Picture for a moment the emergency ward in a hospital. When it’s a matter of life or death, medics rush to patient’s bed, taking with them the crucial equipment for resuscitation on a cart with wheels. A data center crash cart serves a similar purpose, except that the “patient” is now a server or other IT resource that is barely alive and for which remote access is no longer possible.

Accordingly, the equipment on a data center crash cart is what the IT engineer needs to connect locally to the malfunctioning device. The bare minimum is usually a keyboard, monitor for visualization, and a mouse – the so-called KVM combination. The cart serves as a support for these items so that the engineer can work with enough comfort to be efficient and effective in bringing the resource back from the brink of (digital) death.

Revival When You Can’t Go Remote

The immediate goal of a crash cart intervention is typically to recover the server, switch, or another device to the point where remote access is again possible. Remote access is preferable for several reasons, including rapidity and ease of access from a central control console. In many cases, servers in today’s racks offer no local consoles of their own, especially in higher density configurations. The IT engineer physically next to the device will need to connect the necessary cable(s) to get into the device, a further challenge in crash cart situations.

In a colocation center, crash cart services for servers may be offered as an option to customers. The advantage for the customers is that if remote access fails at the server level, they will not have to drive or fly out to the colocation center to fix the problem in person.

“Going Large” on Your Data Center Crash Cart

Crash carts may also have additional features and functions:

• Holding a removable hard disk for booting the malfunctioning device from an externally held operating system and recovery software
• Carrying other backup media, such as tape cartridges, if these are required by the system concerned
• Work surface for using manuals and binders, with storage space for these documents elsewhere on the cart
• A lockable storage area for holding more valuable equipment than the simple KVM combination, for example, a notebook computer
• A seat for the IT engineer to sit on

Construction quality of a crash cart is important too. The cart should be ergonomically designed to facilitate the work of the IT engineer, offer a stable support for the different items used on it, and run smoothly on wheels or casters, for instance, for easily handling and positioning.

The Crash Tray as an Alternative

Mobile (wheeled) crash carts are not the only possibility for dealing with loss of remote access, however. KVM trays are available from some vendors. Such a tray fits into one U of a rack, and can be set up with physical connections to more than one server or switch. The advantages for IT engineers and data centers or colocation centers and their customers are that less or no additional equipment is required to be brought in, and no time is lost in trying to find out which connecting cable should go where.

Which do you favor in a data center, a crash cart on wheels or “crash trays” integrated into racks? Give us your opinion with a few words below.

If you're in the data center, mission critical or cloud services industries, or you sell to the data center industry, don't miss our weekly update newsletter -- Data Center Sales & Marketing Institute (DCSMI) Update Newsletter. Get notified about new reports, events, podcasts, and blog posts.