[improve][ml] Support Bookkeeper batch read

[dao-jun]

Motivation

In BP-62, Bookkeeper starting to support batch read API,
the PR is to support Bookkeeper batch reading on the Pulsar side.

Modifications

1. Support Bookkeeper batch read API
2. Introduce a new flag bookkeeperEnableBatchRead to control enable BK batch read or not, the default value is false.

Verifying this change

• Make sure that the change passes the CI checks.

(Please pick either of the following options)

This change is a trivial rework / code cleanup without any test coverage.

(or)

This change is already covered by existing tests, such as (please describe tests).

(or)

This change added tests and can be verified as follows:

(example:)

• Added integration tests for end-to-end deployment with large payloads (10MB)
• Extended integration test for recovery after broker failure

Does this pull request potentially affect one of the following parts:

If the box was checked, please highlight the changes

• Dependencies (add or upgrade a dependency)
• The public API
• The schema
• The default values of configurations
• The threading model
• The binary protocol
• The REST endpoints
• The admin CLI options
• The metrics
• Anything that affects deployment

Documentation

• doc
• doc-required
• doc-not-needed
• doc-complete

Matching PR in forked repository

PR in forked repository: dao-jun#13

[liudezhi2098]

If maxSize = 0, info log will be printed for each call, which will affect performance.
https://github.com/apache/bookkeeper/blob/7c41204506122ed6904289f4814d4130274874aa/bookkeeper-server/src/main/java/org/apache/bookkeeper/client/LedgerHandle.java#L955-L967

private CompletableFuture<LedgerEntries> batchReadEntriesInternalAsync(long startEntry, int maxCount, long maxSize,
            boolean isRecoveryRead) {
        int nettyMaxFrameSizeBytes = clientCtx.getConf().nettyMaxFrameSizeBytes;
        if (maxSize > nettyMaxFrameSizeBytes) {
            LOG.info(
                "The max size is greater than nettyMaxFrameSizeBytes, use nettyMaxFrameSizeBytes:{} to replace it.",
                nettyMaxFrameSizeBytes);
            maxSize = nettyMaxFrameSizeBytes;
        }
        if (maxSize <= 0) {
            LOG.info("The max size is negative, use nettyMaxFrameSizeBytes:{} to replace it.", nettyMaxFrameSizeBytes);
            maxSize = nettyMaxFrameSizeBytes;
        }

[liudezhi2098]

If isAutoSkipNonRecoverableData is set to true, is the batch read failure consistent with the behavior here?

pulsar/managed-ledger/src/main/java/org/apache/bookkeeper/mledger/impl/OpReadEntry.java

Lines 115 to 129 in 66e1a06

	} else if (cursor.getConfig().isAutoSkipNonRecoverableData()
	&& exception instanceof NonRecoverableLedgerException) {
	log.warn("[{}][{}] read failed from ledger at position:{} : {}", cursor.ledger.getName(), cursor.getName(),
	readPosition, exception.getMessage());
	final ManagedLedgerImpl ledger = (ManagedLedgerImpl) cursor.getManagedLedger();
	Position nexReadPosition;
	Long lostLedger = null;
	if (exception instanceof ManagedLedgerException.LedgerNotExistException) {
	// try to find and move to next valid ledger
	nexReadPosition = cursor.getNextLedgerPosition(readPosition.getLedgerId());
	lostLedger = readPosition.getLedgerId();
	} else {
	// Skip this read operation
	nexReadPosition = ledger.getValidPositionAfterSkippedEntries(readPosition, count);
	}

[dao-jun]

If maxSize = 0, info log will be printed for each call, which will affect performance. https://github.com/apache/bookkeeper/blob/7c41204506122ed6904289f4814d4130274874aa/bookkeeper-server/src/main/java/org/apache/bookkeeper/client/LedgerHandle.java#L955-L967

private CompletableFuture<LedgerEntries> batchReadEntriesInternalAsync(long startEntry, int maxCount, long maxSize,
            boolean isRecoveryRead) {
        int nettyMaxFrameSizeBytes = clientCtx.getConf().nettyMaxFrameSizeBytes;
        if (maxSize > nettyMaxFrameSizeBytes) {
            LOG.info(
                "The max size is greater than nettyMaxFrameSizeBytes, use nettyMaxFrameSizeBytes:{} to replace it.",
                nettyMaxFrameSizeBytes);
            maxSize = nettyMaxFrameSizeBytes;
        }
        if (maxSize <= 0) {
            LOG.info("The max size is negative, use nettyMaxFrameSizeBytes:{} to replace it.", nettyMaxFrameSizeBytes);
            maxSize = nettyMaxFrameSizeBytes;
        }

apache/bookkeeper#4485 I created a PR to change the log level to debug

[liudezhi2098]

It needs to be made clear in the document that enabling batch does not necessarily mean batch reading.

[dao-jun]

Related BK PR: apache/bookkeeper#4485
apache/bookkeeper#4487