Merge pull request #14354 from MicrosoftDocs/FromPublicRepo

huypub · web-flow · commit 8803a04be463 · 2020-04-01T15:10:10.000-07:00
Confirm merge from FromPublicRepo to master to sync with https://github.com/MicrosoftDocs/sql-docs (branch live)
diff --git a/docs/big-data-cluster/deployment-custom-configuration.md b/docs/big-data-cluster/deployment-custom-configuration.md
@@ -462,6 +462,8 @@ azdata bdc config add -c custom-bdc/bdc.json -j "$.spec.resources.zookeeper.spec
 azdata bdc config add -c custom-bdc/bdc.json -j "$.spec.resources.gateway.spec.nodeLabel=bdc-shared"
 azdata bdc config add -c custom-bdc/bdc.json -j "$.spec.resources.appproxy.spec.nodeLabel=bdc-shared"
 ```
+>[!NOTE]
+> Best practice avoids giving the Kubernetes master any of the above BDC roles. If you plan on assigning these roles to the Kubernetes master node anyway, you'll need to [remove its ``master:NoSchedule`` taint.](https://kubernetes.io/docs/concepts/configuration/taint-and-toleration/) Be aware that this could overload the master node and inhibit its ability to perform its Kubernetes management duties on larger clusters. It's normal to see some pods scheduled to the master on any deployment: they already tolerate the ``master:NoSchedule`` taint, and they're mostly used to help manage the cluster. 
 
 ## <a id="jsonpatch"></a> Other customizations using JSON patch files
 
diff --git a/docs/relational-databases/indexes/columnstore-indexes-data-loading-guidance.md b/docs/relational-databases/indexes/columnstore-indexes-data-loading-guidance.md
@@ -41,7 +41,7 @@ Bulk loading has these built-in performance optimizations:
 
 -   **Reduced Logging:** The data that is directly loaded into compressed row groups leads to significant reduction in the size of the log. For example, if data was compressed 10x, the corresponding transaction log will be roughly 10x smaller without requiring TABLOCK or Bulk-logged/Simple recovery model. Any data that goes to a delta rowgroup is fully logged. This includes any batch sizes that are less than 102,400 rows.  Best practice is to use batchsize >= 102400. Since there is no TABLOCK required, you can load the data in parallel. 
 
--   **Minimal logging:** You can get further reduction in logging if you follow the prerequisites for [minimal logging](../import-export/prerequisites-for-minimal-logging-in-bulk-import.md). However, unlike loading data into a rowstore, TABLOCK leads to an X lock on the table rather than a BU (Bulk Update) lock and therefore parallel data load cannot be done. For more information on locking, see [Locking and row versioning[(../sql-server-transaction-locking-and-row-versioning-guide.md).
+-   **Minimal logging:** You can get further reduction in logging if you follow the prerequisites for [minimal logging](../import-export/prerequisites-for-minimal-logging-in-bulk-import.md). However, unlike loading data into a rowstore, TABLOCK leads to an X lock on the table rather than a BU (Bulk Update) lock and therefore parallel data load cannot be done. For more information on locking, see [Locking and row versioning](../sql-server-transaction-locking-and-row-versioning-guide.md).
 
 -   **Locking Optimization:** The X lock on a row group is automatically acquired when loading data into a compressed row group. However, when bulk loading into a delta rowgroup, an X lock is acquired at rowgroup but [!INCLUDE[ssNoVersion](../../includes/ssnoversion-md.md)] still locks the PAGE/EXTENT because X rowgroup lock is not part of locking hierarchy.  
   
diff --git a/docs/relational-databases/views/modify-data-through-a-view.md b/docs/relational-databases/views/modify-data-through-a-view.md
@@ -18,7 +18,7 @@ monikerRange: ">=aps-pdw-2016||=azuresqldb-current||=azure-sqldw-latest||>=sql-s
 ---
 # Modify Data Through a View
 [!INCLUDE[tsql-appliesto-ss2008-asdb-asdw-pdw-md](../../includes/tsql-appliesto-ss2008-all-md.md)]
-  You can modify the data of an underlying base table in [!INCLUDE[ssCurrent](../../includes/sscurrent-md.md)] by using [!INCLUDE[ssManStudioFull](../../includes/ssmanstudiofull-md.md)] or [!INCLUDE[tsql](../../includes/tsql-md.md)].  
+  You can modify the data of an underlying base table in SQL Server by using [!INCLUDE[ssManStudioFull](../../includes/ssmanstudiofull-md.md)] or [!INCLUDE[tsql](../../includes/tsql-md.md)].  
   
   
 ##  <a name="BeforeYouBegin"></a> Before You Begin  
diff --git a/docs/t-sql/functions/sql-variant-property-transact-sql.md b/docs/t-sql/functions/sql-variant-property-transact-sql.md
@@ -1,7 +1,7 @@
 ---
 title: "SQL_VARIANT_PROPERTY (Transact-SQL) | Microsoft Docs"
 ms.custom: ""
-ms.date: "09/12/2017"
+ms.date: "02/25/2020"
 ms.prod: sql
 ms.prod_service: "database-engine, sql-database, sql-data-warehouse, pdw"
 ms.reviewer: ""
@@ -42,9 +42,9 @@ SQL_VARIANT_PROPERTY ( expression , property )
   
 |Value|Description|Base type of sql_variant returned|  
 |-----------|-----------------|----------------------------------------|  
-|**BaseType**|[!INCLUDE[ssNoVersion](../../includes/ssnoversion-md.md)] data type, such as:<br /><br /> **bigint**<br /><br /> **binary**<br /><br /> **char**<br /><br /> **date**<br /><br /> **datetime**<br /><br /> **datetime2**<br /><br /> **datetimeoffset**<br /><br /> **decimal**<br /><br /> **float**<br /><br /> **int**<br /><br /> **money**<br /><br /> **nchar**<br /><br /> **numeric**<br /><br /> **nvarchar**<br /><br /> **real**<br /><br /> **smalldatetime**<br /><br /> **smallint**<br /><br /> **smallmoney**<br /><br /> **time**<br /><br /> **tinyint**<br /><br /> **uniqueidentifier**<br /><br /> **varbinary**<br /><br /> **varchar**|**sysname**<br /><br /> NULL = Input is not valid.|  
-|**Precision**|Number of digits of the numeric base data type:<br /><br /> **datetime** = 23<br /><br />**datetime2** = 27<br /><br /> **smalldatetime** = 16<br /><br /> **float** = 53<br /><br /> **real** = 24<br /><br /> **decimal** (p,s) and **numeric** (p,s) = p<br /><br /> **money** = 19<br /><br /> **smallmoney** = 10<br /><br /> **bigint** = 19<br /><br /> **int** = 10<br /><br /> **smallint** = 5<br /><br /> **tinyint** = 3<br /><br /> **bit** = 1<br /><br /> All other types = 0|**int**<br /><br /> NULL = Input is not valid.|  
-|**Scale**|Number of digits to the right of the decimal point of the numeric base data type:<br /><br /> **decimal** (p,s) and **numeric** (p,s) = s<br /><br /> **money** and **smallmoney** = 4<br /><br /> **datetime** = 3<br /><br />**datetime2** = 7<br /><br /> all other types = 0|**int**<br /><br /> NULL = Input is not valid.|  
+|**BaseType**|[!INCLUDE[ssNoVersion](../../includes/ssnoversion-md.md)] data type, such as:<br /><br /> **bigint**<br /><br /> **binary**<br /><br /> **bit**<br /><br /> **char**<br /><br /> **date**<br /><br /> **datetime**<br /><br /> **datetime2**<br /><br /> **datetimeoffset**<br /><br /> **decimal**<br /><br /> **float**<br /><br /> **int**<br /><br /> **money**<br /><br /> **nchar**<br /><br /> **numeric**<br /><br /> **nvarchar**<br /><br /> **real**<br /><br /> **smalldatetime**<br /><br /> **smallint**<br /><br /> **smallmoney**<br /><br /> **time**<br /><br /> **tinyint**<br /><br /> **uniqueidentifier**<br /><br /> **varbinary**<br /><br /> **varchar**|**sysname**<br /><br /> NULL = Input is not valid.|  
+|**Precision**|Number of digits of the numeric base data type:<br /><br /> **date** = 10<br /><br /> **datetime** = 23<br /><br /> **datetime2** = 27<br /><br /> **datetime2** (s) = 19 when s = 0, else s + 20<br /><br /> **datetimeoffset** = 34<br /><br /> **datetimeoffset** (s) = 26 when s = 0, else s + 27<br /><br /> **smalldatetime** = 16<br /><br /> **time** = 16<br /><br /> **time** (s) = 8 when s = 0, else s + 9<br /><br /> **float** = 53<br /><br /> **real** = 24<br /><br /> **decimal** and **numeric** = 18<br /><br /> **decimal** (p,s) and **numeric** (p,s) = p<br /><br /> **money** = 19<br /><br /> **smallmoney** = 10<br /><br /> **bigint** = 19<br /><br /> **int** = 10<br /><br /> **smallint** = 5<br /><br /> **tinyint** = 3<br /><br /> **bit** = 1<br /><br /> All other types = 0|**int**<br /><br /> NULL = Input is not valid.|  
+|**Scale**|Number of digits to the right of the decimal point of the numeric base data type:<br /><br /> **decimal** and **numeric** = 0<br /><br /> **decimal** (p,s) and **numeric** (p,s) = s<br /><br /> **money** and **smallmoney** = 4<br /><br /> **datetime** = 3<br /><br /> **datetime2** = 7<br /><br /> **datetime2** (s) = s (0 - 7)<br /><br /> **datetimeoffset** = 7<br /><br /> **datetimeoffset** (s) = s (0 - 7)<br /><br /> **time** = 7<br /><br /> **time** (s) = s (0 - 7)<br /><br /> all other types = 0|**int**<br /><br /> NULL = Input is not valid.|  
 |**TotalBytes**|Number of bytes required to hold both the metadata and data of the value. This information would be useful in checking the maximum side of data in a **sql_variant** column. If the value is larger than 900, index creation fails.|**int**<br /><br /> NULL = Input is not valid.|  
 |**Collation**|Represents the collation of the particular **sql_variant** value.|**sysname**<br /><br /> NULL = Input is not valid.|  
 |**MaxLength**|Maximum data type length, in bytes. For example, **MaxLength** of **nvarchar(**50**)** is 100, **MaxLength** of **int** is 4.|**int**<br /><br /> NULL = Input is not valid.|  
