rockeet

> I'm working for remote compaction of RocksDB, which is an experimental feature. When I try to use this feature in MyRocks, I encounter one problem. > > It seems...

I noticed InnoDB has option `innodb_ddl_threads` for this purpose, it is straight forward to implement this feature in myrocks.

Is there any progress?

> This is likely due to the MySQL transaction coordinator. If there are two transaction engines installed and binlog disabled, MySQL uses the TC_LOG_MMAP transaction coordinator, when binlog is enabled...

> > Execute `Rdb_sst_info::commit_sst_file` in dedicated threads, this improves performance: > > ``` > > 1. `Rdb_sst_file_ordered::commit` may use stack to reverse input data, this is time consuming > >...

> > ``` > > 1. In `Rdb_sst_file_ordered::commit()`, if `m_use_stack` is true, it will write kv from stack. > > > > 2. In `Rdb_sst_file_ordered::Rdb_sst_file::commit()`, it calls `m_sst_file_writer->Finish()`, in which...

During creating index, an RocksDB iterator is used for fetching table rows, and there is a Version object held by the Iterator, the Version object pinned a set of MemTable...

> > Yes, by default DDL runs as a single transaction, so it is constrained by memory. > > We recommend using bulk loading -- https://github.com/facebook/mysql-5.6/wiki/Data-Loading -- while this is...

We have an visualization tool which can view structures inside an SST file, by this tool we found **name** is missing, then we added name in (un)materialize. 

# We are eagerly need this project!! + RocksDB on glusterfs We are planing to using glusterfs for our MyRocks(MySQL + RocksDB) fork because glusterfs is fast for random read...