Data bus arrangement and control method for efficiently compensating for faulty signal lines

[0001] This application claims priority under 35 U.S.C. § 119 to my application entitled “DATA BUS SYSTEM AND METHOD FOR CONTROLLING THE SAME”, filed in the Korean Industrial Property Office on 29 Nov. 2001 and assigned Serial No. 2001-75072, the contents of which are hereby incorporated by reference.

[0002] 1. Field of the Invention

[0003] The present invention relates to a data bus system, and more particularly to a data bus having only a single bus having extra signal lines that are used to compensate for faulty, malfunctioning signal lines.

[0004] 2. Description of the Related Art

[0005] In a data bus system using a single data bus having a plurality of signal lines, when one or more signal lines among the plurality of signal lines malfunction, the data bus ordinarily cannot function properly.

[0006] U.S. Patent Application No. 2002/0099980 A1 to Olarig seeks to overcome such a problem when a 64 bit bus is being used. In Olarig '980, the 64 bit bus is composed of a 32 bit upper bus and a 32 bit lower bus. If there are no malfunctioning signal lines, the entire 64 bit bus is used to transmit data. If at least one signal line on the upper 32 bit bus has a parity error, data is then transmitted only via the lower 32 bit bus. If at least one signal line on the lower 32 bit bus has a parity error, data is then transmitted only via the upper 32 bit bus.

[0007] However, a major drawback of the Olarig '980 system is that if there is a single faulty signal line in both of the upper and the lower 32 bit busses that make up the 64 bit bus, the entire 64 bit bus cannot be used to transmit data. Thus, by just having two erroneous signal lines on the 64 bit bus of Olarig '980 can render the entire 64 bit bus in Olarig '980 unusable. This is inefficient. Another drawback of the Olarig '980 system is that a single faulty signal line in Olarig '980 disables 32 signal lines, forcing data to be transmitted on a 32 bit bus instead of a 64 bit bus. This too is inefficient.

[0008] I have not seen a data bus method or apparatus that disables only individual faulty signal lines and substitutes extra supplemental signal lines for the faulty signal lines to enable data transmission on a data bus having one or more faulty signal lines. Therefore, what is needed is a more efficient data bus arrangement that can efficiently compensate for faulty signal lines without consuming enormous resources.

[0009] It is therefore an object of the present invention to provide an improved method and apparatus for compensating for malfunctioning and faulty signal lines on a data bus.

[0010] It is also an object of the present invention to provide a more efficient method and apparatus for compensating for faulty or malfunctioning signal lines on a data bus in such a way so that enormous or unnecessary resources are not allocated for such a data bus.

[0011] It is further an object of the present invention to provide a method and apparatus for detecting individual faulty signal lines on a data bus and substituting individual functional supplementary signal lines for the faulty signal lines on the data bus so that data can continue to be transmitted over the data bus despite the presence of faulty signal lines.

[0012] It is yet another object of the present invention to have a maintenance signal line on the data bus that is used to identify which signal lines are faulty and which signal lines are being substituted for the faulty signal lines to each module connected to the data bus.

[0013] It is still yet another object of the present invention to have a plurality of modules, each having bus interface units (BIU) connected to the novel data bus to send and receive data to and from the novel data bus.

[0014] It is yet further an object of the present invention to have one of the plurality of modules being a master module and the remaining modules being slave modules, the master module communicates which signal lines are carrying data and which signal lines are faulty by using the maintenance signal line.

[0015] It is yet another object of the present invention to have the novel data bus and the modules part of an asynchronous transfer mode (ATM) cell.

[0016] These and other objects may be achieved by providing a data bus system having a single data bus including a plurality of primary signal lines and a plurality of supplemental signal lines and a maintenance signal line. In the absence of faulty signal lines, only the primary signal lines are used to transmit data. Upon detection of one or more faulty signal lines among the primary signal lines, the faulty signal lines are no longer used to transmit data. Instead, supplemental signal lines are substituted for the faulty signal lines and the data bus can continue to transmit data using the same data bus and using the same number of signal lines. For example, if it is detected that three of the primary signal lines are faulty, three supplemental signal lines are selected to be substituted for the three faulty primary signal lines and the remaining working primary signal lines along with the three selected supplemental signal lines are used to transmit data. The three faulty primary signal lines along with supplemental signal lines that have not been selected remain idle. The maintenance signal line carries information regarding which primary signal lines are faulty and which supplemental signal lines have been selected to transmit data in place of the faulty primary signal lines.

[0017] The above-aforementioned bus arrangement can be applied to an ATM cell. In such a scenario, a plurality of modules are connected to the data bus. Only one of the modules is the master module and the other modules are designated as slave modules. Each module has a BUI. The master module has a bus signal monitor that serves to test and detect when a signal line on the bus becomes faulty, to identify the faulty signal line and inform a microprocessor of the master module of the signal line that has become faulty. The microprocessor then selects one of a plurality of supplemental signal lines that is to be used in place of the faulty signal line. Then the microprocessor informs each module of the erroneous signal line and the selected supplemental signal line. In each module, a bus signal selector is then connected to the corresponding signal lines that are used to transmit or carry data. A bus signal transmitter in each module is then directly connected to the bus signal selector for transmitting data to the bus. The bus is continually monitored by the bus signal monitor and a set of working signal lines are routinely updated as signal lines become faulty.

[0023] One solution in overcoming the problem of the possible occurrence of faulty signal lines is to provide two busses to carry data. Each bus includes a plurality of signal lines. FIG. 1 illustrates such an arrangement where two busses are present in an asynchronous transfer mode ATM cell. Connected to both busses 10 and 20 are m modules, each having a BIU. Data bus 10 is called a primary data bus and data bus 20 is called a substitute data bus. In the absence of faulty signal lines, primary bus 10 only is used to carry data while substitute bus 20 remains idle. Upon detection of a faulty signal line in primary data bus 10, substitute bus 20 is then used to carry data while primary bus 10 remains idle.

[0024] In order for such an arrangement to work in an ATM cell, each of the m modules 30-1 through 30-m must have 2 BIU's, a first BIU 40 connected to primary bus 10 and the second BIU 50 connected to substitute bus 20. Thus, when primary bus 10 is carrying data, each BIU 40 in each module 30-1 through 30-m are used while each BIU 50 in each module 30-1 through 30-m are left idle. When a faulty signal line is detected on the primary bus 10 and the substitute bus 20 is used to carry data, each BIU 40 in each module 30-1 through 30-m are left idle while each BIU 50 in each module 30-1 through 30-m are used in the data transmission process.

[0025] Like the reference to Olarig '980, the solution posed in FIG. 1 has the following drawbacks. FIG. 1 requires an enormous amount of redundant circuitry. As a result, at any given time, m BIU's and an entire bus are left idle. This extra circuitry adds to manufacturing cost and is therefore not a very desirable solution in compensating for a faulty signal line.

[0026] Even worse, in the arrangement of FIG. 1, if a faulty signal line is present in both the primary bus 10 and the substitute bus 20, data can no longer be transmitted. Therefore, it is desirable to have an arrangement for an ATM switch that both is less expensive to manufacture by having fewer redundant parts and is more resilient by compensating for the scenario when a plurality of signal lines go faulty.

[0027] Referring to FIG. 2, a data bus system in an ATM cell according to the principles of the present invention is illustrated. Unlike FIG. 1, FIG. 2 has only a single bus 100. Furthermore, although m modules 200-1 through 200-m are present in FIG. 2, each module has only a single BIU 300 because only a single bus 100 is present. Therefore, unlike the arrangement of FIG. 1, FIG. 2 eliminates the need for a second bus, a second set of BIU's and a second set of electrical connections made from a second set of BIU's to a second bus. FIG. 2 achieves the goal of compensating for faulty signal lines on the bus without all the redundant electrical circuitry that adds greatly to manufacturing costs.

[0028] In FIG. 2, only one of the modules 200-1 through 200-m is a master module and the remaining modules are slave modules. FIG. 3 is a detailed block diagram of FIG. 2. In FIG. 3, module 200-1 is assumed to be the master module and modules 200-2 through 200-m are slave modules. FIG. 3 illustrates all of the circuit components of master module 200-1 in detail along with the detailed construction of bus 100.

[0029] Turning to FIG. 3, bus 100 includes a plurality of (n) primary signal lines 110 (S₁ through S_n), a plurality (y) supplemental (or extra) signal lines 120 (S_n+1˜S_n+y) and a maintenance signal line 140. When there are no faulty signal lines, primary signal lines 110 only carry data on bus 100 and all the supplemental signal lines 120 are left idle. When one or more of the primary signal lines are tested to be faulty, a corresponding number of selected supplemental signal lines 120 are used in substitution for the faulty primary signal lines. Therefore, the faulty primary signal lines and non-selected supplemental signal lines are idle while the non-faulty primary signal lines and the selected supplemental signal lines are used to carry data. It is to be appreciated that maintenance signal line 140 is used only to inform other modules which signal lines on bus 100 are being used to carry data, which signal lines on bus 100 are faulty and which signal lines on bus 100 are currently idle.

[0030] Exactly how the components of FIG. 3 achieve the above results will now be explained. In FIG. 3, master module 200-1 includes a bus interface unit (BIU) 300. BIU 300 includes a bus signal monitor 350 and a bus signal selector 310. Bus signal selector 310 in master module 200-1 is directly connected to all of the signal lines on bus 10 with the exception of maintenance signal line 140. Bus signal monitor 350 is used to test for errors, faults and for malfunctioning signal lines. When bus signal monitor 350 detects a faulty signal line, bus signal monitor 350 informs a microprocessor 210 in module 200-1 that a signal line has been found to be faulty and identifies for the microprocessor 210 which signal line is faulty. Then, microprocessor 210 selects which one of the supplemental signal lines S_n+1 through S_n+y is to be used as a substitute for the faulty signal line. After the microprocessor 210 makes this selection, microprocessor 210 places information regarding which signal lines are faulty, which signal lines carry data and which signal lines are idle on to the maintenance signal line 140 of bus 100 to inform the other modules 200-2 through 200-m. In addition, microprocessor 210 informs bus signal selector 310 which signal lines are faulty, which signal lines are carrying data and which signal lines are idle. Bus signal selector 310 connects the signal lines that are now carrying data to bus signal transmitter 330 to enable sending and receipt of data over bus 100. Reference number 240 is a process bus that serves to connect microprocessor 210 to ROM 220 and RAM 230. Reference number 320 are signal lines of bus signal transmitter 330.

[0031] BUI 300 further comprises ATM cell processor 340 that performs signal processing of an ATM cell stream to the data transmitted from the bus signal transmitter 330 and retransmits the data to the bus signal transmitter 330. Master module 200-1 further includes a read only memory ROM 220 for storing a program necessary for an initial state and operations of the master module and a random access memory RAM 230 used as a temporary data memory for operating the master module.

[0032] Slave modules 200-2 through 200-m are constructed similar to master module 200-1 with the following exceptions. Often, slave modules are absent ROM 220, RAM 230 and microprocessor 210. Furthermore, the bus signal selector 310 in slave modules 200-2 through 200-m may be connected to maintenance signal line 140. This enables the microprocessor 210 of the master module 200-1 to inform the bus signal selectors 310 in each of the slave modules 200-2 through 200-m which signal lines on bus 100 are actively carrying data, which signal lines are faulty and which signal lines are idle. It is to be appreciated that the scope of this invention is not to be limited to a particular design for slave modules 200-2 through 200-m as the composition of the slave modules 200-2 through 200-m may vary, for example, to include a microprocessor and memory.

[0033] A description of the method for operating the novel bus 100 and ATM cell illustrated in FIGS. 2 and 3 will now be made with reference to the flow chart of FIG. 4. At first, the bus signal monitor 350 within the BIU 300 of the master module 200-1 tests the functionality of each signal line within the data bus 100 in step S10. Then, the bus signal monitor 350 in BIU 300 of master module 200-1 delivers the test results to the microprocessor 210 of the master module 200-1 in step S20 for an analysis of the test results and to take further action if necessary. In step S30, if the test results show that one or more previously functioning signal lines on bus 100 have started to malfunction, the control passes to step S40 where further action is required of the microprocessor 210 of master module 200-1. If the test results indicate that there is no change in the functional state of the signal lines in bus 100 since the last time they were tested, the control in step S30 passes to step S60 and no further action need be taken at this time by microprocessor 210 of master module 200-1.

[0034] In step S40, the microprocessor 210 of the master module 200-1 selects and substitutes supplementary signal lines for corresponding newly discovered faulty signal lines. For example, when the n^th signal line (S_n) is determined to have become faulty, the microprocessor 210 selects one of the supplementary signal lines 120, for example signal line S_n+y to carry data in substitute for faulty primary signal line S_n in step S40

[0035] After selecting substitute supplementary signal lines for newly discovered faulty primary signal lines, the microprocessor 210 of the master module 200-1 places information onto maintenance signal line 140 of bus 100 regarding which of the primary lines are now faulty, which of the substitute signal lines are now carrying data and which of the substitute signal lines are idle in step S50. In the case that slave modules 200-2 through 200-m contain a microprocessor, the microprocessors of the slave modules are informed by the maintenance signal line 140 on the bus 100 which signal lines carry data, which signal lines are faulty and which signal lines are idle. Then, the microprocessors of each slave module inform the bus signal selectors 310 of the slave modules which signal lines now carry data to enable the bus signal transmitters 330 in the slave modules to send and receive data off the bus 100. In the case that the slave modules are absent a microprocessor, information regarding which signal lines carry data, which signal lines are faulty and which signal lines are idle is fed directly from maintenance signal line 140 to bus signal selector 310 in each of the slave modules in step S50. In step S60, the bus signal selector 310 in each module 200-1 through 200-m connect only signal lines now carrying data on bus 100 to bus signal transmitter 330 to enable the module to send and receive data off bus 100 despite the fact that one or more signal lines on bus 100 may be faulty.

[0036] The above described operations can occur during initialization of the data bus system and during operation of the data bus system. The data bus system according to the principles of the present invention enables modules to send data to one another despite the fact that one or more signal lines on the data bus may be inoperable. This is achieved with minimal extra circuitry thereby reducing complexity, manufacturing costs and the presence of redundant circuits. Furthermore, the present invention enables data transmission over a bus even if there are many inoperable signal lines, making the above design and process resilient.

[0037] It is to be appreciated that the present invention is not limited to just replacing faulty primary signal lines. It is to be understood that the present invention can be used to compensate for faulty supplementary signal lines as well. Also, the present invention can be applied to other systems such as a motherboard in a personnel computer, etc.

[0038] The data bus system in an ATM cell of the present invention comprises a single data bus only. However, the single data bus is enough for the data bus system to operate normally by substituting a supplementary signal line for a malfunctioning signal line when a signal line on the bus is found to be faulty. Sequentially, efficiency of the data bus can be improved. Also, the data bus system of the present invention can be miniaturized, provide lower production cost and provide a higher reliability.

[0039] While the invention has been particularly shown and described with reference to preferred embodiments thereof, it will be understood by those skilled in the art that the foregoing and other changes in form and details may be made therein without departing from the spirit and scope of the invention.