Ask the Infrastructure Specialists: Lesia’s Q & A Corner
Understanding Fault Tolerance
Once your business has the correct application solution, the next step is selecting the correct hardware platform to support that application. In server selection, availability is the most important consideration. For applications where availability is a high priority and retaining data is a must, a fault-tolerant server is the best selection.
A fault-tolerant server is one that is designed to survive any single fault and many cases of multiple faults without any service interruption or loss of data/work in process. A failure should in all cases be completely transparent to the user. Fault-tolerant servers have been measured in the field to have average time between failure that are orders of magnitude – up to a hundred times or more – longer than those experienced by standard high-availability servers. Fault-tolerant servers experience availabilities of more than five 9s. They will be up more than 99.999% of the time which translates into less than five minutes per year of downtime.
These severs are often described as providing continuous availability because they are designed to preempt downtime. This approach has provided comprehensive components and functional redundancy within the footprint of a single server. If one of the server’s components fails, the server continues to process normally. The redundancy ensures the users do not experience any disruption, no transactions or data are lost, and there is no degradation in server performance. Unlike a high availability server which only offers 99.9% (three 9s) of uptime with an average of eight hours of downtime a year with possible costly data loss.
The ftServer line by Stratus Technologies automatically enable RAS (reliability, availability & serviceability) to the fullest extent possible. Replicated components, system software reliability features, and online service features combine to support uninterrupted uptime that has been proven to meet or exceed five 9s. Applications benefit from the servers’ fault-tolerant capabilities from the time you load them on the system without any software application changes or special configurations.
Current fault-tolerant servers eliminate drawbacks associated with older, proprietary systems. By using a standard Microsoft Windows operating system and off-the-shelf hardware components including Intel Processors, Stratus can offer the ftServer line at a very competitive price. Their single-system design permits lower software license costs compared with clusters because a single server does not require multiple copies of the same software and companies can potentially eliminate the need to move to Enterprise Versions of popular programs and databases
The ftServer series keeps the total cost of ownership low as well. They are maintained like ordinary, standalone Windows servers, but with self-monitoring capabilities, remote troubleshooting, and true hot-swap components all built in. These systems do not miss a beat in the face of component failures and system software faults that cause other servers to crash. Their combination of remote-access serviceability and automatic problem reporting allows service engineers to resolve issues online in more than 95% of cases. And when needed, ftServers system can automatically order their own hot-swappable replacements parts, which are typically delivered within 24 hours.
With the understanding of fault-tolerance and needing a system with 24/7 availability, you may think twice about what high availability really means to you and your business.
For more information on Stratus or establishing a fault tolerant server environment, please contact Lesia Bell, your I2 Inside Sales Coordinator at 443.683.8031 or lbell@insourcess.com
To request a copy of "Fault Tolerance for Dummies" please click here.
*Information from Stratus Server Buying Guide
Back to top
Ask the Infrastructure Specialists: Lesia’s Q & A Corner
Understanding Fault Tolerance
Once your business has the correct application solution, the next step is selecting the correct hardware platform to support that application. In server selection, availability is the most important consideration. For applications where availability is a high priority and retaining data is a must, a fault-tolerant server is the best selection.
A fault-tolerant server is one that is designed to survive any single fault and many cases of multiple faults without any service interruption or loss of data/work in process. A failure should in all cases be completely transparent to the user. Fault-tolerant servers have been measured in the field to have average time between failure that are orders of magnitude – up to a hundred times or more – longer than those experienced by standard high-availability servers. Fault-tolerant servers experience availabilities of more than five 9s. They will be up more than 99.999% of the time which translates into less than five minutes per year of downtime.
These severs are often described as providing continuous availability because they are designed to preempt downtime. This approach has provided comprehensive components and functional redundancy within the footprint of a single server. If one of the server’s components fails, the server continues to process normally. The redundancy ensures the users do not experience any disruption, no transactions or data are lost, and there is no degradation in server performance. Unlike a high availability server which only offers 99.9% (three 9s) of uptime with an average of eight hours of downtime a year with possible costly data loss.
The ftServer line by Stratus Technologies automatically enable RAS (reliability, availability & serviceability) to the fullest extent possible. Replicated components, system software reliability features, and online service features combine to support uninterrupted uptime that has been proven to meet or exceed five 9s. Applications benefit from the servers’ fault-tolerant capabilities from the time you load them on the system without any software application changes or special configurations.
Current fault-tolerant servers eliminate drawbacks associated with older, proprietary systems. By using a standard Microsoft Windows operating system and off-the-shelf hardware components including Intel Processors, Stratus can offer the ftServer line at a very competitive price. Their single-system design permits lower software license costs compared with clusters because a single server does not require multiple copies of the same software and companies can potentially eliminate the need to move to Enterprise Versions of popular programs and databases
The ftServer series keeps the total cost of ownership low as well. They are maintained like ordinary, standalone Windows servers, but with self-monitoring capabilities, remote troubleshooting, and true hot-swap components all built in. These systems do not miss a beat in the face of component failures and system software faults that cause other servers to crash. Their combination of remote-access serviceability and automatic problem reporting allows service engineers to resolve issues online in more than 95% of cases. And when needed, ftServers system can automatically order their own hot-swappable replacements parts, which are typically delivered within 24 hours.
With the understanding of fault-tolerance and needing a system with 24/7 availability, you may think twice about what high availability really means to you and your business.
For more information on Stratus or establishing a fault tolerant server environment, please contact Lesia Bell, your I2 Inside Sales Coordinator at 443.683.8031 or lbell@insourcess.com
To request a copy of "Fault Tolerance for Dummies" please click here.
*Information from Stratus Server Buying Guide
Back to top