This CL mainly changes:
1. Avoid repeated sending of common components in Fragments
In the previous implementation, a query may generate multiple Fragments,
these Fragments contain some common information, such as DescriptorTable.
Fragment will be sent to BE in a certain order, so these public information will be sent repeatedly
and generated repeatedly on the BE side.
In some complex SQL, these public information may be very large,
thereby increasing the execution time of Fragment.
So I improved this. For multiple Fragments sent to the same BE, only the first Fragment will carry
these public information, and it will be cached on the BE side, and subsequent Fragments
no longer need to carry this information.
In the local test, the execution time of some complex SQL can be reduced from 3 seconds to 1 second.
2. Add the time-consuming part of FE logic in Profile
Including SQL analysis, planning, Fragment scheduling and sending on the FE side, and the time to fetch data.
1. When WITH_MYSQL is off, load error hub does not suport MySQL load error hub,
we should check its return value.
2. misjudge the return value of `change_row_block` in schema_change.cpp
We make all MemTrackers shared, in order to show MemTracker real-time consumptions on the web.
As follows:
1. nearly all MemTracker raw ptr -> shared_ptr
2. Use CreateTracker() to create new MemTracker(in order to add itself to its parent)
3. RowBatch & MemPool still use raw ptrs of MemTracker, it's easy to ensure RowBatch & MemPool destructor exec
before MemTracker's destructor. So we don't change these code.
4. MemTracker can use RuntimeProfile's counter to calc consumption. So RuntimeProfile's counter need to be shared
too. We add a shared counter pool to store the shared counter, don't change other counters of RuntimeProfile.
Note that, this PR doesn't change the MemTracker tree structure. So there still have some orphan trackers, e.g. RowBlockV2's MemTracker. If you find some shared MemTrackers are little memory consumption & too time-consuming, you could make them be the orphan, then it's fine to use the raw ptr.
Fix: #3946
CL:
1. Add prepare phase for `from_unixtime()`, `date_format()` and `convert_tz()` functions, to handle the format string once for all.
2. Find the cctz timezone when init `runtime state`, so that don't need to find timezone for each rows.
3. Add constant rewrite rule for `utc_timestamp()`
4. Add doc for `to_date()`
5. Comment out the `push_handler_test`, it can not run in DEBUG mode, will be fixed later.
6. Remove `timezone_db.h/cpp` and add `timezone_utils.h/cpp`
The performance shows bellow:
11,000,000 rows
SQL1: `select count(from_unixtime(k1)) from tbl1;`
Before: 8.85s
After: 2.85s
SQL2: `select count(from_unixtime(k1, '%Y-%m-%d %H:%i:%s')) from tbl1 limit 1;`
Before: 10.73s
After: 4.85s
The date string format seems still slow, we may need a further enhancement about it.
Mainly change:
1. Fix the bug in `update_status(status)` of `PlanFragmentExecutor`.
2. When the FE Coordinator executes `execRemoteFragmentAsync()`, if it finds an RPC error, return a Future with an error code instead of exception.
3. Protect the `_status` in RuntimeState with lock
4. Move the `_runtime_profile` of RuntimeState before the `_obj_pool`, so that the profile will be
deconstructed after the object pool.
5. Remove the unused `ObjectPool` param in RuntimeProfile constructor. If I don't remove it,
RuntimeProfile will depends on the `_obj_pool` in RuntimeProfile.
* 1. Add enable spilling in query option, support spill disk in Analytic_Eval_Node, FE can open enable spilling by
set enable_spilling = true;
Now, Sort Node and Analytic_Eval_Node can spill to disk.
2. Delete merge merge_sorter code we do not use now.
3. Replace buffered_tuple_stream by buffered_tuple_stream2 in Analytic_Eval_Node and support spill to disk. Delete the useless code of buffered_block_mgr and buffered_tuple_stream.
4. Add DataStreamRecvr Profile. Move the counter belong to DataStreamRecvr from fragment to DataStreamRecvr Profile to make clear of Running Profile.
* change some hint in code
* replace disable_spill with enable_spill which is better compatible to FE
LSAN detected errors have been fixed by a prior pathch (#3326), but
there are still some ASAN detected errors.
This patch try to fix these errors to make Doris BE more robustness.
And then we can add CI run in LSAN/ASAN mode to detect memory errors
as early as possible.
This CL mainly made the following modifications:
1. Delete Invalid method in Running Profile Class.
2. Move Memlimit Counter from blockmgr to fragment and add PeakMemUsage Counter
3. Fix the bug of buffer pool memlimit counter
4. Call compute_time_in_profile() before pretty_print() to show the _local_time_percent without child running profile
5. Add TransferThread ThreadToken count in AveThreadToken Counter
Remove unused LLVM related codes of directory (step 4):be/src/runtime (#2910)
there are many LLVM related codes in code base, but these codes are not really used.
The higher version of GCC is not compatible with the LLVM 3.4.2 version currently used by Doris.
The PR delete all LLVM related code of directory: be/src/runtime
This variable is mainly for INSERT operation, because INSERT operation has both query and load part.
Using only the exec_mem_limit variable does not make a good distinction of memory limit between the two parts.
The new msg of limit exceed: "Memory exceed limit. %msg, Backend:%ip, fragment:%id Used:% , Limit:%. xxx".
This commit unifies the msg of 'Memory exceed limit' such as check_query_state, RETURN_IF_LIMIT_EXCEEDED and LIMIT_EXCEEDED.
* Change label of broker load txn
1. put broker load label into txn label
2. fix the bug of `label is already used`
3. fix partition error of new broker load
* Fix count error in mini load and broker load
There are three params (num_rows_load_total, num_rows_load_filtered, num_rows_load_unselected) which are used to count dpp.norm.ALL and dpp.abnorm.ALL.
num_rows_load_total is the number rows of source file.
num_rows_load_unselected is the not satisfied (where conjuncts) rows of num_rows_load_total
num_rows_load_filtered is the rows (quality not good enough) of (num_rows_load_total-num_rows_load_unselected)
1. Add broker load error hub
A broker load error hub will collect error messages in load process and saves them as a file to the specified remote storage via broker. In case that in broker/min/streaming load process, user may not be able to access the error log file in Backend directly.
We also add a new header option: 'enable_hub' in streaming load request, and default is false. Because if we enable the broker load error hub, it will significantly slow down the processing speed of streaming load, due to the visit of remote storage via broker. So use can disable the error load hub using this header option, to avoid slowing down the load speed.
2. Show load error logs by using SHOW LOAD WARNINGS stmt
We also provide a more easy way to get load error logs. We implement 'SHOW LOAD WARNINGS ON 'url'' stmt to show load error logs directly. The 'url' in stmt is provided in 'SHOW LOAD' stmt.
eg:
show load warnings on "http://192.168.1.1:8040/api/_load_error_log?file=__shard_2/error_log_xxx";
3. Support now() function in broker load
User can mapping a column to now() in broker load stmt, which means this column will be filled with time when the ETL started.
4. Support more types of wildcard in broker load
Currently, we only support wildcard '*' to match the file names. wildcard like '/path/to/20190[1-4]*' is not support.
1. Print broker address for debug.
2. Do not letting backup job cancelled if it already in state UPLOAD_INFO.
3. Cancel task on Backends when job is cancelled.
4. Show detail progress of backup and restore job.
5. Make 'show snapshot' result more readable.
6. Change upload and download thread num of backup and restore in Backend to 1.
* Add UserFunctionCache to cache UDF's library
This patch replace LibCache with UserFunctionCache. LibCache use HDFS
URL to identify a UDF's Library, and when BE process restart all of
downloaded library should be loaded another time. We use function id
corresponding to a library, and when process restart, all downloaded
libraries can be loaded without another downloading.
* update
2. add 2 new proc '/current_queries' and '/current_backend_instances' to monitor the current running queries.
3. add a manual compaction api on Backend to trigger cumulative or base compaction manually.
4. add Frontend config 'max_bytes_per_broker_scanner' to limit to bytes per one broker scanner. This is to limit the memory cost of a single broker load job
5. add Frontend config 'max_unfinished_load_job' to limit load job number: if number of running load jobs exceed the limit, no more load job is allowed to be submmitted.
6. a log of bug fixed
