when relevant content is
added and updated.
Linux has made impressive inroads into enterprise-class computing environments because of its performance, security, scalability and global open source support network. As a result, more and more enterprise-class workloads—such as analytics, databases and transaction processing—are moving to Linux.
Now, dramatic increases in data volume and velocity are being driven by such trends as big data, pervasive enterprise mobility and the Internet of Things. This surge in data for lightning-fast transactions and transformative business insights has put big pressure on data center infrastructure to handle the load. The question is: What’s the best approach to meet those needs?
Clustering of x86 servers is one option that inevitably comes up. After all, x86 servers come with relatively low price tags, are easy to install and support a wide range of applications. For specific environments, like fast-growing small businesses or departmental applications, clustered x86 servers may be a good way to go.
But for enterprise-class, mission-critical workloads that demand high-end processing power, throughput, infinitely scalable storage, fast network connectivity and rock-solid security, x86 clusters are likely to come up short. There are several reasons why:
- Economics: Despite a typically low initial capital expense, x86 clusters actually are likely to cost more money as those clusters expand and proliferate to meet increased need for more storage, processing power and connectivity. Adding more “oomph” to handle enterprise workloads means adding more servers, as well as more storage in the form of NAS appliances and expensive storage-area networks. Software licenses and support fees also are going to expand in lockstep with increased cluster size.
- Enterprise workloads are built around applications that require a lot of memory (which often means expanding the x86 cluster farms). Also, not all enterprise applications lend themselves to cluster-based architectures—something IT executives and their business users can’t afford to find out after the fact.
- Management and monitoring. More boxes mean more management complexity, especially if those clusters are built on different server brands and operating system versions. That makes automation more challenging, which in turns puts a bigger burden on IT staffs to monitor system operations and application performance.
- For enterprise workloads, organizations want to limit the potential points of entry for hackers and other threat vectors. More servers and more clusters are going to increase the threat vulnerabilities, not reduce them.
Fortunately, there’s a better option: A mainframe-class Linux server with vastly improved economics. IBM’s LinuxONE server is optimized for enterprise-class Linux workloads that demand performance, scalability, security and manageability, as well as an attractive total cost of ownership model.
One major area where the IBM has focused LinuxONE is disaster recovery—a critical requirement for enterprise workloads that cannot afford downtime, or even short interruptions in application availability. “Because LinuxONE systems utilize a shared-everything data architecture, there is no need for multiple copies of files or databases,” according to an IBM-sponsored report issues by IT consultants Robert Frances Group. “This not only eliminates out-of-sync conditions, but also simplifies the set-up and execution of the recovery point objectives and recovery time objectives.”
Three key principles underscore the IBM LinuxONE philosophy:
- Open source delivered on the best environment for your organization—on-premises or in the cloud, be it public, private or hybrid.
- Limitless flexibility and scalability for expanding workloads, combining the best of open source and enterprise computing.
- Risk reduction, utilizing proven hardware and software platforms from a trusted source and heightened security features including blockchain technology.
As the performance, security and cost efficiency of Linux-based solutions becomes increasingly apparent, enterprise workloads are increasingly migrating to Linux. Smart IT and business decision-makers will continue gravitating toward purpose-built Linux solutions, rather than built-on-the-fly x86 clusters, to handle enterprise workloads in modernized data center environments
 “10 reasons LinuxONE is the best choice for Linux workloads,” Robert Frances Group, 2015