fredia commented on code in PR #24812:
URL: https://github.com/apache/flink/pull/24812#discussion_r1628988650
##########
flink-state-backends/flink-statebackend-forst/src/main/java/org/apache/flink/state/forst/ForStWriteBatchOperation.java:
##########
@@ -55,22 +54,44 @@ public class ForStWriteBatchOperation implements
ForStDBOperation {
public CompletableFuture<Void> process() {
return CompletableFuture.runAsync(
() -> {
- try (WriteBatch writeBatch =
- new WriteBatch(batchRequest.size() *
PER_RECORD_ESTIMATE_BYTES)) {
+ try (ForStDBWriteBatchWrapper writeBatch =
+ new ForStDBWriteBatchWrapper(db, writeOptions,
batchRequest.size())) {
for (ForStDBPutRequest<?, ?> request : batchRequest) {
+ ColumnFamilyHandle cf =
request.getColumnFamilyHandle();
if (request.valueIsNull()) {
- // put(key, null) == delete(key)
- writeBatch.delete(
- request.getColumnFamilyHandle(),
- request.buildSerializedKey());
+ if (request instanceof ForStDBBunchPutRequest)
{
+ ForStDBBunchPutRequest<?> bunchPutRequest =
+ (ForStDBBunchPutRequest<?>)
request;
+ byte[] primaryKey =
bunchPutRequest.buildSerializedKey(null);
+ byte[] endKey =
ForStDBBunchPutRequest.nextBytes(primaryKey);
Review Comment:
We concatenate primary key and user key together as the key of rocksdb, and
physically organized like this:
```
< <primary key1 - user key1> , user value1>
< <primary key1 - user key2> , user value2>
< <primary key1 - user key3> , user value3>
< <primary key1 - user key4> , user value4>
```
Here, we use primary key as a prefix to calculate an interval to leverage
`deleteRange`.
But too much `deleteRange` may affect the read performance, and we will
consider changing it iterator.delete later.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
