Jump to content United States-English
HP.com Home Products and Services Support and Drivers Solutions How to Buy
» Contact HP

HP.com home

Data center automation

» 

HP Labs

» Research
» News and events
» Technical reports
» About HP Labs
» Careers @ HP Labs
» Worldwide sites
» Downloads
Content starts here
data center servers
 

Research opportunities

Key to the success of today's enterprises is IT's ability to respond quickly and efficiently to the diverse and often changing needs of a distributed, global, and heterogeneous infrastructure.

The problem is that current IT processes -- from installing a patch across an entire data center to provisioning resources to managing the thousands of desktops in an enterprise-- are often manual, time-consuming, tedious and prone to human-injected errors.

 

Our approach

Automating IT can increase agility, improve performance, mitigate risk and maximize return. Rather than developing IT systems in an ad-hoc manner as is done today, we believe that IT should operate more like the automatic transmission in a car or the autopilot feature of an airplane -- that is, as a closed-loop, self-managed system.

We are taking a structured, holistic approach to automation, using models, high-level policies and service-level objectives to manage the life cycle of systems and their applications.

Research focus

HP Labs' vision is to automate systems so that users can determine what they want and then have the system determine how to accomplish it.

We believe automation has the power to transform traditional IT, allowing organizations to do more with less and to shift the focus of IT staff from maintenance to innovation.

Current work

We are investigating how to use automation to design, deploy and manage a service, including automating such tasks as determining what application components are needed, choosing where components will reside, provisioning each component and making sure that the service meets sevice-level objectives.

In addition, researchers are applying machine learning and statistical learning theory to aid with diagnosis and forecasting in computing systems. Machine learning techniques analyze the vast amount of data collected on systems to produce answers that facilitate faster diagnosis and predictions of application performance problems.

As part of our research, HP Labs has explored the use of automation in several real-world instances:

  • Automation applied to HP-IT's Shared Application Server Utility (SASU) significantly reduced the time and manual labor involved in resource-intensive monthly planning tasks.
  • The configuration and deployment of software and resources accessed, on-demand by animators involved in the SE3D animation showcase demonstrated the critical role of automation in setting the stage for an agile, adaptive, utility-based infrastructure.

We are also exploring how we can use automation to develop a more cost-effective, scalable remote desktop solution.

Future applications

We see automation as a fundamental piece of an adaptive infrastructure and a critical step on the path to 24X7, lights-out computing and the next-generation data center.

Enterprise computing

       
» Data center automation
  » IT resource virtualization  
  » Data center power & cooling  
  » Storage infrastructure & management  
  » IT asset tracking  
  » Advanced architecture  
       
 
 

Related research

»  Service Level agreement-based management
»  SLA-based Management
»  Statistical learning interface & control
»  Systems architecture
 

Learn more

»  Feature story: Intelligent IT
Printable version
Privacy statement Using this site means you accept its terms Feedback to HP Labs
© 2009 Hewlett-Packard Development Company, L.P.