Package com.linkedin.davinci.client
Class AvroGenericDaVinciClient<K,V>
java.lang.Object
com.linkedin.davinci.client.AvroGenericDaVinciClient<K,V>
- All Implemented Interfaces:
DaVinciClient<K,
,V> AvroGenericReadComputeStoreClient<K,
,V> AvroGenericStoreClient<K,
,V> Closeable
,AutoCloseable
- Direct Known Subclasses:
AvroSpecificDaVinciClient
public class AvroGenericDaVinciClient<K,V>
extends Object
implements DaVinciClient<K,V>, AvroGenericReadComputeStoreClient<K,V>
-
Field Summary
FieldsModifier and TypeFieldDescriptionstatic final int
protected final org.apache.logging.log4j.Logger
static final ExecutorService
The following two fields are used to speed up the requests with a big number of keys: 1. -
Constructor Summary
ConstructorsModifierConstructorDescriptionAvroGenericDaVinciClient
(DaVinciConfig daVinciConfig, ClientConfig clientConfig, VeniceProperties backendConfig, Optional<Set<String>> managedClients) protected
AvroGenericDaVinciClient
(DaVinciConfig daVinciConfig, ClientConfig clientConfig, VeniceProperties backendConfig, Optional<Set<String>> managedClients, ICProvider icProvider, AbstractAvroChunkingAdapter<V> chunkingAdapter, Runnable preValidation, Executor readChunkExecutorForLargeRequest) AvroGenericDaVinciClient
(DaVinciConfig daVinciConfig, ClientConfig clientConfig, VeniceProperties backendConfig, Optional<Set<String>> managedClients, ICProvider icProvider, Executor readChunkExecutorForLargeRequest) -
Method Summary
Modifier and TypeMethodDescriptionGet the values associated with the given keys and return them in a map of keys to values.void
close()
Release the internal resources.void
compute
(ComputeRequestWrapper computeRequestWrapper, Set<K> keys, org.apache.avro.Schema resultSchema, StreamingCallback<K, ComputeGenericRecord> callback, long preRequestTimeInNS) compute
(Optional<ClientStats> stats, AvroGenericReadComputeStoreClient computeStoreClient) void
computeWithKeyPrefixFilter
(byte[] keyPrefix, ComputeRequestWrapper computeRequestWrapper, StreamingCallback<org.apache.avro.generic.GenericRecord, org.apache.avro.generic.GenericRecord> callback) protected D2ServiceDiscoveryResponse
Lookup the value by given key, and get(key).get() will return null if it doesn't exist.Similar toAvroGenericStoreClient.get(Object)
except that it allows passing in a instance, to minimize GC.protected AbstractAvroChunkingAdapter<V>
static DaVinciBackend
protected GenericRecordChunkingAdapter
org.apache.avro.Schema
Get key schema.org.apache.avro.Schema
Deprecated.int
Get partition count of a store.protected void
initBackend
(ClientConfig clientConfig, VeniceConfigLoader configLoader, Optional<Set<String>> managedClients, ICProvider icProvider, Optional<ObjectCacheConfig> cacheConfig) protected boolean
isPartitionReadyToServe
(VersionBackend versionBackend, int partition) Check if user partition is ready to serve traffic.protected boolean
isPartitionSubscribed
(VersionBackend versionBackend, int partition) boolean
boolean
isReady()
void
start()
void
streamingBatchGet
(Set<K> keys, StreamingCallback<K, V> callback) Streaming interface forAvroGenericStoreClient.batchGet(Set)
.protected CompletableFuture<Void>
subscribe
(ComplementSet<Integer> partitions) Ingest specific partition/partitions locally.Ingest the entire data (i.e.protected void
toString()
protected void
unsubscribe
(ComplementSet<Integer> partitions) void
unsubscribe
(Set<Integer> partitions) Stop ingesting a partition locally, and drop its associated local states/data.void
Stop ingesting all subscribed partition locally, and drop their associated local states/data.Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait
Methods inherited from interface com.linkedin.venice.client.store.AvroGenericReadComputeStoreClient
compute, compute, compute, computeAggregation
Methods inherited from interface com.linkedin.venice.client.store.AvroGenericStoreClient
streamingBatchGet
-
Field Details
-
logger
protected final org.apache.logging.log4j.Logger logger -
READ_CHUNK_EXECUTOR
The following two fields are used to speed up the requests with a big number of keys: 1. Split the big request into smaller chunks. 2. Execute these chunks concurrently. -
DEFAULT_CHUNK_SPLIT_THRESHOLD
public static final int DEFAULT_CHUNK_SPLIT_THRESHOLD- See Also:
-
-
Constructor Details
-
AvroGenericDaVinciClient
public AvroGenericDaVinciClient(DaVinciConfig daVinciConfig, ClientConfig clientConfig, VeniceProperties backendConfig, Optional<Set<String>> managedClients) -
AvroGenericDaVinciClient
public AvroGenericDaVinciClient(DaVinciConfig daVinciConfig, ClientConfig clientConfig, VeniceProperties backendConfig, Optional<Set<String>> managedClients, ICProvider icProvider, Executor readChunkExecutorForLargeRequest) -
AvroGenericDaVinciClient
protected AvroGenericDaVinciClient(DaVinciConfig daVinciConfig, ClientConfig clientConfig, VeniceProperties backendConfig, Optional<Set<String>> managedClients, ICProvider icProvider, AbstractAvroChunkingAdapter<V> chunkingAdapter, Runnable preValidation, Executor readChunkExecutorForLargeRequest)
-
-
Method Details
-
getSchemaReader
- Specified by:
getSchemaReader
in interfaceAvroGenericReadComputeStoreClient<K,
V>
-
getStoreName
- Specified by:
getStoreName
in interfaceAvroGenericStoreClient<K,
V>
-
getKeySchema
public org.apache.avro.Schema getKeySchema()Description copied from interface:AvroGenericStoreClient
Get key schema.- Specified by:
getKeySchema
in interfaceAvroGenericStoreClient<K,
V>
-
getLatestValueSchema
Deprecated.Description copied from interface:AvroGenericStoreClient
Get the latest value schema known in current store client. This function doesn't guarantee it will return the latest schema if you add a new value schema when current store client is running.- Specified by:
getLatestValueSchema
in interfaceAvroGenericStoreClient<K,
V>
-
getPartitionCount
public int getPartitionCount()Description copied from interface:DaVinciClient
Get partition count of a store.- Specified by:
getPartitionCount
in interfaceDaVinciClient<K,
V> - Returns:
- partition count
-
subscribeAll
Description copied from interface:DaVinciClient
Ingest the entire data (i.e. all partitions) locally.- Specified by:
subscribeAll
in interfaceDaVinciClient<K,
V> - Returns:
- a future which completes when the data is ready to serve
-
subscribe
Description copied from interface:DaVinciClient
Ingest specific partition/partitions locally.- Specified by:
subscribe
in interfaceDaVinciClient<K,
V> - Parameters:
partitions
- the set of partition IDs to subscribe to- Returns:
- a future which completes when the partitions are ready to serve
-
subscribe
-
unsubscribeAll
public void unsubscribeAll()Description copied from interface:DaVinciClient
Stop ingesting all subscribed partition locally, and drop their associated local states/data. If applications intend to keep the states/data for future use, no need to invoke this function before callingAvroGenericStoreClient.close()
.- Specified by:
unsubscribeAll
in interfaceDaVinciClient<K,
V>
-
unsubscribe
Description copied from interface:DaVinciClient
Stop ingesting a partition locally, and drop its associated local states/data. If applications intend to keep the states/data for future use, no need to invoke this function before callingAvroGenericStoreClient.close()
.- Specified by:
unsubscribe
in interfaceDaVinciClient<K,
V> - Parameters:
partitions
- the set of partition IDs to unsubscribe from
-
unsubscribe
-
get
Description copied from interface:AvroGenericStoreClient
Lookup the value by given key, and get(key).get() will return null if it doesn't exist. For now, if any backend error/exception happens, get(Object key).get() will throwExecutionException
, which is a wrapper of the real exception.- Specified by:
get
in interfaceAvroGenericStoreClient<K,
V> - Returns:
-
get
Description copied from interface:AvroGenericStoreClient
Similar toAvroGenericStoreClient.get(Object)
except that it allows passing in a instance, to minimize GC.- Specified by:
get
in interfaceAvroGenericStoreClient<K,
V>
-
split
-
batchGet
Description copied from interface:AvroGenericStoreClient
Get the values associated with the given keys and return them in a map of keys to values. Note that the returned map will only contain entries for the keys which have a value associated with them.- Specified by:
batchGet
in interfaceAvroGenericStoreClient<K,
V> - Returns:
- Throws:
VeniceClientException
-
streamingBatchGet
public void streamingBatchGet(Set<K> keys, StreamingCallback<K, V> callback) throws VeniceClientExceptionDescription copied from interface:AvroGenericStoreClient
Streaming interface forAvroGenericStoreClient.batchGet(Set)
. You can find more info inStreamingCallback
.- Specified by:
streamingBatchGet
in interfaceAvroGenericStoreClient<K,
V> - Throws:
VeniceClientException
-
isProjectionFieldValidationEnabled
public boolean isProjectionFieldValidationEnabled()- Specified by:
isProjectionFieldValidationEnabled
in interfaceAvroGenericReadComputeStoreClient<K,
V>
-
compute
public ComputeRequestBuilder<K> compute(Optional<ClientStats> stats, AvroGenericReadComputeStoreClient computeStoreClient) throws VeniceClientException - Specified by:
compute
in interfaceAvroGenericReadComputeStoreClient<K,
V> - Throws:
VeniceClientException
-
compute
public void compute(ComputeRequestWrapper computeRequestWrapper, Set<K> keys, org.apache.avro.Schema resultSchema, StreamingCallback<K, ComputeGenericRecord> callback, long preRequestTimeInNS) throws VeniceClientException- Specified by:
compute
in interfaceAvroGenericReadComputeStoreClient<K,
V> - Throws:
VeniceClientException
-
computeWithKeyPrefixFilter
public void computeWithKeyPrefixFilter(byte[] keyPrefix, ComputeRequestWrapper computeRequestWrapper, StreamingCallback<org.apache.avro.generic.GenericRecord, org.apache.avro.generic.GenericRecord> callback) - Specified by:
computeWithKeyPrefixFilter
in interfaceAvroGenericReadComputeStoreClient<K,
V>
-
isReady
public boolean isReady() -
isPartitionReadyToServe
Check if user partition is ready to serve traffic. -
isPartitionSubscribed
-
throwIfNotReady
protected void throwIfNotReady() -
getAvroChunkingAdapter
-
getGenericRecordChunkingAdapter
-
discoverService
-
initBackend
protected void initBackend(ClientConfig clientConfig, VeniceConfigLoader configLoader, Optional<Set<String>> managedClients, ICProvider icProvider, Optional<ObjectCacheConfig> cacheConfig) -
getBackend
-
start
public void start()- Specified by:
start
in interfaceAvroGenericStoreClient<K,
V>
-
close
public void close()Description copied from interface:AvroGenericStoreClient
Release the internal resources.- Specified by:
close
in interfaceAutoCloseable
- Specified by:
close
in interfaceAvroGenericStoreClient<K,
V> - Specified by:
close
in interfaceCloseable
-
toString
-