Exchange Server Processor Capacity Planning
Capacity planning mailbox servers, there is a significant change from previous versions of Exchange because Exchange mailbox features in Microsoft Exchange Server 2010 2010 has been reengineered with the concept of a mailbox is changing architecture in order to prevent automatic operation has been in the level of single mailbox database level instead of at the server level. There are two main changes that affect the planning process for the mailbox server role:
- Hosting active and passive database copies on the same server
- Providing database copy count
You can use the information in this topic to learn about these changes and design guidelines when sizing Mailbox servers are configured for mailboxes.
Hosting Active and Passive Database Copies on the Same Server:
Exchange 2010 mailbox resiliency can be configured when the server hosting the active copy of the database. And alias, both on the same server The workload for each processor in a server service is active (active mailbox. Base mounted on) is the mailbox alias (alias database hosting) Perhaps one of the capacity planning mail Exchange 2010, you must consider the requirements for passive mailboxes and databases. A passive database copy uses CPU resources to investigate or verify replication save replay into the database and keep a copy of the content index that is associated with the database. Under normal circumstances, each passive mailbox (hosted on a passive database copy) will be equal to the mail host (host on a copy of a database that is active). Required for CPU usage, 15%
Key points of the plan in order to determine the number of mailbox database copy that you want to start on the basis of flexible time setting for each Exchange 2010 mailbox servers, there are a variety of designs that you can choose from, but we recommend the following versions:
- Design for all database copies activated In this model, you design your server to handle 100 percent of hosted database copies becoming active.
- Design for targeted failure scenarios In this model, you design your server to handle the active mailbox load during the worst failure case.
Design Steps:
Because the scaled new factor. Additional steps are required to configure the server of the mailbox for the mailbox. The General steps are as follows:
- Taking into account the overall solution architecture, high availability requirements. Consider separate solutions, server or database availability Group (DAG) to handle common failure case number, the resilience of the mailbox or the mailbox number of sites required to recover the copies of the database.
- If you use a mailbox resiliency, choose which database to start modeling to the design. (Design for targeted failure scenario or design for all copies of a database enabled).
- The following table lists the estimated CPU and memory, based on the design. Mailbox CPU requirements of liabilities and get additional copies of a mailbox database enabled processor required taking into account the active mailbox CPU and memory requirements. Select boot models to define mailboxes, designed to accommodate the maximum number.
Categories:
0 comments:
Post a Comment