Amazon Science 团队计划于VLDB 2024 (August 26-30 2024) 发布 redset 数据集

数据集介绍

        Redset是一个数据集，包含了三个月的AWS Redshift fleet 中选定实例样本上运行的用户查询元数据。

数据集用途

        Amazon Science 团队打算在VLDB2024 期间开放该部分数据， 虽然目前数据集还没有开放，但是从数据集的Schema 来看，和在VLDB 2024 会议期间公开该数据集。 可以大胆猜测Redshift 开发团队会在VLDB 上有重要论文发布，同时公布其真实用户的提升数据。 

        研究数据集的Schema ，从另外一方面看，顶级数据库大厂，对数据库的核心指标的描述，或者监控维度在这个Schema 里面已经得到应有的表达。该数据集未来一段时间应该会成为数据库领域优化的benchmark 数据集，值得大家关注 。  数据集链接请看https://www.selectdataset.com/dataset/1dfe70fc50251057041a91e5a882eb57。 

        后续数据集公开后，数据库领域感兴趣的小伙伴，可以第一时间去看看。

数据集 Schema

Column	Name Description
instance_id	Uniquely identifies a redshift cluster
cluster_size	Size of the cluster (only available for provisioned)
user_id	Identifies the user that issued the query
database_id	Identifies the database that was queried
query_id	Unique per instance
arrival_time	Timestamp when the query arrived on the system
compile_duration_ms	Time the query spent compiling in milliseconds
queue_duration_ms	Time the query spent queueing in milliseconds
execution_duration_ms	Time the query spent executing in milliseconds
feature_fingerprint	Hash value of the query fingerprint. A proxy for query-likeness, though not based on text. Will overestimate repetition.
was_aborted	Whether the query was aborted during its lifetime
was_cached	Whether the query was answered from result cache
cache_source_query_id	If query was answered from result cache, this is the query id for the query which populated the cache
query_type	Type of query, e.g.., select, copy, ...
num_permanent_tables_accessed	Number of permanent table accesses by the query (regular database table)
num_external_tables_accessed	Number of external tables accessed by the query
num_system_tables_accessed	Number of system tables accessed by the query
read_table_ids	Comma separated list of unique permanent table ids read by the query
write_table_ids	Comma separated list of unique table ids written to by the query
mbytes_scanned	Total number of megabytes scanned by the query
mbytes_spilled	Total number of megabytes spilled by the query
num_joins	Number of joins in the query plan
num_scans	Number of scans in the query plan
num_aggregations	Number of aggregations in the query plan