NebulaGraph v3.2.0 Release Note, many optimizations such as the performance of querying the shortest path

This article was first published on the Nebula Graph Community account

NebulaGraph v3.2.0 supports the extract() function, which optimizes the push-down of edge and point attribute filtering and the performance of querying the shortest path, and fixes problems such as the crash of the Storage service when scanning attributes concurrently.

Optimization

• Supports the extract() function .
• Optimize the configuration file and add some configurations.#4310
• Add optimization rules and remove useless AppendVertices operator.#4277
• Add optimization rules to optimize the pushdown of edge filtering.#4270
• Add optimization rules to optimize the pushdown of point attribute filtering.#4260
• Predictive filter for culling points.#4249
• Reduce the amount of data copying for join operations when moving data.#4283
• Get the attribute value through subscript, reducing the time of attribute query.#4242
• Optimize the performance of querying the shortest path.#4071
• Optimize the loop condition of query subgraph.#4226
• Reduce the amount of data copying for Traverse and AppendVertices operators when moving data.#4176
• Improve optimization rules to remove invalid item operators.#4157
• Use Arena Allocator to optimize memory allocation.#4239

Bug fixes

• Fix the web service crash when receiving some special attack messages.#4334
• Fix Storage service crash when scanning properties concurrently.#4268
• Fix Storage service crash when inserting edges exceeding the limit length.#4305
• Fix service crash when query concurrency mode is enabled.#4288
• Fix Storage service crash when looking for indexes with NULL properties.#4234
• Fix the bug that the standalone daemon exits after restart.#4269
• Fix the bug that the interpretation result of Join point format is incorrect due to two JSON conversions in the Graphviz online tool.#4280
• Fix the bug of property lookup, which does not allow full stop (.) in Schema.#4194
• Fixed the bug that the machine lost the key when restoring data.#4311
• Fixed the bug of displaying BAD TYPE when using the same statement to return different attributes of the same vertex.#4151
• Fix the error message bug of the statement MATCH p=(:team)-->() RETURN p LIMIT 1 when there is no index.#4053
• Enhanced error messages for operators AND and OR.#4304
• Fixed a bug where no statistics were available under index conditions.#4353
• Fix the bug of different time zones in the cluster.#4391
• Fixed crash when deleting full text index.#4384
• Fix the problem that the service crashes when sending a PUT request with an empty request body.#4405
• Fix the problem that Storage service crashes when the length of the VID in the statement exceeds the defined length when deleting points and edges based on an index.#4406

Historical Versions

History

Go to GitHub to experience this release: https://github.com/vesoft-inc/nebula/releases/tag/v3.2.0

Communication of graph database technology?To join the NebulaGraph exchange group, please fill in your NebulaGraph business card, and the NebulaGraph assistant will pull you into the group~~