Implementing SharePoint Server 2010 on Dell vStart Solution - A Reference Architecture for a
SharePoint Server 2010 farm on vStart 100 Hyper-V Solution
18
(MPIO) policy - ―Least Queue Depth‖ - was used for load balancing the iSCSI traffic from the virtual
machine to the storage arrays.
Configuration of Web Front-End (WFE) servers
This SharePoint 2010 farm design included two WFE servers. The software matrix for these WFE servers
is shown in Table 5. SharePoint 2010 Standard Edition and included only out-of-the-box features used
for collaboration workloads. As a part of the SharePoint collaboration workload, only the standard
search service application was deployed and no other service applications, such as Excel and Visio,
were deployed.
All of the WFE servers were configured in a Network Load Balancing
(NLB) cluster. Using NLB,
stateless applications such as SharePoint WFE servers are made scalable by adding additional servers
when the load increases. In this reference architecture, the application server virtual machines were
not a part of the NLB cluster but can be added when additional capacity at the WFE tier is desired.
Redundant network connections were used at the Hyper-V host and also, the virtual switch for NLB
cluster communication was created using this teamed network. Therefore, NLB was configured to use
multicast mode to avoid IP address conflicts
in the farm. To access the SharePoint NLB cluster from
different IP subnets, an ARP entry for the NLB cluster name must be added on the LAN switch fabric.
Search service application configuration
SharePoint 2010 changed the search architecture and introduced high availability at the application
tier; for instance, the search crawler. The new search service application architecture in SharePoint
2010 includes greater redundancy. This new design provides flexibility and lets the query and crawler
roles be scaled-out separately on an as-needed basis. Search crawlers are now stateless; they do not
store a copy of the index. However, the index does still propagate and is stored locally on the query
servers. Two virtual machines hosting both crawler and query roles were used in the SharePoint 2010
farm.
The farm configuration used the search service application configuration shown in Figure 5. The search
crawler and query roles were hosted on both of the application servers to enable high availability and
provide improved search performance. Both the crawlers were associated with the same crawl
database, and indexed the same content source.
The search architecture shown in Figure 5 is a logical representation of how the SharePoint Search
service was configured. This configuration provides complete redundancy for both search crawler and
query roles.
NLB Deployment Guide: http://technet.microsoft.com/en-us/library/cc732855(WS.10).aspx
Using teamed adapters with network load balancing may cause network problems:
http://support.microsoft.com/kb/278431
(23 páginas)
Manymanuals.com
Manymanuals.de
Manymanuals.fr
Manymanuals.it
Manymanuals.pl
Manymanuals.cz
Manymanuals.es
Manymanuals-pt.com
Comentários a estes Manuais