SQL Server 各个版本支持的 Analysis Services 功能Analysis Services features supported by SQL Server editions

适用于: 是 SQL Server 2016 及更高版本 Analysis Services  不 Azure Analysis Services  Power BI PremiumAPPLIES TO: yesSQL Server 2016 and later Analysis Services noAzure Analysis Services noPower BI Premium

本文介绍 SQL Server 2016、2017、2019 Analysis Services 的不同版本支持的功能。This article describes features supported by different editions of SQL Server 2016, 2017, 2019 Analysis Services. 评估版支持企业版功能。Evaluation edition supports Enterprise edition features.

Analysis Services (服务器)Analysis Services (servers)

FeatureFeature EnterpriseEnterprise StandardStandard WebWeb Express with Advanced ServicesExpress with Advanced Services Express with ToolsExpress with Tools ExpressExpress 开发人员Developer
可伸缩共享数据库Scalable shared databases Yes Yes
备份/还原和附加/分离数据库Backup/Restore & Attach/Detach databases Yes Yes Yes
同步数据库Synchronize databases Yes Yes
故障转移群集实例Failover cluster instances Yes

节点数是操作系统支持的最大值Number of nodes is the operating system maximum
Yes

支持 2 个节点Support for 2 nodes
Yes

节点数是操作系统支持的最大值Number of nodes is the operating system maximum
可编程性(AMO、ADOMD.Net、OLEDB、XML/A、ASSL、TMSL)Programmability (AMO, ADOMD.Net, OLEDB, XML/A, ASSL, TMSL) Yes Yes Yes

表格模型Tabular models

FeatureFeature EnterpriseEnterprise StandardStandard WebWeb Express with Advanced ServicesExpress with Advanced Services Express with ToolsExpress with Tools ExpressExpress 开发人员Developer
层次结构Hierarchies Yes Yes Yes
KPIKPIs Yes Yes Yes
透视Perspectives Yes Yes
翻译Translations Yes Yes Yes
DAX 计算、DAX 查询、MDX 查询DAX calculations, DAX queries, MDX queries Yes Yes Yes
行级安全性Row-level security Yes Yes Yes
多个分区Multiple partitions Yes Yes
计算组Calculation groups 是(从 SQL Server 2019 开始)Yes (beginning with SQL Server 2019) 是(从 SQL Server 2019 开始)Yes (beginning with SQL Server 2019) 是(从 SQL Server 2019 开始)Yes (beginning with SQL Server 2019)
查询交叉Query interleaving 是(从 SQL Server 2019 开始)Yes (beginning with SQL Server 2019) 是(从 SQL Server 2019 开始)Yes (beginning with SQL Server 2019) 是(从 SQL Server 2019 开始)Yes (beginning with SQL Server 2019)
多对多关系Many-to-many relationships 是(从 SQL Server 2019 开始)Yes (beginning with SQL Server 2019) 是(从 SQL Server 2019 开始)Yes (beginning with SQL Server 2019) 是(从 SQL Server 2019 开始)Yes (beginning with SQL Server 2019)
内存中存储模式In-memory storage mode Yes Yes Yes
DirectQuery 模式DirectQuery mode Yes 是(从 SQL Server 2019 开始)Yes (beginning with SQL Server 2019) Yes

多维模型Multidimensional models

FeatureFeature EnterpriseEnterprise StandardStandard WebWeb Express with Advanced ServicesExpress with Advanced Services Express with ToolsExpress with Tools ExpressExpress 开发人员Developer
半累加性度量值Semi-additive measures Yes 1No 1 Yes
层次结构Hierarchies Yes Yes Yes
KPIKPIs Yes Yes Yes
透视Perspectives Yes Yes
操作Actions Yes Yes Yes
帐户智能Account intelligence Yes Yes Yes
时间智能Time intelligence Yes Yes Yes
自定义汇总Custom rollups Yes Yes Yes
写回多维数据集Writeback cube Yes Yes Yes
写回维度Writeback dimensions 2Yes 2 2Yes 2
写回单元Writeback cells Yes Yes Yes
钻取Drillthrough Yes Yes Yes
高级层次结构类型(父子、不规则层次结构)Advanced hierarchy types (parent-child and ragged hierarchies) Yes Yes Yes
高级维度(引用维度、多对多维度)Advanced dimensions (reference dimensions, many-to-many dimensions) Yes Yes Yes
链接度量值和维度Linked measures and dimensions Yes 3Yes 3 Yes
翻译Translations Yes Yes Yes
AggregationsAggregations Yes Yes Yes
多个分区Multiple partitions Yes 支持,最多 3 个Yes, up to 3 Yes
主动缓存Proactive caching Yes Yes
自定义程序集(存储过程)Custom assemblies (stored procedures) Yes Yes Yes
MDX 查询和脚本MDX queries and scripts Yes Yes Yes
DAX 查询DAX queries Yes Yes Yes
基于角色的安全模型Role-based security model Yes Yes Yes
维度和单元级别的安全性Dimension and cell-level security Yes Yes Yes
可扩展字符串存储Scalable string storage Yes Yes Yes
MOLAP、ROLAP 和 HOLAP 存储模式MOLAP, ROLAP, and HOLAP storage models Yes Yes Yes
二进制和压缩的 XML 传输Binary and compressed XML transport Yes Yes Yes
推送模式处理Push-mode processing Yes Yes
度量值表达式Measure expressions Yes Yes

[1]标准版支持 LastChild 半累加性度量值,但不支持其他半累加性度量值,例如 None、FirstChild、FirstNonEmpty、LastNonEmpty、AverageOfChildren 和 ByAccount。[1] The LastChild semi-additive measure is supported in Standard edition, but other semi-additive measures, such as None, FirstChild, FirstNonEmpty, LastNonEmpty, AverageOfChildren, and ByAccount, are not. 在所有版本上都支持累加性度量值(如 Sum、Count、Min 和 Max)和非累加性度量值 (DistinctCount)。Additive measures, such as Sum, Count, Min, Max, and non-additive measures (DistinctCount) are supported on all editions.

[2]在2019和更高版本中已停止写回维度 SQL Server Analysis Services。[2] Writeback dimensions are discontinued in SQL Server Analysis Services 2019 and later.

[3]标准版支持在同一个数据库中链接度量值和维度,但不支持来自其他数据库或实例。[3] Standard edition supports linking measures and dimensions within the same database, but not from other databases or instances.

Power Pivot for SharePointPower Pivot for SharePoint

FeatureFeature EnterpriseEnterprise StandardStandard WebWeb Express with Advanced ServicesExpress with Advanced Services Express with ToolsExpress with Tools ExpressExpress 开发人员Developer
基于共享服务体系结构的 SharePoint 场集成SharePoint farm integration based on shared service architecture Yes Yes
使用情况报告Usage reporting Yes Yes
运行状况监视规则Health monitoring rules Yes Yes
PowerPivot 库Power Pivot gallery Yes Yes
PowerPivot 数据刷新Power Pivot data refresh Yes Yes
PowerPivot 数据馈送Power Pivot data feeds Yes Yes

数据挖掘Data Mining

备注

SQL Server Analysis Services 2017 中弃用了数据挖掘。Data mining is deprecated in SQL Server Analysis Services 2017.

功能名称Feature Name EnterpriseEnterprise StandardStandard WebWeb Express with Advanced ServicesExpress with Advanced Services Express with ToolsExpress with Tools ExpressExpress 开发人员Developer
标准算法Standard algorithms Yes Yes Yes
数据挖掘工具(向导、编辑器、查询生成器)Data mining tools (Wizards, Editors, Query Builders) Yes Yes Yes
交叉验证Cross validation Yes Yes
挖掘结构数据的筛选子集的模型Models on filtered subsets of mining structure data Yes Yes
时序:ARTXP 和 ARIMA 方法之间的自定义混和Time series: Custom blending between ARTXP and ARIMA methods Yes Yes
时序:使用新数据的预测Time series: Prediction with new data Yes Yes
无限制并发数据挖掘查询Unlimited concurrent DM queries Yes Yes
数据挖掘算法的高级配置和优化选项Advanced configuration & tuning options for data mining algorithms Yes Yes
支持插件算法Support for plug-in algorithms Yes Yes
并行模型处理Parallel model processing Yes Yes
时序:跨序列预测Time series: cross-series prediction Yes Yes
关联规则的无限制属性Unlimited attributes for association rules Yes Yes
序列预测Sequence prediction Yes Yes
Naïve Bayes、神经网络和逻辑回归的多个预测目标Multiple prediction targets for naïve Bayes, neural network and logistic regression Yes Yes