mirror of https://gitee.com/bigwinds/arangodb
removed one known issue that was fixed in the meantime
This commit is contained in:
parent
b9c000253d
commit
3bb8367ff4
|
@ -32,7 +32,7 @@ are present in the MMFiles engine:
|
||||||
are split into multiple smaller RocksDB transactions that are committed individually.
|
are split into multiple smaller RocksDB transactions that are committed individually.
|
||||||
The entire user transaction will not necessarily have ACID properties in this case.
|
The entire user transaction will not necessarily have ACID properties in this case.
|
||||||
|
|
||||||
The thresholds values for transaction sizes can be configured globally using the
|
The threshold values for transaction sizes can be configured globally using the
|
||||||
startup options
|
startup options
|
||||||
|
|
||||||
* `--rocksdb.intermediate-commit-size`: if the size of all operations in a transaction
|
* `--rocksdb.intermediate-commit-size`: if the size of all operations in a transaction
|
||||||
|
@ -55,10 +55,7 @@ The following known issues will be resolved in future releases:
|
||||||
|
|
||||||
* the RocksDB engine is not yet performance-optimized and potentially not well configured
|
* the RocksDB engine is not yet performance-optimized and potentially not well configured
|
||||||
|
|
||||||
* the edge index in the RocksDB engine is not memory-optimized. It has a cache that will
|
* collections for which a geo index is present will use collection-level write locks
|
||||||
currently use more memory than expected. This will be changed after 3.2.beta is released
|
|
||||||
|
|
||||||
* collections for which a geo index is present will use a collection-level write locks
|
|
||||||
even with the RocksDB engine. Reads from these collections can still be done in parallel
|
even with the RocksDB engine. Reads from these collections can still be done in parallel
|
||||||
but no writes
|
but no writes
|
||||||
|
|
||||||
|
|
Loading…
Reference in New Issue