Browse Prior Art Database

Shared diagnostic LEDs for a chassis that contains multiple independent compute nodes

IP.com Disclosure Number: IPCOM000166857D
Original Publication Date: 2008-Jan-25
Included in the Prior Art Database: 2008-Jan-25
Document File: 3 page(s) / 29K

Publishing Venue

IBM

Abstract

In a chassis that contains 2 or more independent compute notes, each compute node normally requires a set of diagnostic LEDs to speed fault identification. This means that each node incurs the cost of these LEDs and the LEDs occupy valuable real estate on the front of the system. Described is a single operator panel on the chassis that allows the user to know which node has the fault and the nature of the fault, while limiting the actual number of diagnostic LEDS to that which would be required for a single node. The example used below shows only two nodes, but there could be more. It also only shows 5 diagnostic LEDs, but there can be many more. Some servers have supported up to 18 diagnostic LEDs per node.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 54% of the total text.

Page 1 of 3

Shared diagnostic LEDs for a chassis that contains multiple independent compute nodes

Concept: An operator panel that will share diagnostic LEDs across one or more compute nodes located in a single chassis. It is assumed in this example that the planars are located horizontally one above the other. Were they vertically oriented the operator panel would be appropriately laid out to reflect this.
.

Green LEDs indicate an operating state (on solid) or standby state (blinking).

Yellow LEDs indicate faults or states that requires human intervention.

The Green Power LED is indicates that power is applied to a node. There is one per node. In the example shown, the LED is combined with a power button, however the control and display functions may be separated as long as they are clearly associated one with the other.

The yellow fault LED, shown in the square with the exclamation point icon, is lit when a fault has been detected in a compute node. There will be one such Fault LED per compute node for however many compute notes there are co-located in a given chassis.

The diagnostic LEDs may indicate such things Over Temperature, DIMM failure, Fan Failure, Power Supply Failure. Each one have an appropriate icon associated with it. The number of diagnostic LEDs will vary with the complexity of the compute nodes.

Operator panel behavior for non-catastrophic faults (node still operating)

Figure 1 illustrates one example of layout of control panel (all LEDs lit).

Figure 1

Figure 2 illustrates the appearance when both nodes are working correctly .

Figure 2

Figure 3 illustrates the appearance when the top node has deve...