• 04/13/2007
    4:00 AM
  • Network Computing
  • News
  • Connect Directly
  • Rating: 
    0 votes
    Vote up!
    Vote down!

Analysis: Out-of-Band Management

Resolving server problems over remote access is tough if the network is unavailable. A true out-of-band system can improve network uptime. We examine several setups.


It never fails. After a 50-hour week you finally settle down for a quiet Saturday with the spouse and kids--a good movie and a big bowl of popcorn--and your beeper goes off. The system is completely down at a branch office 100 miles away, a terminal session just won't work, and there's much wailing and gnashing of teeth among the employees trying to get work done. Unfortunately, even the best in-band remote-management tools live and die with the status of your network, and if it's unavailable, the only option is to saddle up your pony and ride.

Vendors are pushing system-management processors as a way of giving you complete remote control of the server. They're even working up standards like SMASH/IPMI (Systems Management Architecture for Server Hardware/Intelligent Platform Management Interface) to further that mission. But these are only partial solutions because they're ineffective if the main network is accessible. True OBM (out-of-band management) systems can reduce network downtime--and save you travel time--by providing some form of remote access to key systems regardless of the network's status.

OBM approaches include KVM-IP systems, SCSs (serial console servers) and intelligent remote power management. Although each approach has pros and cons, the first obstacle is usually obtaining budget. With the number of in-band remote-management platforms available, it may be hard to justify the added cost and complexity of a separate OBM system. Most system admins have been using RDP (Remote Desktop Protocol) and other in-band management tools for years, and these have worked well for the majority of remote problems, except in the rare instances when network connectivity is lost. The system redundancy being built into server and network hardware and the improved stability of OS software are making catastrophic failures far less common, thereby reducing the need for redundant OBM systems. But for those sites with stringent uptime requirements, options are wide-ranging.

Continue Reading This Story...

Click image to view image
bulletAnalysis: Out-of-Band Management
Get the full PDF of this article at NWC Reports.

bulletData Center Power Issues
We provide practical, hands-on guidance on how to design a modular data center that will future-proof your investment. Bonus: Includes free PDF of NWC's popular "This Old Data Center" feature.

It's All Inside

We welcome your comments on this topic on our social media channels, or [contact us directly] with questions about the site.

Log in or Register to post comments