When it comes down to selecting the x86 server platform for your applications and services, you first think about reliability – you need highly reliable system for your tasks and processes to run 24 hours per day, 7 days a week (24x7), without any downtime. And the question you might ask is: What are the most reliable x86 servers to fulfill your needs?
While there are many different ways to estimate server reliability and gather the required data to evaluate different offerings, one of the approaches is to look at the industry analyst reports that are based on client surveys. Who else other than your peers – clients who actually use the hardware (and, perhaps, software) – can give you a better perspective on this topic!?
Recently, I’ve found an interesting report from Information Technology Intelligence Consulting Corp. (ITIC) titled “ITIC 2014 - 2015 Global Server Hardware, Server OS Reliability Report.” (The report can be found here: http://public.dhe.ibm.com/common/ssi/ecm/en/xsl03126usen/XSL03126USEN.PDF). This report contains the results of the latest independent global survey about server reliability conducted across IT managers and C-level executives from around the world during March-April 2014.
I encourage you to read the full report to get the details of the methodology used and reliability results referred before making a purchasing decision. In the meanwhile, I want to share few key things from this report.
First, and the most important, IBM x86 servers (IBM System x) were recognized as the most reliable x86 servers in the world by the clients, according to this report. This is due to the fact that only 5% of the IBM x86 servers experienced downtime of more than 40 minutes but fewer than 4 hours in the past year. (For other vendors mentioned in this report, the percentage is double-digit). Also, only 1% of System x servers experienced more than 4 hours downtime in the past year.
Second, the reliability numbers achieved by the System x servers are close to the numbers shown by higher-end IBM platforms such as IBM System z (mainframes). IBM X-Architecture principles that are used to build IBM System x servers pay off. The ultimate design goal for IBM System x servers as a part of the IBM X-Architecture strategy is to prevent hardware faults from causing an outage. Part selection for reliability, redundancy, predictive failure alerts, recovery and self-healing techniques are used in a fault resilient IBM Reliability, Availability, and Serviceability (RAS) strategy to avoid application outages.
Ilya Krutov is an IBM Redbooks Project Leader. He writes books and papers on various topics related to IBM System x, IBM BladeCenter, and IBM Flex System. Follow Ilya on Twitter at @IlyaAtRedbooks.