Find White Papers
Home About Contact Help
Free Membership Member Login
Search the Library                  Advanced Search

Going Beyond HA for Business-Critical Virtualization

Stratus Technologies
By : Stratus Technologies
INFORMATION
Published : Dec 05, 2007
Length : 7
Type : White Paper
 
Download Now
Save for Later
  Email This Page
Overview :

This paper details how VMware and Stratus together address the full spectrum of planned and unplanned downtime challenges: server hardware failures, operating system failures, server maintenance, operating system maintenance, disaster recovery, support and services. Also presented are scenarios describing how you can use VMware®  Infrastructure 3 and Stratus ftServer® systems at your data center and remote sites.

Download this paper now. 

View All Items By This Company
Browse Related Categories :

High Availability

,

Infrastructure

,

Server Virtualization

 
The combination of VMware® Infrastructure 3 (VI3) and Stratus® Continuous Availability servers provides the most comprehensive solution for IT infrastructure availability in the industry. VMware VI3, encompassing VMware ESX Server and other components, is the most widely deployed software suite for optimizing and managing industry-standard IT environments through virtualization. Stratus Technologies is the company whose fault-tolerant servers set the standard for uptime.
This paper details how VMware and Stratus together address the full spectrum of planned and unplanned downtime challenges: server hardware failures, operating system failures, server maintenance, operating system maintenance, disaster recovery, support and services. Also presented are scenarios describing how you can use VMware VI3 and Stratus ftServer® systems at your data center and remote sites.
Server Hardware Failure
VMware ESX Server with VMware High Availability (HA), which is now included with the VMware Infrastructure 3 (Standard Edition), can provide automatic restart of virtual machines running on a failed conventional server, but users will be faced with downtime of a few minutes, loss of in-process transactions and loss of in-memory application data as well as application state information.
Stratus ftServer systems eliminate downtime associated with server hardware failures and protect in-process transactions as well as in-memory application data. These Continuous Availability, fault-tolerant servers rely on a fully duplicated hardware architecture with core CPU-memory components that run together in lockstep. The systems automatically detect their own hardware failures and remove a malfunctioning component from service while a partner component continues uninterrupted operation. Proven in the field to deliver 99.999% or better uptime, the ftServer architecture uses Intel® Xeon® processors and industry-standard operating systems.
Operating System Failure
VMware ESX Server provides a highly reliable, small-footprint hypervisor that manages physical x86 hardware and I/O devices. Operating systems, such as Microsoft® Windows®, Linux® and Solaris™, run as guests. Each virtualized guest OS instance typically runs a reduced set of applications compared with a non-virtualized server environment. A reduction in application interaction combined with virtualization of hardware and I/O dramatically reduces the complexity of the guest OS environment, resulting in improved operating system availability. In the event of an unpreventable OS crash, VMware ESX Server allows the associated virtual machine to be restarted quickly for fast recovery.
Stratus Continuous Availability technology adds hardened drivers to ESX Server to further reduce the chances of driver-induced failures. Stratus ftServer systems also recover transparently from hardware component failures and mask these failures from the hypervisor and guest operating systems, eliminating another possible cause of operating system failure.
Application Failure
VMware ESX Server allows applications to run in fully isolated virtual machines. This eliminates application conflicts typical of non-virtualized environments. Such conflicts are one of the leading causes of application failure. This is particularly true of Windows environments, which helps explain the predominance of those environments in the initial move to virtualization. Recovery from unpreventable application failures can be performed rapidly by restarting the associated virtual machine on the same server or on a different server, which minimizes application downtime.
Stratus ftServer systems further improve application availability by masking all hardware component errors from the application and by eliminating server failures that would otherwise introduce application downtime.
Server Hardware Maintenance
To allow a server to be taken offline for maintenance, VMware ESX Server allows all virtual machines to be moved from a server to other servers within a VMware cluster. VMware VMotion™ can be used to move virtual machines online to eliminate all application downtime. After you complete maintenance on the target server, virtual machines can be moved back to the server to rebalance load within the cluster.
Stratus ftServer systems support online maintenance for most hardware maintenance operations. This capability reduces the operational time associated with offloading and rebalancing virtual machines for servers that must be taken down for maintenance.
Operating System Maintenance
The most common maintenance operation associated with operating systems is the application of minor updates and patches. The introduction of new service packs, major update releases or new operating system versions is less frequent, but can result in more significant interruptions to your IT infrastructure.
VMware ESX Server virtualization significantly reduces downtime and improves the reliability of operating system maintenance by allowing operating system updates, either major or minor, to be implemented and tested in a non-disruptive virtual machine test environment that accurately duplicates the production environment.
Search the Library                  Advanced Search
About Us Contact Us List Your Papers Partner With Us Site Map