org.openspaces.grid.esm
Class NullElasticScaleHandler

java.lang.Object
  org.openspaces.grid.esm.NullElasticScaleHandler

All Implemented Interfaces:

ElasticScaleHandler

public class NullElasticScaleHandler
extends Object
implements ElasticScaleHandler

Constructor Summary

NullElasticScaleHandler()

Method Summary

boolean	accept(Machine machine) A machine can be accepted/not-accepted (filtered) for whatever reason at any point in time.
void	init(ElasticScaleHandlerConfig config) An initialization call, parameterized with the configuration used at deployment.
void	scaleIn(Machine machine) A scale in request to terminate a machine.
void	scaleOut(ElasticScaleHandlerContext context) A scale out request to start a machine.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

NullElasticScaleHandler

public NullElasticScaleHandler()

Method Detail

init

public void init(ElasticScaleHandlerConfig config)

Description copied from interface: ElasticScaleHandler

An initialization call, parameterized with the configuration used at deployment.

Specified by:

init in interface ElasticScaleHandler

Parameters:

config - The scale configuration.

accept

public boolean accept(Machine machine)

Description copied from interface: ElasticScaleHandler

A machine can be accepted/not-accepted (filtered) for whatever reason at any point in time. This will prevent from a GSC to be started on it.

Specified by:

accept in interface ElasticScaleHandler

Parameters:

machine - The machine to accept to filter.

Returns:

true to use this machine to start a GSC on; false to skip it.

scaleOut

public void scaleOut(ElasticScaleHandlerContext context)

Description copied from interface: ElasticScaleHandler

A scale out request to start a machine. Implementation may choose to block until it has scaled out (which will cause the ESM to wait until the call has returned, preventing it from handling other processing units). It's advised to scale out asynchronously due to long startup delays. As long as the machine has not been discovered, the ESM will repeatedly call this method until a new machine has been allocated. It's up to the implementation class to maintain the state and not allocate new resources before the pending allocations have completed. When a new machine is started, it should start a Grid Service Agent (GSA).

Specified by:

scaleOut in interface ElasticScaleHandler

Parameters:

context - Context details

scaleIn

public void scaleIn(Machine machine)

Description copied from interface: ElasticScaleHandler

A scale in request to terminate a machine. Implementation may choose to ignore it, or decide to terminate considering cost/benefit factors. Will be called only once per-last GSC on this machine to terminate. Implementation may choose to block until it has scaled in (which will cause the ESM to wait until the call has returned, preventing it from handling other processing units).

Specified by:

scaleIn in interface ElasticScaleHandler

Parameters:

machine - The machine to terminate.