Question

我需要结合两个查询的帮助。查询1查看所有节点的主列表（Nodes.Caption）。但是，所有节点都不存在请求的数据（网络流量）。我希望组合查询从InterfaceTraffic表中提取网络流量（请参阅查询1）。但是，如果此表中没有现有数据，则从VIM_VMStatistics表（查询2）中提取数据。

查询1

SELECT Caption
    ,Round(InAVG, 2) AS NetInAverage_Kbps
    ,Round(InMAX, 2) AS NetInMax_Kbps
    ,Round(OutAVG, 2) AS NetOutAverage_Kbps
    ,Round(OutMAX, 2) AS NetOutMax_Kbps
FROM (
    SELECT caption
        ,nodeid
    FROM Nodes
    ) n
INNER JOIN (
    SELECT nodeid
        ,AVG(In_Averagebps) / 1024 AS InAVG
        ,MAX(In_Maxbps) / 1024 AS InMAX
        ,AVG(Out_Averagebps) / 1024 AS OutAVG
        ,MAX(Out_Maxbps) / 1024 AS OutMAX
    FROM InterfaceTraffic
    WHERE DATETIME >= DATEADD(MONTH, datediff(MONTH, 0, getdate()) - 1, 0)
        AND DATETIME < DATEADD(MONTH, datediff(MONTH, 0, getdate()), 0)
    GROUP BY nodeid
    ) l ON n.NodeID = l.NodeID
ORDER BY Caption

查询2

SELECT NAME
    ,Round(InAVG, 2) AS NetInAverage_Kbps
    ,Round(InMAX, 2) AS NetInMax_Kbps
    ,Round(OutAVG, 2) AS NetOutAverage_Kbps
    ,Round(OutMAX, 2) AS NetOutMax_Kbps
FROM (
    SELECT NAME
        ,VirtualMachineID
            ,NodeID
    FROM VIM_VirtualMachineNodes
    ) n
INNER JOIN (
    SELECT VirtualMachineID
        ,AVG(AvgNetworkReceiveRate) AS InAVG
        ,MAX(MaxNetworkReceiveRate) AS InMAX
        ,AVG(AvgNetworkTransmitRate) AS OutAVG
        ,MAX(MaxNetworkTransmitRate) AS OutMAX
    FROM VIM_VMStatistics
    WHERE DATETIME >= DATEADD(MONTH, datediff(MONTH, 0, getdate()) - 1, 0)
        AND DATETIME < DATEADD(MONTH, datediff(MONTH, 0, getdate()), 0)
    GROUP BY VirtualMachineID
    ) l ON n.VirtualMachineID = l.VirtualMachineID

Answer 1

这些可爱的小选择不需要。这将完全相同

SELECT n.Caption
    ,Round(InAVG, 2) AS NetInAverage_Kbps
    ,Round(InMAX, 2) AS NetInMax_Kbps
    ,Round(OutAVG, 2) AS NetOutAverage_Kbps
    ,Round(OutMAX, 2) AS NetOutMax_Kbps
FROM Nodes n
INNER JOIN (
  SELECT nodeid
        ,AVG(In_Averagebps) / 1024 AS InAVG
        ,MAX(In_Maxbps) / 1024 AS InMAX
        ,AVG(Out_Averagebps) / 1024 AS OutAVG
        ,MAX(Out_Maxbps) / 1024 AS OutMAX
  FROM InterfaceTraffic
  WHERE DATETIME >= DATEADD(MONTH, datediff(MONTH, 0, getdate()) - 1, 0)
    AND DATETIME < DATEADD(MONTH, datediff(MONTH, 0, getdate()), 0)
  GROUP BY nodeid
) l ON n.NodeID = l.NodeID
ORDER BY Caption

SELECT n.NAME
    ,Round(InAVG, 2) AS NetInAverage_Kbps
    ,Round(InMAX, 2) AS NetInMax_Kbps
    ,Round(OutAVG, 2) AS NetOutAverage_Kbps
    ,Round(OutMAX, 2) AS NetOutMax_Kbps
FROM VIM_VirtualMachineNodes n 
INNER JOIN (
  SELECT VirtualMachineID
        ,AVG(AvgNetworkReceiveRate) AS InAVG
        ,MAX(MaxNetworkReceiveRate) AS InMAX
        ,AVG(AvgNetworkTransmitRate) AS OutAVG
        ,MAX(MaxNetworkTransmitRate) AS OutMAX
  FROM VIM_VMStatistics
  WHERE DATETIME >= DATEADD(MONTH, datediff(MONTH, 0, getdate()) - 1, 0)
    AND DATETIME < DATEADD(MONTH, datediff(MONTH, 0, getdate()), 0)
  GROUP BY VirtualMachineID
) l ON n.VirtualMachineID = l.VirtualMachineID

我将假设VIM_VirtualMachineNodes有一个像节点一样的nodeid字段

有很多方法可以做到这一点，这是我认为最清晰的方式......

首先，您需要一个重要的节点列表。

SELECT DISTINCT nodeid 
FROM (
  SELECT nodeid from nodes
  UNION ALL
  SELECT nodeid from VIM_VirtualMachineNodes
) as node_list

现在我们将此连接到我们的两个列表并使用coalesce选择项目

SELECT 
  COALESCE(l1.nodeid, l2.nodeid, 0) as nodeid,
  COALESCE(l1.InAVG,  l2.InAVE, 0.0) AS NetInAverage_Kbps,
  COALESCE(l1.InMAX,  l2.InMAX, 0.0) AS NetInMax_Kbps,
  COALESCE(l1.OutAVG, l1.OutAVG, 0.0) AS NetOutAverage_Kbps,
  COALESCE(l1.OutMAX, l1.OutMAX, 0.0) AS NetOutMax_Kbps
FROM (
  SELECT DISTINCT nodeid 
  FROM (
    SELECT nodeid from nodes
    UNION ALL
    SELECT nodeid from VIM_VirtualMachineNodes
  ) as node_list
) as N
LEFT JOIN (
  SELECT nodeid
        ,AVG(In_Averagebps) / 1024 AS InAVG
        ,MAX(In_Maxbps) / 1024 AS InMAX
        ,AVG(Out_Averagebps) / 1024 AS OutAVG
        ,MAX(Out_Maxbps) / 1024 AS OutMAX
  FROM InterfaceTraffic
  WHERE DATETIME >= DATEADD(MONTH, datediff(MONTH, 0, getdate()) - 1, 0)
    AND DATETIME < DATEADD(MONTH, datediff(MONTH, 0, getdate()), 0)
  GROUP BY nodeid
) l1 ON n.NodeID = l1.NodeID
LEFT JOIN (
  SELECT nodeid
        ,AVG(AvgNetworkReceiveRate) AS InAVG
        ,MAX(MaxNetworkReceiveRate) AS InMAX
        ,AVG(AvgNetworkTransmitRate) AS OutAVG
        ,MAX(MaxNetworkTransmitRate) AS OutMAX
  FROM VIM_VMStatistics
  WHERE DATETIME >= DATEADD(MONTH, datediff(MONTH, 0, getdate()) - 1, 0)
    AND DATETIME < DATEADD(MONTH, datediff(MONTH, 0, getdate()), 0)
  GROUP BY VirtualMachineID
) l2 ON n.NodeID = l2.NodeID

正如您所看到的，我没有打扰这个名字，但如果您愿意，可以加入子查询中的那些。

Answer 2

您可以在具有左外连接的节点上对query1和query2进行查询。为简单起见，我们假设：

您有Query1和Query2
Query1和Query2每个节点标题最多返回一行

然后查询看起来像这样：

SELECT caption,
       (CASE WHEN Q1.Caption IS NULL THEN Q2.NetInAverage_Kbps
             ELSE Q1.NetInAverage_Kbps END) AS NetInAverage_Kpbs,
       (CASE WHEN Q1.Caption IS NULL THEN Q2.NetInMax_Kbps
             ELSE Q1.NetInMax_Kbps END) AS NetInMax_Kbps,
       (CASE WHEN Q1.Caption IS NULL THEN Q2.NetOutAverage_Kbps
             ELSE Q1.NetOutAverage_Kbps END) AS NetOutAverage_Kbps,
       (CASE WHEN Q1.Caption IS NULL THEN Q2.NetOutMax_Kbps
             ELSE Q1.NetOutAverage_Kbps END) AS NetOutMax_Kbps
FROM Nodes N
     LEFT OUTER JOIN Query1 Q1 ON N.Caption=Q1.Caption
     LEFT OUTER JOIN Query2 Q2 ON N.Caption=Q2.Name

此查询检查Query1中是否有行，如果没有，则使用Query2，否则使用Query1。

需要帮助组合查询

2 个答案: