Who’s tops in server reliability? Ask 750 enterprise users
Imagine what would happen if one of your critical, revenue-generating applications suddenly became unavailable. What if the systems running your core transaction processing services went down or suffered a security breach? Each hour of downtime typically costs between $150,000 and $400,000, and each “average breach” lasting from one to 60 minutes could cost upwards of $200,000 to $700,000 in lost productivity, lost, damaged or destroyed data and remediation to return and restore operations, according to Information Technology Intelligence Consulting (ITIC). And it’s important to note that these figures could conceivably be much higher if the organization is unable to restore or recoup the data, or if customers and business partners bring litigation in the event they were impacted by the breach.
System reliability and security are crucial from the standpoints of both financial performance and brand reputation. That’s why we put so much emphasis on both in our IBM Z and IBM Power Systems. And it’s why we are proud of the results from the ITIC 2017 Global Server Hardware and Server OS Reliability survey of 750 organizations worldwide. The survey includes actual user experiences for the most currently available commercial technology, including the latest IBM Power8 Systems and the IBM z13 mainframe.
Results of this year’s reliability survey
Corporate enterprises surveyed ranked IBM Z mainframe-class servers the most reliable hardware platform available on the market today. In its distributed server category, the survey found that IBM Power Systems achieved the highest levels of reliability when compared among 14 server hardware platforms and 11 server hardware virtualization platforms
In fact, IBM Z mainframes have ranked number one for nine straight years. That recognition demonstrates why many leading banks, insurers, retailers and others choose this platform for running enterprise processes, applications and services with respondents reporting less than one (<1) successful security breach during the last 12 to 24 months – and just 0.96 minutes of unplanned downtime per server, per year. That metric equates to 8 seconds per month.
Key factors behind the ranking
Reliability is built into IBM Z servers throughout the entire stack. It starts with the underlying system components, which undergo amazingly rigorous testing—up to and including simulated earthquakes and extreme hot and cold environments. Advanced automation and self-management heighten availability by optimizing the use of system resources, which are redeployed as needed for operational efficiency.
In addition, IBM Operations Analytics for IBM Z helps prevent or reduce outages by using machine learning to proactively detect unusual system or application behavior in near–real time. The machine-learning algorithm sifts through vast amounts of data, identifying correlations and patterns that would be virtually impossible for a human to see. This tool provides both predictive analytics and trend identification based on past operational data and is designed to help reduce risk and restore service up to 60 percent faster through advanced IT analytics.
As we look forward, we focus on continuous improvement with every new generation of the IBM Z platform and we expect more of the same market leadership with the recently announced IBM z14 system and with future generations. To learn more about what you can expect with the new IBM Z, click here.
With mainstream Linux server platforms, IBM Power Systems have the least amount of unplanned downtime at 2.5 minutes per server, per year, according to the ITIC 2017 reliability survey. IBM hardware and Linux operating system distributions were either first or second in every reliability category, including virtualization and security.
From redundant connections to sophisticated firmware, built-in high availability is the reason why IBM Power Systems servers placed first among 14 mainstream hardware platforms. Power Systems are designed and optimized for demanding analytics workloads, with key Linux-class reliability, availability and serviceability (RAS) features. For example, the Open Power Abstraction Layer (OPAL) firmware provides a hypervisor and operating system–independent layer that exploits robust error detection and self-healing functions built into the IBM POWER8 processor.
As a key usability advantage, many of the high-availability features in IBM Power Systems require very little handling or customization. In many cases, line-of-business users are not even aware that the availability of their application is protected by built-in RAS features.
The optimal continuity plan
Today, the consequences of downtime and security breaches aren’t restricted to loss of business opportunities, productivity and income. Many affected customers are quick to vent on social media, where news of your data breach, downed server and/or interrupted service can become public knowledge quickly. Having a plan to remediate downtime and security issues is important. Disaster recovery and data protection are essential. And we think a good business plan starts with the most reliable servers on the market.
Learn how to combat cyber threats from infrastructure to endpoint.
The post Who’s tops in server reliability? Ask 750 enterprise users appeared first on IBM Systems Blog: In the Making.