GB2355824A

GB2355824A - Monitoring burn racks

Info

Publication number: GB2355824A
Application number: GB0028977A
Authority: GB
Inventors: Subha R Ram; Roger W Wong; Richard D Amberg
Original assignee: Dell USA LP
Current assignee: Dell USA LP
Priority date: 1998-10-22
Filing date: 1999-09-22
Publication date: 2001-05-02
Anticipated expiration: 2019-09-22
Also published as: GB0028977D0; GB2368944A; GB2368944B; GB2355824B; GB0129641D0

Abstract

A test device including a plurality of burn racks 10 each including a plurality of workcells. A monitor R is connected to receive and display data from each workcell, in particular a visual occupancy representation, and is also connected to a burn rack database DB. Remote monitor R2 is provided to select and access data from any of the monitors R relating to systems in the workcells.

Description

2355824 TROUBLESHOOTING COMPUTER SYSTEMS DURING MANUFACTURING USING STATE

AND ATTRIBUTE INFORMATION This invention is related generally to building computer systems and more particularly to the preparation of build-to-order computer systems.

This application relates to co-pending British Patent Applications 9816129.2, 9816126.8 and 9918882.3.

2 Many methods have been devised for tracking inventory. In U.S. Patent 5,434,775, the locations of a plurality of devices are tracked using a network of communication links, each of which corresponds to a location. Each device is given a tag that identifies the device with respect to other devices and that is connectable to a communication link when the device is disposed at the location to which the link corresponds. Each tag that is connected to each commun cation link is detected, and the location of each device is deter-mined based oil the detection. One feature of the technique is additionally determining the conditions of the devices by correlating one or more communication linkswith conditions. The technique is simple to use and a highly effective technique in tracldng devices stored at various locations throughout a facility. Device location and condition are monitored continuously, thereby reducing the risk that the removal of a device from storage win go undetected.

A present trend among some computer manufactuTers is to provide a customer with a custom-built computer system in which the customer has designated that certain components and capabilities be included in the system being ordered- It is therefore important to mx6dmize efficiency at every step of the build-to- order process.

That efficiency begins at the time the order is placed and processed, and ontinues throughout the assembly, testing and shipment of the custom-built unit.

During production of build-to-order computer systems, specific parts for a computer are pulled from stock and taken to an assembly pod where those specific parts are assembled in the computer chassis. Following assembly, the chassis is moved to a qxrick-test area where tests are conducted to quickly determine whether the correct parts for that order are installed, and whether the parts are operative.

Following the quick test procedure, assembled chassis are moved to a burn rack where the parts are 4burned in7 and where operational errors may be detected.

Mahy units are simultaneously tested on the burn racks zind the tests may take a 3 couple of hours to complete. With many units in production waiting to be tested, it is important that the burn rack spaces available for testing are used efficiently.

Therefore, it is important that the computers or devices under test (DUT) are tested in a manner which quickly and efficiently determines whether a DUT is satisfactorily operational and if not, which quickly and efficiently identifies operational deficiencies so that the DUT may be removed from the burn rack to free up the occupied burn rack space for another DUT to be tested. DUTs are loaded on the burn racks by manual verification of empty workcells. An operator is required to walk up and down rows between burn racks to visually verify empty workcells. This practice is both labor intensive and time consuming. Also, historic infonnation of burn rack utilization is not available.

When a DUT is on the burn rack, the software ordered with the system is also downloaded to the DUT from a server. Personnel monitor the burn rack test units for visual and audible indications, i.e. LED's and beeps, of how the testing and downloading procedures are progressing. A red LED indication accompanied by an audible beep indicates a failed DUT which is returned to quick test where it is thoroughly tested by a technician. A green LED indication means that a unit is ready to be moved on to a final test to check the screen and the operating system prior to shipping the unit.

When software downloading is to be accomplished, the DUT is identified to the server for download of the appropriate software. Each DUT is identified by a lifetime identifier (serial number) in the form of a'bar code. When the DUT is on the burn rack, its physical location is also identified by a rack, a column on the rack aad a row In the column. Each burn rack location is serviced by a location specific cable and a networkdevice connector which interconnects the cable to the DUT. :However, although the cable can only service a specific rack location, the network device can and does sometimes become detached from one cable and attached to another. Each network device has a MAC address which is mapped to a location in terms of the rack, the column and the row. The mapping information is stored in a database in the network environment. The DUT can communicate with the database. As a result, the exact location of the DUT can be determined. Therefore, if the connector is moved to a different rack location and is connected to another cable, the information in the database will be inconsistent with the exact location of the DUT.

During the manufacturing process, problems with components do sometimes occur. For example, an incorrect component may have been installed, and may need to be replaced Also, an installed component may not pass the test phase and may -need to be replaced. Preferably such events are corrected during manufacture. If not, a costly recall may be generated.

Failure of a computer system under test requires identification of the system and identification of the syste&s bum-rack location. A recent development provides an automatic means for determining the location of a DUT by-mapping a DUT, connected to a simple network management protocol (SNTVM) enabled network to a physical location. As a result, a device and method are provided to track a DUT during the manufacturing process. The burn rack includes several work cells- An SNMP switch -device is provided adjacent the burn rack. The switch device includes several ports. Cables are provided such that a respective cable interconnects a respective port of the switch device and a respective work cell. A monitor is -provided adjacent the bum rack and is connected to a port of the switch device.

Therefore, what is needed is an enhanced burn rack monitor system which enables each systems information to be stored in a centralized database, and pr:ovides the ability to track the physical burn rack location of each system and indicates 'each system!s state and attribute informatiori.

In our British Patent Application GB 2346463, from which the present case is divided, We-disclose and claim a test device comprising a burn rack including a plurality of work cells, a simple network management protocol enabled network device adjacent the burn rack, the network device including a plurality of ports, a plurality of cables for interconnecting a respective port of the network device and a respective workcell, a monitor connected to the network device, and a display provided on a screen of the monitor to display a visual occupancy representation of each workcell.

According to the present invention, a bum rack test device comprises:

a plurality of test devices in accordance with any one of the preceding claims, each test device including: a burn rack database connected to each monitor, the burn rack test device further comprising:

a remote monitor coupled for selecting and accessing data from any of the burn rack monitors relating to systems in the workcells.

A principal advantage of these embodiments is that this information is accessible in real-time to manufacturing. Along with the bum rack monitors ability to track the physical location of each system being built in' manufacturing, a location map is provided to operators to permit them to pinpoint the location of a specific system for special handling and/or troubleshooting, determine the test phase (state), determine an incorrect component or component failure, (attribute), determine total work cell space available, and access each work cell in an entire group of bum racks.

Preferred examples of the present invention will be described in accordance with the accompanying drawings, in which:

6 Fig. 1 is a diagrammatic view illustrating a burn rack test device; Fig. 2 is a pictorial replication illustrating a burn rack monitor display; Fig. 2A is a diagrammatic view illustrating a bank of burn racks; Fig. 3 is a pictorial replication illustrating another burn rack monitor display; and, Figs. 4-8 are pictorial replications illustrating further embodiments of displays providing burn rack data and information.

Fig. 1 shows a burn rack 10 divided into workcells 12 arranged in columns 14 and rows 16. An SNMP enabled network switch device 18 is adjacent rack 10 and has several ports P connected by cables C to computer systems, or DUTs, designated M in each workcell 12. Another port P connects network device 18 to monitor R which is in turn connected to a database DB.

A burn rack monitor display 20, Fig. 2, on monitor R provides a graphical user interface indication of the DUTs in each workcell 12 for the associated rack. That is, display 20 includes a visual occupancy representation of each workcell 12 so as to indicate whether a workcell 12 is empty or is occupied by a DUT. As it can be seen, the display 20 indicates workcells, on an A side of the rack 10 and the workcells on a B side of the rack 10. Each of the A and B sides includes a plurality of horizontal rows 1-4 and vertical columns 1-6. A system identifier is displayed in each workcell position 12 which contains a DUT. For example, side A, row 1, column 4, contains a DUT having the system identifier CT 314, and side B, row 3, column 1, contains a DUT having the system identifier CT5WY. Thus, each workcell 12 containing a DUT is visible on the display 20, and each DUT can be readily identified by its specific identifier displayed on a background having a specific shape, for example, a rectangular shape. Workcells 12 which display no 7 system identifier provide an indication that that specific workcell 12 is unoccupied, e.g. side A, row 1, column 2, and side B, row 3, column 2 are empty workcell locations.

Fig. 2A shows an example of a test device in accordance with the present invention. In Fig. 2A, a bank of burn racks 10 each have a dedicated monitor R connected to the database DB. Thus, the display 20 such as that illustrated in Fig. 1 may only be viewed for a specific burn rack 10 to which the monitor R is connected. A remote monitor R2, however, may be used by for example, a supervisor, at a remote location to select and access any of the dedicated monitors R and thus monitor any system M at any workcell 12.

An enhanced version of display 20 is illustrated in Fig. 3. Note that one of the workcells 12 on side A, row 3, column 2 indicates a system identifier CSLPO. However, the system identifier CSLPO appears on an oval-shaped background rather than a rectangular background as indicated at each other workcell 12. This enhancement, or the like, may be used to identify each system which is part of a special order or a mass order which is being processed, so as to provide a quick visual reference to locate or identify a specific system. Also, note that a line of information 22 is provided below the side-column row information. This information is provided by merely using a mouse to point to a system identifier in a specific workcell 12 on display 20.1 The information revealed includes, for example, a bar code identifier, workcell location information, and the current test running at the time of inquiry for a specific DUT. In addifion to background shape differences, the background may be colour coded to indicate an in test mode, a test failure mode, a test completion mode, etc.

Additional information can be obtained by using a mouse to point and click on a selected system identifier in a specific workcell 12 on display 20.

This will result in a system status display 24, Fig. 4, which includes detailed 8 activity on the system selected. The detailed activity includes information about the bar code, the current test running, and the location. In addition, a point and click on a "More" button 26, on status display 24, provides a detailed transaction history display 28, Fig. 5, which may be viewed for tracking all the 5 tests performed on the selected system.

Referring again to Figs. 2 and 3, a point and click on a "Utilities" main menu display 30 provides options added to the main menu. This selection provides options illustrated in Fig. 6. A point and click on an option designated "Switch Rack" 32 provides the flexibility to view any rack 10, Fig.

2A, from the remote monitor R2. This enables the supervisory user with a. list of several burn racks as illustrated in Fig. 7. As a result, the supervisory user can point and click to a specific burn rack and view, from the remote monitor R2, any burn rack monitor display, such as that illustrated in Figs. 1 and 2.

Furthermore, for a given bar code information about a location, a point and click on an option designated "Query by Bar code" 34, Fig. 6, provides information, illustrated at Fig. 8, regarding location, row, column and side. This feature is useful to find out the last known location of the system which is tracked throughout a manufacturing facility by bar code references.

In operation, these embodiments provide a means to determine vacant workcell locations in the burn rack area. A monitor display is provided in front of each burn rack indicting the workcells which are either occupied or empty. This provides a signal to enable loading devices to be tested in an empty location. The information provided may be used to determine times of peak and minimal usage of the workcells. In the case of resource intensive devices to be tested on the burn rack, space usage information is used to arrange the devices efficiently for balanced load sharing purposes. Capacity planning for burn racks is accomplished by using the space usage information.

9 The monitor displays in front of each burn rack provides an indication of the state of the DUTs in the workcells- The workcells which are unoccupied are indicated in the form of blank rectangles having no system identifier displayed therein. After a DUT undergoes all of the diagnostics, the display 5 indicates that the DUT is ready to be moved out of the display area. Automatic end of test detection can be made, so that the display shows an available spot and a new DUT can be loaded on the burn rack. Because the information is available for the entire diagnostic area, efficient routing decisions can be made. This results in optimum usage of the burn racks saving waiting times outside the burn racks during peak usage periods.

As it can be seen, the principal advantages of the invention are that a specific DUT can be identified and located, burn rack space available can be determined, state and attribute information regarding a DUT can be readily determined, and it is possible to access an entire burn rack system status.

Information is accessible in real-time to manufacturing. Along with the burn rack monitor's ability to track the physical location of each system being built in manufacturing, a location map is provided to operators to permit them to pinpoint the location of a specific system for special handling and/or troubleshooting, determine the test phase (state), determine an incorrect component or component failure (attribute), determine total work cell space 1 availability, and access each work cell in an entire group of burn racks. Efficient factory space management is achieved by placement of a DUT based on its physical attributes and the current distribution of other DUTs in the burn racks based on predefined criteria for each factory.

Claims

1. A burn rack test device comprising:

a plurality of test devices in accordance with any one of the preceding claims, each test device including a burn rack database connected to each monitor, the bum rack test device further comprising a remote monitor coupled for selecting and accessing data from any of the burn rack monitors relating to systems in the workcells.

2. The device as defined in Claim 1, wherein each.monitor includes a screen having a display thereon, the display including a visual occupancy representation of each workcell.

3. The device as defined in Claim 1 or Claim 2, wherein the display includes a main menu providing for a selection of options for viewing a selected bum rack.

4. The device as defined in any one of the preceding claims, wherein the display includes a main menu provided for a selection of options for viewing a last known location of a device under test.