|
Article Directory :: Computers & Technology Articles
To ensure high availability servers should include redundant hardware components with the dual power supplies , dual network connections and mirrored system disks typically used in enterprise environments. Servers should also have multiple connections to alternate storage devices through fibre channel switches and a minimum of two independent connections to the SAN. In most cases servers should feature dual active or hot stand by configuration with automatic failover capabilities.
The next single point of failure to consider after the server is the path between the server and the storage. Potential points of failure on this path might include HBA failures , cable issues , fabric issues or storage connection problems. The HBA is the fibre channel interconnect between the server and the SAN. Using a dual redundant HBA configuration helps ensure that a path is always available. In addition to providing redundancy this configuration might enable overall higher performance due to the additional SAN connectivity.
To achieve fault tolerance , multiple paths are connected to alternate locations within the SAN or even to a completely redundant SAN. Server based software for path failover enables the use of multiple HBA's and typically allows a dual active configuration that can divide workload between multiple HBA thereby improving performance. The software monitors the health of available storage , servers and physical paths and automatically reroutes data traffic to an alternate path if a failure occurs.
In the event of an HBA failure host server software detects that the data path is no longer available and transfers the failed HBA workload to an active one. The remaining HBA then assumes the workload until the failed HBA is repaired or replaced. After identifying failed path or failed over storage devices and resolving the problem , the software automatically initiates failback and restores the dual path without impacting applications. If desired the administrator can manually perform the failback to verify the process.
The sofware that performs this failover is typically provided by system vendors , storage vendors or value added software developers. Software solutions such as Compaq , Secure Path , Veritas , EMC Powerpath ensure that data traffic can continue despite a path failure. These types of software products effectively remove connections , components and devices as single points of failure in the SAN to improve availability of enterprise applications.
To help eliminate unnecessary failover the software distinguishes between actual failures and other network events that might appear to be failures. By recognizing false failures the software can help prevent unnecessary failover/failback effects caused by marginal or intermittent conditions. After detecting an actual failure the software typically waits to determine whether the events is an actual failure.
Eddison Sherry had been working in Linux and other Unix flavours for long years. He had been writing blogs and article on the Linux Storage , Linux Commands , Linux Server Administration , Apache Configuration , NAS and SAN etc. You can learn to use Linux and its usage at Linux Technical Blog.
 EasyPublish™ this article - publishers click here
More articles by Eddison Sherry
|
Free Report!
Ten Essential Secrets Of Article Marketing ... Grab Your Free
Copy Now:
Need Content?
Regular Top Quality Content for your Blog, Ezine or Website ...
Delivered Direct,
For Free!
Click For Details
Arts & Entertainment
Automotive
Business - General
Computers & Technology
Finance & Investment
Food & Drink
Health & Fitness
Home & Family
Internet Marketing/Online Business
Legal
Pets & Animals
Politics & Government
Reference & Education
Religion & Faith
Self-Improvement/Motivation
Social
Sports & Recreation
Travel & Leisure
Writing & Speaking
|
