Files
doris/be/src/exec/kudu_scan_node.cpp
2019-06-14 23:38:31 +08:00

345 lines
12 KiB
C++

// Licensed to the Apache Software Foundation (ASF) under one
// or more contributor license agreements. See the NOTICE file
// distributed with this work for additional information
// regarding copyright ownership. The ASF licenses this file
// to you under the Apache License, Version 2.0 (the
// "License"); you may not use this file except in compliance
// with the License. You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing,
// software distributed under the License is distributed on an
// "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
// KIND, either express or implied. See the License for the
// specific language governing permissions and limitations
// under the License.
#include "exec/kudu_scan_node.h"
#include <boost/algorithm/string.hpp>
#include <kudu/client/row_result.h>
#include <kudu/client/schema.h>
#include <kudu/client/value.h>
#include <thrift/protocol/TDebugProtocol.h>
#include <vector>
#include "exec/kudu_scanner.h"
#include "exec/kudu_util.h"
#include "exprs/expr.h"
// #include "gutil/gscoped_ptr.h"
// #include "gutil/strings/substitute.h"
// #include "gutil/stl_util.h"
#include "runtime/mem_pool.h"
#include "runtime/runtime_state.h"
#include "runtime/row_batch.h"
#include "runtime/string_value.h"
#include "runtime/tuple_row.h"
// #include "util/disk-info.h"
// #include "util/jni-util.h"
// #include "util/periodic-counter-updater.h"
// #include "util/runtime-profile-counters.h"
// #include "common/names.h"
using boost::algorithm::to_lower_copy;
using kudu::client::KuduClient;
using kudu::client::KuduColumnSchema;
using kudu::client::KuduPredicate;
using kudu::client::KuduRowResult;
using kudu::client::KuduSchema;
using kudu::client::KuduTable;
using kudu::client::KuduValue;
using kudu::Slice;
namespace doris {
const string KuduScanNode::KUDU_ROUND_TRIPS = "TotalKuduScanRoundTrips";
const string KuduScanNode::KUDU_REMOTE_TOKENS = "KuduRemoteScanTokens";
KuduScanNode::KuduScanNode(ObjectPool* pool, const TPlanNode& tnode,
const DescriptorTbl& descs)
: ScanNode(pool, tnode, descs),
_tuple_id(tnode.kudu_scan_node.tuple_id),
_next_scan_token_idx(0),
_num_active_scanners(0),
_done(false),
_thread_avail_cb_id(-1) {
DCHECK(KuduIsAvailable());
int max_row_batches = config::kudu_max_row_batches;
// if (max_row_batches <= 0) {
// TODO: See comment on hdfs-scan-node.
// This value is built the same way as it assumes that the scan node runs co-located
// with a Kudu tablet server and that the tablet server is using disks similarly as
// a datanode would.
// max_row_batches = 10 * (DiskInfo::num_disks() + DiskIoMgr::REMOTE_NUM_DISKS);
//}
_materialized_row_batches.reset(new RowBatchQueue(max_row_batches));
}
KuduScanNode::~KuduScanNode() {
DCHECK(is_closed());
}
Status KuduScanNode::prepare(RuntimeState* state) {
RETURN_IF_ERROR(ScanNode::prepare(state));
_runtime_state = state;
_scan_ranges_complete_counter =
ADD_COUNTER(runtime_profile(), _s_scan_ranges_complete_counter, TUnit::UNIT);
_kudu_round_trips = ADD_COUNTER(runtime_profile(), KUDU_ROUND_TRIPS, TUnit::UNIT);
_kudu_remote_tokens = ADD_COUNTER(runtime_profile(), KUDU_REMOTE_TOKENS, TUnit::UNIT);
DCHECK(state->desc_tbl().get_tuple_descriptor(_tuple_id) != NULL);
_tuple_desc = state->desc_tbl().get_tuple_descriptor(_tuple_id);
return Status::OK();
}
Status KuduScanNode::set_scan_ranges(const std::vector<TScanRangeParams>& scan_ranges) {
// Initialize the list of scan tokens to process from the TScanRangeParams.
// int num_remote_tokens = 0;
for (const TScanRangeParams& params: scan_ranges) {
// if (params.__isset.is_remote && params.is_remote) ++num_remote_tokens;
_scan_tokens.push_back(params.scan_range.kudu_scan_token);
}
// COUNTER_SET(kudu_remote_tokens_, num_remote_tokens);
return Status::OK();
}
Status KuduScanNode::open(RuntimeState* state) {
RETURN_IF_ERROR(ExecNode::open(state));
RETURN_IF_CANCELLED(state);
// RETURN_IF_ERROR(query_maintenance(state));
SCOPED_TIMER(_runtime_profile->total_time_counter());
const KuduTableDescriptor* table_desc =
static_cast<const KuduTableDescriptor*>(_tuple_desc->table_desc());
RETURN_IF_ERROR(CreateKuduClient(table_desc->kudu_master_addresses(), &_client));
uint64_t latest_ts = static_cast<uint64_t>(
std::max<int64_t>(0, state->query_options().kudu_latest_observed_ts));
VLOG_RPC << "Latest observed Kudu timestamp: " << latest_ts;
if (latest_ts > 0) _client->SetLatestObservedTimestamp(latest_ts);
KUDU_RETURN_IF_ERROR(_client->OpenTable(table_desc->table_name(), &_table),
"Unable to open Kudu table");
_num_scanner_threads_started_counter =
ADD_COUNTER(runtime_profile(), _s_num_scanner_threads_started, TUnit::UNIT);
// Reserve one thread.
// state->resource_pool()->ReserveOptionalTokens(1);
// if (state->query_options().num_scanner_threads > 0) {
// state->resource_pool()->set_max_quota(
// state->query_options().num_scanner_threads);
// }
// thread_avail_cb_id_ = state->resource_pool()->AddThreadAvailableCb(
// bind<void>(mem_fn(&KuduScanNode::thread_available_cb), this, _1));
thread_available_cb(state->resource_pool());
return Status::OK();
}
Status KuduScanNode::get_next(RuntimeState* state, RowBatch* row_batch, bool* eos) {
DCHECK(row_batch != NULL);
// RETURN_IF_ERROR(ExecDebugAction(TExecNodePhase::GETNEXT, state));
RETURN_IF_CANCELLED(state);
// RETURN_IF_ERROR(QueryMaintenance(state));
SCOPED_TIMER(_runtime_profile->total_time_counter());
SCOPED_TIMER(materialize_tuple_timer());
if (reached_limit() || _scan_tokens.empty()) {
*eos = true;
return Status::OK();
}
*eos = false;
RowBatch* materialized_batch = _materialized_row_batches->GetBatch();
if (materialized_batch != NULL) {
row_batch->acquire_state(materialized_batch);
_num_rows_returned += row_batch->num_rows();
COUNTER_SET(_rows_returned_counter, _num_rows_returned);
if (reached_limit()) {
int num_rows_over = _num_rows_returned - _limit;
row_batch->set_num_rows(row_batch->num_rows() - num_rows_over);
_num_rows_returned -= num_rows_over;
COUNTER_SET(_rows_returned_counter, _num_rows_returned);
*eos = true;
std::unique_lock<mutex> l(_lock);
_done = true;
_materialized_row_batches->shutdown();
}
delete materialized_batch;
} else {
*eos = true;
}
Status status;
{
std::unique_lock<mutex> l(_lock);
status = _status;
}
return status;
}
Status KuduScanNode::close(RuntimeState* state) {
if (is_closed()) {
return Status::OK();
}
SCOPED_TIMER(_runtime_profile->total_time_counter());
// PeriodicCounterUpdater::StopRateCounter(total_throughput_counter());
// PeriodicCounterUpdater::StopTimeSeriesCounter(bytes_read_timeseries_counter_);
// if (thread_avail_cb_id_ != -1) {
// state->resource_pool()->RemoveThreadAvailableCb(thread_avail_cb_id_);
// }
if (!_done) {
std::unique_lock<mutex> l(_lock);
_done = true;
_materialized_row_batches->shutdown();
}
_scanner_threads.join_all();
// scanner_threads_.JoinAll();
DCHECK_EQ(_num_active_scanners, 0);
_materialized_row_batches->Cleanup();
ExecNode::close(state);
return Status::OK();
}
void KuduScanNode::debug_string(int indentation_level, stringstream* out) const {
string indent(indentation_level * 2, ' ');
*out << indent << "KuduScanNode(tupleid=" << _tuple_id << ")";
}
const string* KuduScanNode::get_next_scan_token() {
std::unique_lock<mutex> lock(_lock);
if (_next_scan_token_idx >= _scan_tokens.size()) return NULL;
const string* token = &_scan_tokens[_next_scan_token_idx++];
return token;
}
Status KuduScanNode::get_conjunct_ctxs(vector<ExprContext*>* ctxs) {
return Expr::clone_if_not_exists(_conjunct_ctxs, _runtime_state, ctxs);
}
void KuduScanNode::thread_available_cb(ThreadResourceMgr::ResourcePool* pool) {
while (true) {
std::unique_lock<mutex> lock(_lock);
// All done or all tokens are assigned.
if (_done || _next_scan_token_idx >= _scan_tokens.size()) {
break;
}
// Check if we can get a token.
// if (!pool->TryAcquireThreadToken()) break;
++_num_active_scanners;
COUNTER_UPDATE(_num_scanner_threads_started_counter, 1);
// Reserve the first token so no other thread picks it up.
const string* token = &_scan_tokens[_next_scan_token_idx++];
// string name = Substitute("scanner-thread($0)",
// _num_scanner_threads_started_counter->value());
std::stringstream ss;
ss << "scanner-thread(" << _num_scanner_threads_started_counter->value() << ")";
std::string name = ss.str();
VLOG_RPC << "Thread started: " << name;
_scanner_threads.create_thread(
boost::bind(&KuduScanNode::run_scanner_thread, this, name, token));
// _scanner_threads.AddThread(new Thread("kudu-scan-node", name,
// &KuduScanNode::RunScannerThread, this, name, token));
}
}
Status KuduScanNode::process_scan_token(KuduScanner* scanner, const string& scan_token) {
RETURN_IF_ERROR(scanner->open_next_scan_token(scan_token));
bool eos = false;
while (!eos) {
std::auto_ptr<RowBatch> row_batch(new RowBatch(
row_desc(), _runtime_state->batch_size(), mem_tracker()));
RETURN_IF_ERROR(scanner->get_next(row_batch.get(), &eos));
while (!_done) {
scanner->keep_kudu_scanner_alive();
if (_materialized_row_batches->AddBatchWithTimeout(row_batch.get(), 1000000)) {
row_batch.release();
break;
}
}
}
if (eos) scan_ranges_complete_counter()->update(1);
return Status::OK();
}
void KuduScanNode::run_scanner_thread(KuduScanNode *scanNode, const string& name, const string* initial_token) {
scanNode->run_scanner(name, initial_token);
}
void KuduScanNode::run_scanner(const string& name, const string* initial_token) {
DCHECK(initial_token != NULL);
// SCOPED_THREAD_COUNTER_MEASUREMENT(scanner_thread_counters());
// SCOPED_THREAD_COUNTER_MEASUREMENT(runtime_state_->total_thread_statistics());
// Set to true if this thread observes that the number of optional threads has been
// exceeded and is exiting early.
bool optional_thread_exiting = false;
KuduScanner scanner(this, _runtime_state);
const string* scan_token = initial_token;
Status status = scanner.open();
if (status.ok()) {
while (!_done && scan_token != NULL) {
status = process_scan_token(&scanner, *scan_token);
if (!status.ok()) break;
/*
// Check if the number of optional threads has been exceeded.
if (_runtime_state->resource_pool()->optional_exceeded()) {
std::unique_lock<mutex> l(_lock);
// Don't exit if this is the last thread. Otherwise, the scan will indicate it's
// done before all scan tokens have been processed.
if (_num_active_scanners > 1) {
--_num_active_scanners;
optional_thread_exiting = true;
break;
}
}
*/
scan_token = get_next_scan_token();
}
}
scanner.close();
{
std::unique_lock<mutex> l(_lock);
if (!status.ok()) {
if (_status.ok()) {
_status = status;
_done = true;
}
}
// Decrement num_active_scanners_ unless handling the case of an early exit when
// optional threads have been exceeded, in which case it already was decremented.
if (!optional_thread_exiting) --_num_active_scanners;
if (_num_active_scanners == 0) {
_done = true;
_materialized_row_batches->shutdown();
}
}
// lock_ is released before calling ThreadResourceMgr::ReleaseThreadToken() which
// invokes ThreadAvailableCb() which attempts to take the same lock.
VLOG_RPC << "Thread done: " << name;
// runtime_state_->resource_pool()->release_thread_token(false);
}
} // namespace impala