com.bigdata.mdi.IPartitionMetadata Maven / Gradle / Ivy
/**
Copyright (C) SYSTAP, LLC DBA Blazegraph 2006-2016. All rights reserved.
Contact:
SYSTAP, LLC DBA Blazegraph
2501 Calvert ST NW #106
Washington, DC 20008
[email protected]
This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; version 2 of the License.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program; if not, write to the Free Software
Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
*/
/*
* Created on Apr 23, 2007
*/
package com.bigdata.mdi;
import com.bigdata.btree.ICounter;
import com.bigdata.btree.IndexMetadata;
import com.bigdata.service.DataService;
import com.bigdata.service.IDataService;
import com.bigdata.service.IMetadataService;
/**
* A description of the metadata state for a partition of a scale-out index.
*
* Each index partition has a distinct partitionId. This partitionId is assigned
* by a centralized service - the {@link IMetadataService} for the scale-out
* index for that index partition. A centralized service is required in order to
* obtain distinct int32 partition identifiers because those partition
* identifiers are used in turn to support scale-out partition local
* {@link ICounter}s - the partitionId forms the upper word of the int64
* counter value.
*
* The scale-out index name and the left and right separator keys are a complete
* identifier for an index partition - any two index partitions which share
* those three properties MUST be the same index partition. However, access to
* index partitions is generally in terms of the name of the scale-out index and
* the partitionId. See {@link DataService#getIndexPartitionName(String, int)}
* which returns the name under which an index partition will be registered and
* the name that must be used when requesting operations on that index partition
* using an {@link IDataService}.
*
* An index partition has additional state, including:
*
* - the {@link IDataService} on which it resides.
* - the {@link IResourceMetadata}[] describing the resources required to
* materialize a view of the index partition.
* - those resources themselves, which contain the index partition data.
*
*
* The left and right separator keys define the half-open key range of the index
* partition. The separator keys are available directly as the keys of
* the {@link MetadataIndex}, therefore they are not stored in the index
* partition records within the metadata index. However, the separator keys are
* stored in the index partition description within the {@link IndexMetadata}
* records so that they are available locally with the index partition data.
*
* If the client knows a key or key range of interest for a scale-out index then
* they can obtain the relevant index partition descriptions and a data service
* locator either either by flooding the query to the {@link IDataService}s or
* from the {@link MetadataIndex}.
*
* @author Bryan Thompson
* @version $Id$
*/
public interface IPartitionMetadata extends ISeparatorKeys {
/**
* The unique partition identifier.
*/
public int getPartitionId();
/**
* Return {@link #getPartitionId()}
*/
public int hashCode();
}