Early Design Documentation: https://shimo.im/docs/DT6JXDRkdTvdyV3G
Implement a new way of memory statistics based on TCMalloc New/Delete Hook,
MemTracker and TLS, and it is expected that all memory new/delete/malloc/free
of the BE process can be counted.
1.
The methods in the IndexChannel are called back in the RpcClosure in the NodeChannel.
However, this callback may occur after the whole task is finished (e.g. due to network latency),
and by that time the IndexChannel may have been destructured, so we should not call
the IndexChannel methods anymore, otherwise the BE will crash.
Therefore, we use the `_is_closed` variable and `_closed_lock` to ensure that the RPC callback function
will not call the IndexChannel's method after the NodeChannel is closed.
2.
Do not add IndexChannel to the ObjectPool.
Because when deconstruct IndexChannel, it may call the deconstruction of NodeChannel.
And the deconstruction of NodeChannel maybe time consuming(wait rpc finished).
But the ObjectPool will hold a SpinLock to destroy the objects, so it may cause CPU busy.
Due to unlimited queue in OlapScanNode and NodeChannel, memory usage can be
very large for reading and writing large table, e.g 'insert into tableB select * from tableA'.
Modify the implementation of MemTracker:
1. Simplify a lot of useless logic;
2. Added MemTrackerTaskPool, as the ancestor of all query and import trackers, This is used to track the local memory usage of all tasks executing;
3. Add cosume/release cache, trigger a cosume/release when the memory accumulation exceeds the parameter mem_tracker_consume_min_size_bytes;
4. Add a new memory leak detection mode (Experimental feature), throw an exception when the remaining statistical value is greater than the specified range when the MemTracker is destructed, and print the accurate statistical value in HTTP, the parameter memory_leak_detection
5. Added Virtual MemTracker, cosume/release will not sync to parent. It will be used when introducing TCMalloc Hook to record memory later, to record the specified memory independently;
6. Modify the GC logic, register the buffer cached in DiskIoMgr as a GC function, and add other GC functions later;
7. Change the global root node from Root MemTracker to Process MemTracker, and remove Process MemTracker in exec_env;
8. Modify the macro that detects whether the memory has reached the upper limit, modify the parameters and default behavior of creating MemTracker, modify the error message format in mem_limit_exceeded, extend and apply transfer_to, remove Metric in MemTracker, etc.;
Modify where MemTracker is used:
1. MemPool adds a constructor to create a temporary tracker to avoid a lot of redundant code;
2. Added trackers for global objects such as ChunkAllocator and StorageEngine;
3. Added more fine-grained trackers such as ExprContext;
4. RuntimeState removes FragmentMemTracker, that is, PlanFragmentExecutor mem_tracker, which was previously used for independent statistical scan process memory, and replaces it with _scanner_mem_tracker in OlapScanNode;
5. MemTracker is no longer recorded in ReservationTracker, and ReservationTracker will be removed later;
In some scenarios, users cannot find a suitable hash key to avoid data skew, so we need to provide an additional data distribution for olap table to avoid data skew
example:
CREATE TABLE random_table
(
siteid INT DEFAULT '10',
citycode SMALLINT,
username VARCHAR(32) DEFAULT '',
pv BIGINT SUM DEFAULT '0'
)
AGGREGATE KEY(siteid, citycode, username)
DISTRIBUTED BY random BUCKETS 10
PROPERTIES("replication_num" = "1");
Co-authored-by: caiconghui1 <caiconghui1@jd.com>
Support thread pool per disk for scanners to prevent pool performance from some high ioutil disks happening
key point:
1. each disk has a thread pool for scanners
2. whenever a thread pool of one disk runs out of local work, tasks can be retrieved from other threads(disks). This is done round-robin.
performance testing:
vec version: 25% faster than single thread pool in a high io util disk test case
normal version: 8% faster than single thread pool in a high io util disk test case
1. Fix the problem of BE crash caused by destruct sequence. (close#8058)
2. Add a new BE config `compaction_task_num_per_fast_disk`
This config specify the max concurrent compaction task num on fast disk(typically .SSD).
So that for high speed disk, we can execute more compaction task at same time,
to compact the data as soon as possible
3. Avoid frequent selection of unqualified tablet to perform compaction.
4. Modify some log level to reduce the log size of BE.
5. Modify some clone logic to handle error correctly.
1. set both `tuple_offsets` and `new_tuple_offsets` in PRowBatch for compatibility
2. set FE config `repair_slow_replica` default to false
Avoid impacting the load process after upgrading.
Eg, if there are only 2 replicas, one is with high version count. After upgrade,
that replica will be set to bad, so that the load process will be stopped
because only 1 replica is alive.
3. Fix a bug that NodeChannel may be blocked at `close_wait()`
Forget to set `add_batch_finish` flag after the last rpc finished.
4. Fix a NPE of RoutineLoadScheduler
1. Added http interface return example in table-schema-action.md.
2. Correct typos in the document in error.md.
3. Modify the content of the code comments in the text_converter.hpp file.
Support implement UDF through GRPC protocol. This brings several benefits:
1. The udf implementation language is not limited to c++, users can use any familiar language to implement udf
2. UDF is decoupled from Doris, udf will not cause doris coredump, udf computing resources are separated from doris, and doris services are not affected
But RPC's UDF has a fixed overhead, so its performance is much slower than C++ UDF, especially when the amount of data is large.
Create function like
```
CREATE FUNCTION rpc_add(INT, INT) RETURNS INT PROPERTIES (
"SYMBOL"="add_int",
"OBJECT_FILE"="127.0.0.1:9999",
"TYPE"="RPC"
);
```
Function service need to implement `check_fn` and `fn_call` methods
Note:
THIS IS AN EXPERIMENTAL FEATURE, THE INTERFACE AND DATA STRUCTURE MAY BE CHANGED IN FUTURE !!!
This PR mainly changes:
1. Fix bug when enable `transfer_data_by_brpc_attachment`
In `data_stream_sender`, we will send a serialized PRowBatch data to multiple Channels.
And if `transfer_data_by_brpc_attachment` is enabled, we will mistakenly clear the data in PRowBatch
after sending PRowBatch to the first Channel.
As a result, the following Channel cannot receive the correct data, causing an error.
So I use a separate buffer instead of `tuple_data` in PRowBatch to store the serialized data
and reuse it in multiple channels.
2. Fix bug that the the offset in serialized row batch may overflow
Use int64 to replace int32 offset. And for compatibility, add a new field `new_tuple_offsets` in PRowBatch.
Currently, if we encounter a problem with a replica of a tablet during the load process,
such as a write error, rpc error, -235, etc., it will cause the entire load job to fail,
which results in a significant reduction in Doris' fault tolerance.
This PR mainly changes:
1. refined the judgment of failed replicas in the load process, so that the failure of a few replicas will not affect the normal completion of the load job.
2. fix a bug introduced from #7754 that may cause BE coredump
This PR mainly changes:
1. Help to Cancel the load job ASAP when encounter unqualified data.
Solution is described in #6318 .
Also replace some std::stringstream with fmt::memory_buffer to avoid performance issues.
2. fix a NPE bug when create user with empty host
3. fix compile warning after rebasing the master(vectorization)
# Proposed changes
Issue Number: close#6238
Co-authored-by: HappenLee <happenlee@hotmail.com>
Co-authored-by: stdpain <34912776+stdpain@users.noreply.github.com>
Co-authored-by: Zhengguo Yang <yangzhgg@gmail.com>
Co-authored-by: wangbo <506340561@qq.com>
Co-authored-by: emmymiao87 <522274284@qq.com>
Co-authored-by: Pxl <952130278@qq.com>
Co-authored-by: zhangstar333 <87313068+zhangstar333@users.noreply.github.com>
Co-authored-by: thinker <zchw100@qq.com>
Co-authored-by: Zeno Yang <1521564989@qq.com>
Co-authored-by: Wang Shuo <wangshuo128@gmail.com>
Co-authored-by: zhoubintao <35688959+zbtzbtzbt@users.noreply.github.com>
Co-authored-by: Gabriel <gabrielleebuaa@gmail.com>
Co-authored-by: xinghuayu007 <1450306854@qq.com>
Co-authored-by: weizuo93 <weizuo@apache.org>
Co-authored-by: yiguolei <guoleiyi@tencent.com>
Co-authored-by: anneji-dev <85534151+anneji-dev@users.noreply.github.com>
Co-authored-by: awakeljw <993007281@qq.com>
Co-authored-by: taberylyang <95272637+taberylyang@users.noreply.github.com>
Co-authored-by: Cui Kaifeng <48012748+azurenake@users.noreply.github.com>
## Problem Summary:
### 1. Some code from clickhouse
**ClickHouse is an excellent implementation of the vectorized execution engine database,
so here we have referenced and learned a lot from its excellent implementation in terms of
data structure and function implementation.
We are based on ClickHouse v19.16.2.2 and would like to thank the ClickHouse community and developers.**
The following comment has been added to the code from Clickhouse, eg:
// This file is copied from
// https://github.com/ClickHouse/ClickHouse/blob/master/src/Interpreters/AggregationCommon.h
// and modified by Doris
### 2. Support exec node and query:
* vaggregation_node
* vanalytic_eval_node
* vassert_num_rows_node
* vblocking_join_node
* vcross_join_node
* vempty_set_node
* ves_http_scan_node
* vexcept_node
* vexchange_node
* vintersect_node
* vmysql_scan_node
* vodbc_scan_node
* volap_scan_node
* vrepeat_node
* vschema_scan_node
* vselect_node
* vset_operation_node
* vsort_node
* vunion_node
* vhash_join_node
You can run exec engine of SSB/TPCH and 70% TPCDS stand query test set.
### 3. Data Model
Vec Exec Engine Support **Dup/Agg/Unq** table, Support Block Reader Vectorized.
Segment Vec is working in process.
### 4. How to use
1. Set the environment variable `set enable_vectorized_engine = true; `(required)
2. Set the environment variable `set batch_size = 4096; ` (recommended)
### 5. Some diff from origin exec engine
https://github.com/doris-vectorized/doris-vectorized/issues/294
## Checklist(Required)
1. Does it affect the original behavior: (No)
2. Has unit tests been added: (Yes)
3. Has document been added or modified: (No)
4. Does it need to update dependencies: (No)
5. Are there any changes that cannot be rolled back: (Yes)
If an load task has a relatively short timeout, then we need to ensure that
each RPC of this task does not get blocked for a long time.
And an RPC is usually blocked for two reasons.
1. handling "memory exceeds limit" in the RPC
If the system finds that the memory occupied by the load exceeds the threshold,
it will select the load channel that occupies the most memory and flush the memtable in it.
this operation is done in the RPC, which may be more time consuming.
2. close the load channel
When the load channel receives the last batch, it will end the task.
It will wait for all memtables flushes to finish synchronously. This process is also time consuming.
Therefore, this PR solves this problem by.
1. Use timeout to determine whether it is a high-priority load task
If the timeout of an load task is relatively short, then we mark it as a high-priority task.
2. not processing "memory exceeds limit" for high priority tasks
3. use a separate flush thread to flush memtable for high priority tasks.
1. fix core dump when using multi explode_bitmap #7716
2. fix bug that json array extract by json path is wrong #7717
3. fix bug that after lateral view, the null value become non-null value #7718
4. fix bug that lateral view may return error: couldn't resolve slot descriptor 1. #7719
5. fix error result when using lateral view with where predicate #7720
Support merge IN predicate when exist remote target(e.g. shuffle hash join).
Remote the code that IN predicate implicit conversion to Bloom filter then exist remote target.
Close related #7546
1. Consider the responsibility of Reader, Rename Reader to TabletReader, I think the new name TabletReader can represent its function exactly, it is more suitable and meaningful
2. add virtual keyword for the destructor of OlapScanner, because VOlapScanner is derived from it
3. refactor struct ReaderParams and KeysParam as TabletReader's inner struct，guard by TabletReader name scope, it's also more reasonable
4. reduce OlapScanner's member data amount, just use _parent->member_data is simpler
5. bugfix: TupleReader has the same memeber data _collect_iter to its parent class Reader, this usage is dangerous, the writer may make some mistake, so i delete TupleReader::_collect_iter to fix it.
6. call set_tablet_reader() in OlapScanner::prepare() to setup _tablet_reader, VOlapScanner should override set_tablet_reader to new BlockReader instead, use this way to avoid new Reader twice by reset unique_ptr _tablet_reader
7. if the member data is a inseparable part of a class, i suggest using normal variable while not pointer variable, because pointer bring a indirect lay and must handle coping and destructing carefully, it's not necessary
8. some other small changes for readability or design
1. Delete useless variables
2. Add const modifier for read-only function
3. Delete the empty destructor, the compiler will automatically generate it, refer to the 3/5/0 rule:
[https://en.cppreference.com/w/cpp/language/rule_of_three]
4. It is recommended to add the override keyword (instead of the virtual keyword) to the subclass virtual function.
Override will let the compiler help check and improve security. This is also the reason why C++11 introduces override
