Overview: AutoSPInstaller is a great tool for building SharePoint 2010 ans 2013 farms. One section of AutoSPInstaller relates to installing SharePoint 2013 search. This post examines how to configure the xml that drives AutoSPInstaller and supplemental Ps to build the Search farm you require.
Below is an example of a search xml config that will setup Search with full redunancy on 2 servers in the SP farm. This design provides High Availaibility (HA). MS recommend keeping each index partition under 10 million items. So in this scenario if you follow MS recommendations you are limited to having 10 million items in your index (hardware, moving outher roles to more servers, usage of search to query all affect how quickly results will be returned so with tuning you can get your indexes to be larger and provide a good end user experience).
Tip: Create a seperate SQL aliase fo your 4 Search databases, this will alow you to move the dbs to a new instance or at 40mill documents MS recommend you consider a dedicate search farm. The SQL Alaise gives you room to adapt.
Tip: Group your search roles onto servers such as:
- Index & Query Processing
- Analytics & Content Processing
- Crawl, Content processing & Search Admin
$WFE2="WFE2"
$WFE4="WFE4"
$ssa = Get-SPEnterpriseSearchServiceApplication #Assuming you have a single Service application
$active = Get-SPEnterpriseSearchTopology -SearchApplication $ssa -Active
$clone = New-SPEnterpriseSearchTopology -SearchApplication $ssa -Clone –SearchTopology $active
New-SPEnterpriseSearchIndexComponent -SearchTopology $clone -SearchServiceInstance $WFE2 -IndexPartition 1
New-SPEnterpriseSearchIndexComponent -SearchTopology $clone -SearchServiceInstance $WFE4 -IndexPartition 1
Set-SPEnterpriseSearchTopology -Identity $clone
More Info:
https://autospinstaller.codeplex.com/discussions/453501
Example:
4 Server Search farm created using AutoSPInstaller. AutoSPInstaller version 3.87 only supports creating a farm using one partion index as shown below.
To add "Partion Index 1" to the farm using WFE1 and WFE2 to hold the index components use the following PowerShell:
$ssa = Get-SPEnterpriseSearchServiceApplication #Assuming you have a single Service application
$active = Get-SPEnterpriseSearchTopology -SearchApplication $ssa -Active
$clone = New-SPEnterpriseSearchTopology -SearchApplication $ssa -Clone –SearchTopology $active
New-SPEnterpriseSearchIndexComponent -SearchTopology $clone -SearchServiceInstance "WFE1" -IndexPartition 1
New-SPEnterpriseSearchIndexComponent -SearchTopology $clone -SearchServiceInstance "WFE3" -IndexPartition 1
Set-SPEnterpriseSearchTopology -Identity $clone
$active = Get-SPEnterpriseSearchTopology -SearchApplication $ssa -Active
$clone = New-SPEnterpriseSearchTopology -SearchApplication $ssa -Clone –SearchTopology $active
New-SPEnterpriseSearchIndexComponent -SearchTopology $clone -SearchServiceInstance "WFE1" -IndexPartition 1
New-SPEnterpriseSearchIndexComponent -SearchTopology $clone -SearchServiceInstance "WFE3" -IndexPartition 1
Set-SPEnterpriseSearchTopology -Identity $clone
Results in a farm that can handle up to 20 million documents due to the partioned index.
Another Example:
No comments:
Post a Comment