类 ml.shifu.guagua.worker.WorkerContext
的使用

使用 WorkerContext 的软件包

ml.shifu.guagua.worker

This package contains guagua worker implementation.

ml.shifu.guagua.worker 中 WorkerContext 的使用

参数类型为 WorkerContext 的 ml.shifu.guagua.worker 中的方法

WORKER_RESULT	WorkerComputable.compute(WorkerContext<MASTER_RESULT,WORKER_RESULT> context) Computation for each iteration.
WORKER_RESULT	AbstractWorkerComputable.compute(WorkerContext<MASTER_RESULT,WORKER_RESULT> workerContext)
abstract WORKER_RESULT	AbstractWorkerComputable.doCompute(WorkerContext<MASTER_RESULT,WORKER_RESULT> workerContext) Real computation logic after data loading.
abstract void	AbstractWorkerComputable.init(WorkerContext<MASTER_RESULT,WORKER_RESULT> workerContext) Initialization work for the whole computation
protected WORKER_RESULT	GuaguaWorkerService.iterate(WorkerContext<MASTER_RESULT,WORKER_RESULT> context, int initialIteration, Progressable progress) Call each iteration computation and preIteration, postIteration in intercepters.
abstract void	AbstractWorkerComputable.load(KEY currentKey, VALUE currentValue, WorkerContext<MASTER_RESULT,WORKER_RESULT> workerContext) Load data one by one before computation.
void	WorkerTimer.postApplication(WorkerContext<MASTER_RESULT,WORKER_RESULT> context)
void	WorkerInterceptor.postApplication(WorkerContext<MASTER_RESULT,WORKER_RESULT> context) The hook point for each application or each mapreduce job which is after all iterations completed.
void	WorkerComputableTimer.postApplication(WorkerContext<MASTER_RESULT,WORKER_RESULT> context)
void	TraceWorkerInterceptor.postApplication(WorkerContext<MASTER_RESULT,WORKER_RESULT> context)
void	MemoryStatsWorkerInterceptor.postApplication(WorkerContext<MASTER_RESULT,WORKER_RESULT> context)
void	BasicWorkerInterceptor.postApplication(WorkerContext<MASTER_RESULT,WORKER_RESULT> context)
void	AbstractWorkerCoordinator.postApplication(WorkerContext<MASTER_RESULT,WORKER_RESULT> context)
void	WorkerTimer.postIteration(WorkerContext<MASTER_RESULT,WORKER_RESULT> context)
void	WorkerInterceptor.postIteration(WorkerContext<MASTER_RESULT,WORKER_RESULT> context) The hook point for each iteration which is after WorkerComputable.compute(WorkerContext).
void	WorkerComputableTimer.postIteration(WorkerContext<MASTER_RESULT,WORKER_RESULT> context)
void	TraceWorkerInterceptor.postIteration(WorkerContext<MASTER_RESULT,WORKER_RESULT> context)
void	SyncWorkerCoordinator.postIteration(WorkerContext<MASTER_RESULT,WORKER_RESULT> context)
void	MemoryStatsWorkerInterceptor.postIteration(WorkerContext<MASTER_RESULT,WORKER_RESULT> context)
void	InternalWorkerCoordinator.postIteration(WorkerContext<MASTER_RESULT,WORKER_RESULT> context)
void	BasicWorkerInterceptor.postIteration(WorkerContext<MASTER_RESULT,WORKER_RESULT> context)
void	AsyncWorkerCoordinator.postIteration(WorkerContext<MASTER_RESULT,WORKER_RESULT> context)
protected void	AbstractWorkerComputable.postLoad(WorkerContext<MASTER_RESULT,WORKER_RESULT> workerContext) Do some post work before loading data.
void	WorkerTimer.preApplication(WorkerContext<MASTER_RESULT,WORKER_RESULT> context)
void	WorkerInterceptor.preApplication(WorkerContext<MASTER_RESULT,WORKER_RESULT> context) The hook point for each application or each mapreduce job which is before all iterations started.
void	WorkerComputableTimer.preApplication(WorkerContext<MASTER_RESULT,WORKER_RESULT> context)
void	TraceWorkerInterceptor.preApplication(WorkerContext<MASTER_RESULT,WORKER_RESULT> context)
void	SyncWorkerCoordinator.preApplication(WorkerContext<MASTER_RESULT,WORKER_RESULT> context)
void	MemoryStatsWorkerInterceptor.preApplication(WorkerContext<MASTER_RESULT,WORKER_RESULT> context)
void	InternalWorkerCoordinator.preApplication(WorkerContext<MASTER_RESULT,WORKER_RESULT> context)
void	BasicWorkerInterceptor.preApplication(WorkerContext<MASTER_RESULT,WORKER_RESULT> context)
void	AsyncWorkerCoordinator.preApplication(WorkerContext<MASTER_RESULT,WORKER_RESULT> context)
void	WorkerTimer.preIteration(WorkerContext<MASTER_RESULT,WORKER_RESULT> context)
void	WorkerInterceptor.preIteration(WorkerContext<MASTER_RESULT,WORKER_RESULT> context) The hook point for each iteration which is before WorkerComputable.compute(WorkerContext).
void	WorkerComputableTimer.preIteration(WorkerContext<MASTER_RESULT,WORKER_RESULT> context)
void	TraceWorkerInterceptor.preIteration(WorkerContext<MASTER_RESULT,WORKER_RESULT> context)
void	MemoryStatsWorkerInterceptor.preIteration(WorkerContext<MASTER_RESULT,WORKER_RESULT> context)
void	InternalWorkerCoordinator.preIteration(WorkerContext<MASTER_RESULT,WORKER_RESULT> context)
void	BasicWorkerInterceptor.preIteration(WorkerContext<MASTER_RESULT,WORKER_RESULT> context)
void	AsyncWorkerCoordinator.preIteration(WorkerContext<MASTER_RESULT,WORKER_RESULT> context)
void	AbstractWorkerCoordinator.preIteration(WorkerContext<MASTER_RESULT,WORKER_RESULT> context)
protected void	AbstractWorkerComputable.preLoad(WorkerContext<MASTER_RESULT,WORKER_RESULT> workerContext) Do some pre work before loading data.
protected void	AbstractWorkerCoordinator.setMasterResult(WorkerContext<MASTER_RESULT,WORKER_RESULT> context, String appMasterNode, String appMasterSplitNode)

参数类型为 WorkerContext 的 ml.shifu.guagua.worker 中的构造方法

AbstractWorkerCoordinator.FailOverCoordinatorCommand(WorkerContext<MASTER_RESULT,WORKER_RESULT> context)