomplish? What other alternatives have you considered?
        Providing context helps us come up with a solution that is most useful in the real world
    validations:

When fetching certificates, ztunnel will authenticate to the CA with its own identity, but request the identity of another workload.
Critically, the CA must enforce that the ztunnel has permission to request that identity.
Requests for identities not running on the node are rejected.
This is critical to ensure that a compromised node does not compromise the entire mesh.

available to TensorFlow is powerful enough that you should assume that the
TensorFlow process effectively executes arbitrary code.

The risk of loading untrusted checkpoints depends on the code or graph that you
are working with. When loading untrusted checkpoints, the values of the traced
variables from your model are also going to be untrusted. That means that if

   * of the function, thus time. Thus, the RateLimiter becomes /faster/ after a period of
   * underutilization. If, on the other hand, we pick a function that goes /above/ that horizontal
   * line, then it means that the area (time) is increased, thus storedPermits are more costly than
   * fresh permits, thus the RateLimiter becomes /slower/ after a period of underutilization.
   *

#### JVM platform

This is a platform that builds on the core and software platforms to add support for developing software that runs on the JVM.
This includes software that is implemented using Java, Kotlin or some other JVM language.

This platform provides specific support for Java, Groovy and Scala, and includes the foojay toolchain plugin. 

#### Extensibility platform