org.gridgain.grid.spi.loadbalancing.adaptive

Class GridAdaptiveLoadBalancingSpi

java.lang.Object

org.gridgain.grid.spi.GridSpiAdapter

org.gridgain.grid.spi.loadbalancing.adaptive.GridAdaptiveLoadBalancingSpi

All Implemented Interfaces:

GridSpi, GridSpiManagementMBean, GridAdaptiveLoadBalancingSpiMBean, GridLoadBalancingSpi

@GridSpiMultipleInstancesSupport(value=true)
public class GridAdaptiveLoadBalancingSpi
extends GridSpiAdapter
implements GridLoadBalancingSpi, GridAdaptiveLoadBalancingSpiMBean

Load balancing SPI that adapts to overall node performance. It proportionally distributes more jobs to more performant nodes based on a pluggable and dynamic node load probing.

Adaptive Node Probe

This SPI comes with pluggable algorithm to calculate a node load at any given point of time. The algorithm is defined by GridAdaptiveLoadProbe interface and user is free to provide custom implementations. By default GridAdaptiveCpuLoadProbe implementation is used which distributes jobs to nodes based on average CPU load on every node.

The following load probes are available with the product:

• GridAdaptiveCpuLoadProbe - default
• GridAdaptiveProcessingTimeLoadProbe
• GridAdaptiveJobCountLoadProbe

Note that if GridAdaptiveLoadProbe.getLoad(GridNode, int) returns a value of 0, then implementation will assume that load value is simply not available and will try to calculate an average of load values for other nodes. If such average cannot be obtained (all node load values are 0), then a value of 1 will be used.

When working with node metrics, take into account that all averages are calculated over metrics history size defined by GridConfiguration.getMetricsExpireTime() and GridConfiguration.getMetricsHistorySize() grid configuration parameters. Generally the larger these configuration parameter values are, the more precise the metrics are. You should tune these values based on the level of accuracy needed vs. the additional memory that would be required for storing metrics.

You should also keep in mind that metrics for remote nodes are delayed (usually by the heartbeat frequency). So if it is acceptable in your environment, set the heartbeat frequency to be more inline with job execution time. Generally, the more often heartbeats between nodes are exchanged, the more precise the metrics are. However, you should keep in mind that if heartbeats are exchanged too often then it may create unnecessary traffic in the network. Heartbeats (or metrics update frequency) can be configured via underlying GridDiscoverySpi used in your grid.

Here is an example of how probing can be implemented to use number of active and waiting jobs as probing mechanism:

 public class FooBarLoadProbe implements GridAdaptiveLoadProbe {
     // Flag indicating whether to use average value or current.
     private int useAvg = true;

     public FooBarLoadProbe(boolean useAvg) {
         this.useAvg = useAvg;
     }

     // Calculate load based on number of active and waiting jobs.
     public double getLoad(GridNode node, int jobsSentSinceLastUpdate) {
         GridNodeMetrics metrics = node.getMetrics();

         if (useAvg) {
             double load = metrics.getAverageActiveJobs() + metrics.getAverageWaitingJobs();

             if (load > 0) {
                 return load;
             }
         }

         return metrics.getCurrentActiveJobs() + metrics.getCurrentWaitingJobs();
     }
 }
 

Which Node Probe To Use

There is no correct answer here. Every single node probe will work better or worse in different environments. CPU load probe (default option) is the safest approach to start with as it simply attempts to utilize every CPU on the grid to the maximum. However, you should experiment with other probes by executing load tests in your environment and observing which probe gives you best performance and load balancing.

Task Coding Example

If you are using GridComputeTaskSplitAdapter then load balancing logic is transparent to your code and is handled automatically by the adapter. Here is an example of how your task will look:

 public class MyFooBarTask extends GridComputeTaskSplitAdapter<Object, Object> {
    @Override
    protected Collection<? extends GridComputeJob> split(int gridSize, Object arg) throws GridException {
        List<MyFooBarJob> jobs = new ArrayList<MyFooBarJob>(gridSize);

        for (int i = 0; i < gridSize; i++) {
            jobs.add(new MyFooBarJob(arg));
        }

        // Node assignment via load balancer
        // happens automatically.
        return jobs;
    }
    ...
 }
 

If you need more fine-grained control over how some jobs within task get mapped to a node and use affinity load balancing for some other jobs within task, then you should use GridComputeTaskAdapter. Here is an example of how your task will look. Note that in this case we manually inject load balancer and use it to pick the best node. Doing it in such way would allow user to map some jobs manually and for others use load balancer.

 public class MyFooBarTask extends GridComputeTaskAdapter<String, String> {
    // Inject load balancer.
    @GridLoadBalancerResource
    GridComputeLoadBalancer balancer;

    // Map jobs to grid nodes.
    public Map<? extends GridComputeJob, GridNode> map(List<GridNode> subgrid, String arg) throws GridException {
        Map<MyFooBarJob, GridNode> jobs = new HashMap<MyFooBarJob, GridNode>(subgrid.size());

        // In more complex cases, you can actually do
        // more complicated assignments of jobs to nodes.
        for (int i = 0; i < subgrid.size(); i++) {
            // Pick the next best balanced node for the job.
            jobs.put(new MyFooBarJob(arg), balancer.getBalancedNode())
        }

        return jobs;
    }

    // Aggregate results into one compound result.
    public String reduce(List<GridComputeJobResult> results) throws GridException {
        // For the purpose of this example we simply
        // concatenate string representation of every
        // job result
        StringBuilder buf = new StringBuilder();

        for (GridComputeJobResult res : results) {
            // Append string representation of result
            // returned by every job.
            buf.append(res.getData().string());
        }

        return buf.string();
    }
 }
 

Configuration

In order to use this load balancer, you should configure your grid instance to use GridJobsLoadBalancingSpi either from Spring XML file or directly. The following configuration parameters are supported:

Mandatory

This SPI has no mandatory configuration parameters.

Optional

This SPI has the following optional configuration parameters:

• Adaptive node load probing implementation (see setLoadProbe(GridAdaptiveLoadProbe)). This configuration parameter supplies a custom algorithm for probing a node's load. By default, GridAdaptiveCpuLoadProbe implementation is used which takes every node's CPU load and tries to send proportionally more jobs to less loaded nodes.

Below is Java configuration example:

 GridAdaptiveLoadBalancingSpi spi = new GridAdaptiveLoadBalancingSpi();

 // Configure probe to use latest job execution time vs. average.
 GridAdaptiveProcessingTimeLoadProbe probe = new GridAdaptiveProcessingTimeLoadProbe(false);

 spi.setLoadProbe(probe);

 GridConfiguration cfg = new GridConfiguration();

 // Override default load balancing SPI.
 cfg.setLoadBalancingSpi(spi);

 // Starts grid.
 G.start(cfg);
 

Here is how you can configure GridJobsLoadBalancingSpi using Spring XML configuration:

 <property name="loadBalancingSpi">
     <bean class="org.gridgain.grid.spi.loadBalancing.adaptive.GridAdaptiveLoadBalancingSpi">
         <property name="loadProbe">
             <bean class="org.gridgain.grid.spi.loadBalancing.adaptive.GridAdaptiveProcessingTimeLoadProbe">
                 <constructor-arg value="false"/>
             </bean>
         </property>
     </bean>
 </property>
 

For information about Spring framework visit www.springframework.org

Constructor Summary

Constructors

Constructor and Description
GridAdaptiveLoadBalancingSpi()

Method Summary

Methods

Modifier and Type	Method and Description
GridNode	getBalancedNode(GridComputeTaskSession ses, List<GridNode> top, GridComputeJob job) Gets balanced node for specified job within given task session.
String	getLoadProbeFormatted() Gets text description of current load probing implementation used.
protected void	onContextDestroyed0() Method to be called in the beginning of onContextDestroyed() method.
protected void	onContextInitialized0(GridSpiContext spiCtx) Method to be called in the end of onContextInitialized method.
void	setLoadProbe(GridAdaptiveLoadProbe probe) Sets implementation of node load probe.
void	spiStart(String gridName) This method is called to start SPI.
void	spiStop() This method is called to stop SPI.
String	toString()

Methods inherited from class org.gridgain.grid.spi.GridSpiAdapter

assertParameter, checkConfigurationConsistency0, configInfo, createSpiAttributeName, getConsistentAttributeNames, getGridGainHome, getLocalNodeId, getName, getNodeAttributes, getSpiContext, getStartTimestamp, getStartTimestampFormatted, getUpTime, getUpTimeFormatted, injectables, onContextDestroyed, onContextInitialized, registerMBean, setName, startInfo, startStopwatch, stopInfo, unregisterMBean

Methods inherited from class java.lang.Object

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

Methods inherited from interface org.gridgain.grid.spi.GridSpi

getName, getNodeAttributes, onContextDestroyed, onContextInitialized

Methods inherited from interface org.gridgain.grid.spi.GridSpiManagementMBean

getGridGainHome, getLocalNodeId, getName, getStartTimestamp, getStartTimestampFormatted, getUpTime, getUpTimeFormatted

Constructor Detail

GridAdaptiveLoadBalancingSpi

public GridAdaptiveLoadBalancingSpi()

Method Detail

getLoadProbeFormatted

public String getLoadProbeFormatted()

Gets text description of current load probing implementation used.

Specified by:

getLoadProbeFormatted in interface GridAdaptiveLoadBalancingSpiMBean

Returns:

Text description of current load probing implementation used.

setLoadProbe

@GridSpiConfiguration(optional=true)
public void setLoadProbe(GridAdaptiveLoadProbe probe)

Sets implementation of node load probe. By default GridAdaptiveProcessingTimeLoadProbe is used which proportionally distributes load based on the average job execution time on every node.

Parameters:

probe - Implementation of node load probe

spiStart

public void spiStart(@Nullable
            String gridName)
              throws GridSpiException

This method is called to start SPI. After this method returns successfully kernel assumes that SPI is fully operational.

Specified by:

spiStart in interface GridSpi

Parameters:

gridName - Name of grid instance this SPI is being started for (null for default grid).

Throws:

GridSpiException - Throws in case of any error during SPI start.

spiStop

public void spiStop()
             throws GridSpiException

This method is called to stop SPI. After this method returns kernel assumes that this SPI is finished and all resources acquired by it are released.

Note that this method can be called at any point including during recovery of failed start. It should make no assumptions on what state SPI will be in when this method is called.

Specified by:

spiStop in interface GridSpi

Throws:

GridSpiException - Thrown in case of any error during SPI stop.

onContextInitialized0

protected void onContextInitialized0(GridSpiContext spiCtx)
                              throws GridSpiException

Method to be called in the end of onContextInitialized method.

Overrides:

onContextInitialized0 in class GridSpiAdapter

Parameters:

spiCtx - SPI context.

Throws:

GridSpiException - In case of errors.

onContextDestroyed0

protected void onContextDestroyed0()

Method to be called in the beginning of onContextDestroyed() method.

Overrides:

onContextDestroyed0 in class GridSpiAdapter

getBalancedNode

public GridNode getBalancedNode(GridComputeTaskSession ses,
                       List<GridNode> top,
                       GridComputeJob job)
                         throws GridException

Gets balanced node for specified job within given task session.

Specified by:

getBalancedNode in interface GridLoadBalancingSpi

Parameters:

ses - Grid task session for currently executing task.

top - Topology of task nodes from which to pick the best balanced node for given job.

job - Job for which to pick the best balanced node.

Returns:

Best balanced node for the given job within given task session.

Throws:

GridException - If failed to get next balanced node.

toString

public String toString()

Overrides:

toString in class Object