»  Search

Large, scale-out server environments are quickly becoming commonplace. Yesterday’s monolithic, costly server farms are now being replaced by vast arrays of cost-efficient servers – often blades – running open source operating systems. The reason is simple: putting server blade technologies together with Linux® software provides considerable computing bang for the buck.

However, it also presents some notable challenges.

“Scaling and managing a Linux-based blade environment can be a thorny undertaking,” says Michael Schulz, Product Marketing Manager for Linux on Blades at HP. “Without some help, IT organisations may face a labour-intensive and complex road ahead.”

Deploying new server blades and operating system images can be challenging, he explains. The process requires several people and numerous manual processes that can be time consuming and error prone. And once this kind of environment has been implemented, managing it can prove to be difficult as well, since older management tools were not designed with blades in mind.

“Legacy infrastructure management tools are geared toward static, dedicated servers,” notes Schulz. “However, blades represent a different paradigm; they are more flexible and virtualised. Organisations need management tools that reflect this new paradigm and graphically map the environment, so they can easily make adjustments and maximise blade resources.”

Schulz also says that it may be challenging to integrate several tools to work together. The result: provisioning, performing updates and making changes can be arduous and complex. “It’s fine if an organisation uses one Linux utility to deploy images and another utility to monitor the environment,” says Schulz. “But if those utilities aren’t well integrated, how does the monitoring utility know when new images have been added?” This scenario is all too common, Schulz says, and makes the IT manager’s job more complicated and time consuming than it has to be.

Schulz recommends that companies facing these and other obstacles can turn to the HP BladeSystem when deploying Linux in a scale-out environment. “It provides a fully integrated set of offerings, from hardware to operating system to support and services,” he explains. “And we offer Linux-specific software for managing blade-based environments.”

Schulz also points to HP Insight Control Linux Edition, an integrated software package for deploying and managing HP blade environments from a Linux-based console. The software provides extensive provisioning, asset tracking, data centre monitoring and failure prevention capabilities. The solution is built entirely on Linux, but since most organisations don’t run the open source software exclusively, Insight Control Linux Edition supports both Linux and Windows® server blades.

Developed using the previous HP Control Tower product, which was designed for Linux and focused specifically on the HP BladeSystem, Insight Control Linux Edition provides features and functions designed to match the way Linux users like to deploy and manage their servers. It uses a safe and isolated management network for discovery, monitoring and remote administration.

“Linux on HP BladeSystem combines the advantages of an open source platform with advanced server technology to reduce the overall cost of IT operations while delivering an open and adaptive infrastructure,” says Schulz. “And with HP Insight Control Linux Edition, organisations now have a comprehensive tool for easily scaling, monitoring and controlling their multiplatform blade environment.”

HP BladeSystem c-Class
HP Insight Control Linux Edition
Free book available in limited quantities – “Hot Air Rises and Heat Sinks: Everything You Know about Cooling Electronics is Wrong”
Transforming Your Enterprise Magazine – HP BladeSystem Special Edition

Microsoft and Windows are U. S. registered trademarks of Microsoft Corporation. Windows Vista is either a registered trademark or trademark of Microsoft Corporation in the United States and/or other countries. Linux is a U.S. registered trademark of Linus Torvalds.