During load process, the same operation are performed on all replicas such as sort and aggregation,
which are resource-intensive.
Concurrent data load would consume much CPU and memory resources.
It's better to perform write process (writing data into MemTable and then data flush) on single replica
and synchronize data files to other replicas before transaction finished.
BE can not graceful exit because some threads are running in endless
loop. This patch do the following optimization:
- Use the well encapsulated Thread and ThreadPool instead of std::thread
and std::vector<std::thread>
- Use CountDownLatch in thread's loop condition to avoid endless loop
- Introduce a new class Daemon for daemon works, like tcmalloc_gc,
memory_maintenance and calculate_metrics
- Decouple statistics type TaskWorkerPool and StorageEngine notification
by submit tasks to TaskWorkerPool's queue
- Reorder objects' stop and deconstruct in main(), i.e. stop network
services at first, then internal services
- Use libevent in pthreads mode, by calling evthread_use_pthreads(),
then EvHttpServer can exit gracefully in multi-threads
- Call brpc::Server's Stop() and ClearServices() explicitly
ISSUE-2069: This kind of query could be stuck.
The sender failed to send the last packet to receiver.
Also, the failure does not be reportted to FE , so the query is not cancelled.
The error log sames as "body_size=xxxx from xxx:xxx is too large".
The reason of the socket is that the packet of the query is too big which is more then the max_body_size of brpc.
This commit add a config named brpc_max_body_size whcih is used to change the max_body_size of brpc.
Also, user can change the max_body_size directly on-the-fly by "http://host:brpc_port/flags".
1. Implement Backend http server using libevent instead of mongoose.
2. Remove Old Hypertable rpc framework, use brpc instead.
3. Change rpc from FE to BE to brpc.
4. Fs broker support HDFS HA.
5. add more metrics to monitor.
6. Lots of bug fixed.
