AU702953B2 - Loading method for multi computer system - Google Patents

Loading method for multi computer system Download PDF

Info

Publication number
AU702953B2
AU702953B2 AU34287/95A AU3428795A AU702953B2 AU 702953 B2 AU702953 B2 AU 702953B2 AU 34287/95 A AU34287/95 A AU 34287/95A AU 3428795 A AU3428795 A AU 3428795A AU 702953 B2 AU702953 B2 AU 702953B2
Authority
AU
Australia
Prior art keywords
data records
loader
slaves
unit
control unit
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
AU34287/95A
Other versions
AU3428795A (en
Inventor
Tamer Dumanli
Karl-Heinz Legat
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Alcatel Lucent NV
Original Assignee
Alcatel NV
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Alcatel NV filed Critical Alcatel NV
Publication of AU3428795A publication Critical patent/AU3428795A/en
Application granted granted Critical
Publication of AU702953B2 publication Critical patent/AU702953B2/en
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04QSELECTING
    • H04Q3/00Selecting arrangements
    • H04Q3/42Circuit arrangements for indirect selecting controlled by common circuits, e.g. register controller, marker
    • H04Q3/54Circuit arrangements for indirect selecting controlled by common circuits, e.g. register controller, marker in which the logic circuitry controlling the exchange is centralised
    • H04Q3/545Circuit arrangements for indirect selecting controlled by common circuits, e.g. register controller, marker in which the logic circuitry controlling the exchange is centralised using a stored programme
    • H04Q3/54508Configuration, initialisation
    • H04Q3/54516Initialization, software or data downloading
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F15/00Digital computers in general; Data processing equipment in general
    • G06F15/16Combinations of two or more digital computers each having at least an arithmetic unit, a program unit and a register, e.g. for a simultaneous processing of several programs
    • G06F15/163Interprocessor communication
    • G06F15/17Interprocessor communication using an input/output type connection, e.g. channel, I/O port
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L9/00Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols
    • H04L9/40Network security protocols
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04QSELECTING
    • H04Q2213/00Indexing scheme relating to selecting arrangements in general and for multiplex systems
    • H04Q2213/1305Software aspects
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04QSELECTING
    • H04Q2213/00Indexing scheme relating to selecting arrangements in general and for multiplex systems
    • H04Q2213/13109Initializing, personal profile
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04QSELECTING
    • H04Q2213/00Indexing scheme relating to selecting arrangements in general and for multiplex systems
    • H04Q2213/13376Information service, downloading of information, 0800/0900 services

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Computer Security & Cryptography (AREA)
  • Computer Hardware Design (AREA)
  • Signal Processing (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Software Systems (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)
  • Information Transfer Between Computers (AREA)
  • Stored Programmes (AREA)

Description

7 P/00/011 28/5/91 Regulation 3.2
AUSTRALIA
Patents Act 1990 a .o Ao *s *o ta
ORIGINAL
COMPLETE SPECIFICATION STANDARD PATENT Invention Title: "LOADING METHOD FOR MULTI COMPUTER SYSTEM" The following statement is a full description of this invention, including the best method of performing it known to us:- This invention relates to a loading method for a multi-computer system for loading a number of data records from a central load control unit via a communication network into a number of computer modules, With a multi-computer system consisting of several interconnected computer modules, it is increasingly required that the software for the individual computer modules is loaded into the computer modules by a centralised service. With this, it is customary to load the software for the individual computer modules, that is the data records for the individual computer modules, by means of a central loading unit in which they are stored, into the associated computer modules, one by one.
This procedure, however, has the disadvantage that, for a large multicomputer system with a multiplicity of computer modules, much time is required to complete the loading process. Because generally, during the loading process, the S multi-computer system is not operational and thus cannot be used, there is considerable interest in keeping this time as small as possible.
In another known loading method intended especially for switching systems, since these represent a multi-computer system with a large number of computer modules, a data record which is to be loaded into a set of several computer modules is reproduced by the "avalanche method" and stored in all these computer modules.
For this, the data record is first loaded into several predetermined computer modules and then from these copied again into several computer modules, as determined by a distribution list. In the load phase the different data records are sent one after the other to predetermined computer modules and, when required, are immediately afterwards copied to several other computer modules by the "avalanche method" described above. By this method it is achieved that a data record which has to be loaded into several computer modules is very rapidly loaded into them. But if the data records destined for computer modules are different, then this loading method provides no advantage and the loading time is indeed longer than in the first described loading method.
An object of the present invention is, therefore, to load a number of data records into a number of computer modules in as short a time as possible.
According to a first aspect of the invention, there is provided a loading method for a multi-computer system for loading a number of data records from a central load control unit via a communication network into a number of computer modules, wherein, in a first phase, the data records are allocated to n groups of data records and each group is sent from the load control unit to one of n loader slaves and stored there, and that the n loader slaves, in a second phase, simultaneously distribute the stored data records via the communication network to the computer modules and the data records are stored there.
According to a second aspect of the invention, there is provided a load control unit for a loading unit for loading different data records in a multi-computer system via a communication network into a number of computer module, where the load S control unit is provided with a memory unit for storing data records and with a send unit for transmitting data records wherein the load control unit is provided with a grouping unit for allocating data records to n groups of data records and with a o: control unit which is so designed that, in a first phase, it causes each group to be sent to one of n loader slaves and, in a second phase, it causes control instructions to be sent to the n loader slaves, where the control instructions are so selected that they cause the n loader slaves to distribute the data records simultaneously to the computer modules.
According to a third aspect of the invention, there is provided a program module for a load control unit for loading different data records in a multi-computer system via a communication network into a number of computer modules, where the program module contains a set of control instructions wherein the set of control instructions is so structured that when it is executed, it controls the operating sequence in the load control unit in such a way that, during a first phase, the data records are allocated to n groups of data records and each group is sent to one of n loader slaves, and that during a second phase control instructions are sent to the n loader slaves which are so chosen that they cause the loader slaves to distribute the stored data records simultaneously via the communication network to the computer modules.
According to a fourth aspect of the invention, there is provided a multicomputer system with a loading unit, with a number of computer modules and with a communication network by which the loading unit is connected to the computer modules wherein the loading unit is provided with a load control unit and n loader slaves, that the load control unit is provided with a send unit for transmitting data records, a memory unit to store the data records, a grouping unit for allocating the data records to n groups of data records and a control unit which is so designed that, in a first phase, it causes each group to be sent to one of the loader slaves, and in a second phase it causes control instructions to be sent to the n loader slaves, where the control instructions are so chosen that they cause the n loader slaves to distribute the data records simultaneously to the computer modules, and that the n loader slaves are each provided with a receiver unit to receive a group of data records, a memory unit for storing the group, a send unit for distributing the data records of the group to the computer modules via the communication network, and a control unit which is so .6 designed that it causes distribution of the data records in response to a control S instruction from the load control unit.
According to a fifth aspect of the invention, there is provided a switching system with a loading unit, with a number of computer modules to carry out switching functions, and with a communication network by which the loading unit is connected with the computer modules wherein the loading unit is provided with a load control S unit and with n loader slaves, that the load control unit is provided with a send unit for transmitting data records, a memory unit for storing the data records, a grouping unit for allocating the data records to n groups of data records, and a control unit which is so designed that, in a first phase, it causes each group to be sent to one of the loader slaves, and in a second phase it causes control instructions to be sent to the n loader slaves, where the control instructions are so chosen that they cause the n loader slaves to distribute the data records simultaneously to the computer modules, and that the n loader slaves are each provided with a receiver unit to receive a group of data records, a memory unit for storing the group, a send unit for distributing the data records of the group to the computer modules via the communication network, and a control unit which is so designed that it causes distribution of the data records in response to a control instruction from the load control unit.
The basic idea of the invention is that in a load preparation phase the different data records are distributed from a central load control unit to several loader slaves and stored there and, after all the different data records have been distributed, in a load phase all the loader slaves distribute the different stored data records simultaneously to the computer modules. By this means it is achieved that the data records are distributed in the load phase simultaneously from several load sources.
The extra actions required for this, that is the distribution to, and storage in, the loader slaves, take place in the load preparation phase and therefore before the actual load phase. Thus the time required for this is not included in the actual loading time.
Advantageous design versions of the invention are given in the subsidiary S claims.
As already mentioned above, an advantage of the invention is that in the load phase the data records are distributed simultaneously from several load sources, without the preparation of these load sources requiring additional time during the load phase. By using several load sources, on the one hand the speed with which data records can be read out from the internal storage locations of the loading unit is multiplied since data records can be read out from several locations in parallel. The readout speed is then increased again considerably if the data records are stored in a S• fast working store in the loader slaves. On the other hand, the transmission capacity from the loading unit to the computer modules is multiplied since transmission takes place from several points (greater connection capacity, more even loading of the network). Here it is an advantage to locate a loader slave and the computer modules to be loaded by it close together in the network topology, as then the traffic load on the network is kept low. The loading time and therefore the downtime of the multicomputer system is then reduced.
The use of data compression in accordance with the invention has the additional advantage that the quantity of data to be transmitted for a data record is reduced and thus the loading time is shortened again. Here it is especially advantageous to carry out the data compression in the load preparation phase in the loader slaves, and to store the data records there in compressed form, so that the time required for the data compression is not included in the loading time and the processing capacity available for the data compression is multiplied because of the number of loader slaves, which is very advantageous in view of the large quantity of data to be processed.
In order that the invention may be readily carried into effect, embodiments thereof will now be described in relation to the accompanying drawings.
Figure 1 shows a block schematic diagram of a switching system according to the invention with a loading unit and several computer modules according to the invention.
Figure 2 shows a detailed block schematic diagram of a section of the switching system according to the invention shown in Figure 1.
In the design example the use of the method of the invention will now be explained in a switching system according to the invention, with a loading unit 1.6 5 according to the invention which contains a load control unit according to the invention and a program module according to the invention. A switching system is only chosen here as an example of a multi-computer system. It is also possible to use the invention in other multi-computer systems, for example in a large computer consisting of many computer modules, or in a syslem in which distributed individual computers are working together.
Figure 1 shows the switching system with a loading unit LS, a communication network KN and nine computer modules M1 to M9. The computer modules M1 to M3, M4 and M5, and M6 to M9 respectively form sets of computer module groups SET1, SET2 and SET3.
The loading unit LS and the computer modules M1 to M9 exchange data via the communication network KN. The communication n 9twork KN is formed from one, or several, digital switching networks interconnected by jurinction lines. It serves not only for the transmission of information for telephone and data connections, and thus the actual switching task of the switching system, but also for the exchange of data between the computer modules M1 to M9 and the loading unit LS.
It is also possible for the communication network KN to be designed differently.
It can be formed from any arbitrary communication network with distributed or centralised switching which guarantees the information exchange between the loading unit LS and the computer modules M1 to M9.
The computer modules M1 to M9 are, for example, connection modules for analogue subscribers, MF signalling (MF Multi-frequency), clock pulse and tones, or control and supervision. They each consist of at least one processor, at least one memory store, and appropriate peripheral components. Furthermore, they are provided with at least one connection to the communication network KN, over which they can receive data and control instructions.
The nine computer modules, the allocation of the computer modules M1 to M9 to the sets SET1 to SET3 and the three sets, are only chosen here as an example. It is also possible that the computer modules fulfil additional or other functions. This is not Smaterial to the invention.
The detailed construction of a digital switching network or of the computer S e..
•1.15 modules M1 to M9 can be seen, for example, in the article 'Hardware Struktur" [Hardware Structure], pages 135 to 147, of the journal "Elektrisches Nachrichtenwesen", Volume 58, Number 2/3, 1981 or in other articles in this volume.
The loading unit LS has a load control unit LC and three loader slaves LS1 to LS3 which form a set SET of loader slaves. The load control unit LC and the loader t slaves LS1 to LS3 exchange data via the communication network KN.
The load control unit LC and the loader slaves LS1 to LS3 each consist of at least one computer with appropriate peripheral units and each has at least one S: connection to the communication network KN. The operating sequence in the load control unit LC and in the loader slaves LS1 to LS3 is in each case controlled by a program module which can be written in any arbitrary language.
In the load control unit LC nine data records are stored, one for each of the computer modules M1 to M9. Each of these data records contains the control programs as well as the data which is required for the operation of the respective computer module. Such a data record can also be considered as the software controlling the function of a computer module. These data records are sent from the load control unit LC to the loader slaves LS1 to LS3, where the data records for the computer modules M1 to M3 are sent to the loader slave LS1, those for the computer modules M4 and M5 to the loader slave LS2 and those for the computer modules M6 to M9 to the loader slave LS3.
It is also possible that a data record is not stored in the loading control unit LC for every computer module because, for example, the same data record is sent to several computer modules, or because data records are sent only to a portion of the computer modules M1 to M9. The first can quite frequently be the case in multicomputer systems, in which a large number of computer modules have the same function (for example connection modules for analogue subscribers), and the second can be the case when a change of software is only required for a portion of the computer modules.
In the loader slaves LS2 to LS3 the data records are stored temporarily and, in response to a control instruction from the load control unit LC, then sent on to the respective computer modules Ml to M9.
.165 It is also possible for the loading unit LS to be constructed differently. For example, it is possible that the load control unit LC and the loader slaves LS1 to LS3 are not physically separate units, but that they are considered as logical entities. Also, the exchange of data between the load control unit LC and the loader slaves LS1 to LS3 does not have to take place over the communication network KN, but can be carried out over any other arbitrary communication network which is suitable for the transmission of data. In that case the load control unit LC would not be equipped with a connection to the communication network KN.
It is furthermore possible that not all loader slaves LS1 to LS3 have their own connection to the communication network KN, but that a group of the loader slaves LS1 to LS3 share one connection.
If now new software is to be loaded into the computer modules M1 to M9, the procedure is as follows: In a first phase, the load preparation phase, the load control unit LC sends the data records for the computer modules Ml to M9 to the loader slaves LS1 to LS3, in which they are then temporarily stored. In a second phase, the load phase, the switching system, or parts of the switching system, are brought into the loading state.
The load control unit LC then sends control instructions to the loader slaves LS1 to LS3 which cause them to send the data records stored in them simultaneously to those computer modules M1 to M9 for which they are intended. These data records are received by the computer modules M1 to M9 and then stored in them. After this, the switching system is brought into the operational state again and the computer modules immediately start to work in accordance with the new software in the data records.
It is also possible for this loading method to be combined with other loading processes. For instance, it is possible that the load control unit LC only sends a part of the data records to the loader slaves LS1 to LS3, and distributes the rest to the computer modules by itself. Furthermore, it would also be possible for the load control unit LC to separate a data record into two partial data records and only to load one Spartial data record with the method of the invention into the computer module for which the data record is intended. The other partial data record would be loaded with another loading method into this computer module, for example direct from the load 4* *0 S: control unit LC. In the computer module the partial data records would then be combined again into a data record.
Furthermore, it is possible for the computer modules to carry out their function even during the loading state. This can, for example, be achieved by loading the new "2b2 software in parallel to the old and only stopping the execution of the old software o" after the completion of the loading process, and then starting to execute the new software.
The detailed sequence of the loading process will now be explained with the aid of Figure 2 for the loading of data records into the computer modules M1 to M3 as an example. The loading process proceeds similarly for the computer modules M4 to M9.
Figure 2 shows a section of the switching system with the load control unit LC, the loader slave LS1 and the set SET1 with the computer modules M1 to M3. The load control unit LC, the loader slave LS1 and the computer modules M1 to M3 exchange data respectively via the communication network KN.
The load control unit LC has a memory unit MEM1, a grouping unit GROUP, a send unit SEND1 and a control unit CONTR1. The grouping unit GROUP exchanges data with the memory unit MEM1 and sends data to the send unit SEND1. The control unit CONTR1 exchanges control data with the grouping unit GROUP and the send unit SEND1. The send unit SEND1 exchanges data via the communication network
KN.
The memory unit MEM1 stores the data records for the computer modules M1 to M9, data concerning the allocation of these data records to the computer modules, as well as data about the system configuration. For the memory unit MEM1 a memory unit is used which can store a large amount of data permanently, for example a fixed disk store.
The grouping unit GROUP reads out the data records and the other data sequentially from the memory unit MEM1, forms groups of data records DS and generates for each group of data records a distribution list DL which indicates for which of the computer modules M1 to M9 the respective data records of the group .o1 .5 are destined. Then it transfers the data sets DS of a group, and the associated distribution list DL, to the send unit SEND1.
For the formation of the groups of data records the grouping unit GROUP makes use of the data concerning the allocation of the data records to computer modules, and the data about the system configuration. From the data about the B system configuration it determines data about the physical location of the computer S• modules M1l to M9 and of the loader slaves LS1 to LS3. With this data it forms three sets of computer modules, here the sets SET1 to SET3, which each lie close to a loader slave, here the loader slaves LS1 to LS3. From the data concerning the allocation of data records to the computer modules it then determines, from these sets of computer modules, groups of data records, for example here from the sets SET1 to SET3 three groups of data records, which are respectively allocated to the computer modules of the sets SET1 to SET3. It is also possible to apply other, or additional, criteria for the formation of the groups. For example, the data records destined for one exchange could form one group, or the formation of the group could take place on the basis of the most uniform loading of the communication network KN. Furthermore, it is possible that the memory unit MEM1 holds no data about the system configuration and that the formation of the groups follows the sequence of the read-out from the memory unit MEM1I. A formation of the groups is also possible according to the criterion that the total quantity of data of the groups is as equal as possible.
It is also possible for data concerning the allocation of data records to groups already to be stored in the memory unit MEMi. It would then also not be necessary to keep data about the system configuration in the memory unit MEM1.
The send unit SEND provides the communication facilities required for sending data over the communication network KN and carries out the transmission of data and control instructions to the loader slaves LS1 to LS3 and the computer modules M1 to M9. The send unit SEND receives from the grouping unit GROUP groups of data records DS and distribution lists DL, which are respectively allocated to such a group.
Additionally, it receives from the grouping unit data DR concerning the allocation of groups of data records to loader slaves. It establishes connection paths via the
S..
communication network KN and transmits each group of data records and the •15 associated distribution list to the loader slave for which it is intended.
oo It is also possible for the send unit SEND to transmit several groups of data records simultaneously and thus to service a number of different connection paths simultaneously.
The control unit CONTR1 controls the flow of the loading process. If it is instructed to start the loading process by a control instruction, for example from a network management unit not shown here, it then instructs the grouping unit GROUP and the send unit SEND in the first phase, the load preparation phase, to distribute the data records to the loader slaves LS1 to LS3, as described above. When this distribution is complete, the load phase begins. The completion of the distribution is here determined by an appropriate acknowledge signal from the grouping unit GROUP or the send unit SEND which indicates that all the data records have been distributed. However, it is also possible for acknowledge signals from the loader slaves LS1 to LS3 to first indicate the end of the load preparation phase. Such an acknowledge signal would here show that the respective group of data records and the distribution list have been correctly received by the respective loader slave and have been stored.
In the second phase, the load phase, the control unit CONTR1 sends control
I
12 instructions via the send unit SEND to the loader slaves LS1 to LS3 which cause these simultaneously to distribute the data records stored in them.
It is also possible that the loader slaves LS1 to LS3 are not available at the beginning of the load preparation phase to carry out the loading process and that they must first be requested by the control unit CONTR1 from a unit which is responsible for management of the operating resources. The allocation of the loader slaves by such a unit would in particular be considered when the operation of the loader slaves is determined solely by the respective control software.
It is furthermore possible that the number of loader slaves is not fixed in advance, but that it is determined according to the requirement, i.e. that the number of loader slaves depends on the number and size of the data records.
The loader slave LS1 has a receiver unit REC1, a send unit SEND2, a compression unit COMP, a memory unit MEM2 and a control unit CONTR2. The o 15 receiver unit REC1 exchanges data via the communication network KN and sends data records via the compression unit COMP to the memory unit MEM2, and data about a distribution list to the memory unit MEM2. The send unit SEND2 receives data records and data about a distribution list from the memory unit MEM2 and exchanges data via the communication network KN. The control unit CONTR2 exchanges control instructions with the receiver unit REC1 and with the send unit SEND2.
The receiver unit REC1 and the send unit SEND2 respectively provide the communication services required for the reception or transmission of data via the communication network KN.
Via the communication network KN, the receiver unit REC1 receives a group of data records DS which it then transfers to the compression unit COMP. It also receives a distribution list DL, which is allocated to this group, and transfers this list to the memory unit MEM2 where it is stored.
From the memory unit MEM2, the send unit SEND2 reads out the group of data records DS' and the associated distribution list DL and sends the data records to the allocated computer modules in accordance with the distribution list DL.
The compression unit COMP carries out a data compression on the individual data records DS of the group. Compression and decompression algorithms suitable II II 13 for this are, for example, the algorithms LZ 77 or LZ 78 which are known to the expert. An additional algorithm is, for example, described in the US Patent US 4 558 302. The compressed data records DS' are then stored in the memory unit MEM2.
It is, however, also possible to omit the data compression and decompression, and thus to omit the data compression unit COMP.
The control unit CONTR2 controls the operating sequence in the loader slave LS1. It monitors the reception and storing of the data records in the memory unit MEM2 and, in response to a control instruction from the load control unit LC, instructs the send unit SENL)2 to distribute the data records DS' to the respective computer modules in accordance with the distribution list.
It is possible that the send unit SEND2 simultaneously sends several data Srecords DS' to different computer modules and therefore simultaneously serves several different connection paths. This can be useful when the transmission capacity of the loader slave LS1 is greater than that of the individual computer module.
'The computer module Ml has a receiver unit REC2, a decompression unit DKOMP, a memory unit MEM3 and a control unit CONTR3.
The receiver unit REC2 exch.-iges data via the communication network KN and sends data records to the decompression unit DKOMP. The memory unit MEM3 receives data from the decompression unit DKOMP and exchanges data with 1he control unit CONTR2. The control unit CONTR2 exchanges control data with the receiver unit REC2.
The receiver unit REC2 provides the necessary communication services in order to receive data records via the communication network KN. When it receives a data record DS' sent from one of the loader slaves LS1 to LS3 to the computer module M1, it transfers the data record DS' to the decompression unit DKOMP.
The decompression unit DKOMP reverses the data compression carried out in the compression unit COMP and stores the data record, which is now present again in its original form, in the memory unit MEM3. The compressed data record DS' is thus converted back into the data record DS and stored in this form.
The control unit CONTR3 controls the loading process during the loading state in the computer module M1 and, during the operating state, carries out the function
I
14 of the computer module by means of peripheral components which are not shown here, example as a subscriber connection module. In response to a first control instruction from the load control unit LC, which is received via the receiver unit REC2, the computer module changes from the operating state to the loading state. When the data record received from the loader slave LS1 has been loaded into the memory unit MEM3 and a second control instruction is received from the load control unit LC, it changes from the loading state to the operating state and from this moment on carries out its function in accordance with the data record now stored in the memory unit MEM3, i.e. under the control of the software contained therein.
The loading process for the computer modules Ml to M3 takes place as follows: .In the load preparation phase the control unit CONTR1 causes the grouping unit GROUP to read out the data records from the memory unit MEM1. The data records for the computer modules Ml to M3 are then collected into a group and, together with a distribution list which describes the allocation of these data records to the computer modules Ml to M3, are sent to the loader slave LS1. There the data records are compressed in the compression unit COMP and, together with the distribution list, stored in the memory unit MEM2. When all the data records have been sent to the loader slaves LS1 to LS3, the control unit CONTR1 changes to the load phase and sends control instructions to the computer modules Ml to M9 which flag the start of the loading state. Then it sends control instructions to the loader slaves LS1 to LS3 which instruct them to distribute the stored data records. The loader slave LS1 thus distributes the data records stored in the memory unit MEM2 to the computer modules M1 to M3, in accordance with the stored distribution list. The computer module M1, for example, receives a data record from the loader slave LS1, decompresses it in the decompression unit DKOMP and stores it in the memory unit MEM3. At the end of the load phase the control unit CONTR1 sends control instructions to the computer modules Ml to M9 which cause them to return to the operating state again. The computer module M1, for example, then receives this control instruction and operates immediately in accordance with the data record newly stored in the memory unit MEM3.

Claims (14)

1. A loading method for a multi-computer system for loading a number of data records from a central load control unit via a communication network into a number of computer modules, wherein, in a first phase, the data records are allocated to n groups of data records and each group is sent from the load control unit to one of n loader slaves and stored there, and that the n loader slaves, in a second phase, simultaneously distribute the stored data records via the communication network to the computer modules and the data records are stored there.
2. A method as claimed in Claim 1, wherein to each group of data records a distribution list is allocated which, for each data record of the group, designates at least one computer module as the target computer module, that the distribution lists are sent to the respective loader slaves and that the loader slaves distribute the data records according to the respective distribution list.
3. A method as claimed in Claim 1, wherein at the start of the second phase the load control unit sends a control instruction to the loader slaves which causes the latter to start distributing the data records.
4. A method as claimed in Claim 1, wherein the load control unit sends the data records via the communication network to the loader slaves.
A method as claimed in Claim 1, wherein a loader slave simultaneously 20 establishes several connections to computer modules and simultaneously sends data records to different computer modules.
6. A method as claimed in Claim 1, wherein the data records are compressed in the loader slaves according to a data compression algorithm, wherein the compressed data records are sent to the computer modules and wherein the compressed data records are decompressed in the computer modules before being Sstored there.
7. A method as claimed in Claim 1, wherein the data records are stored in the loader slaves in a compressed state.
8. A method as claimed in Claim 1, wherein the number of groups, and therefore the number of loader slaves, is determined by the load control unit according to the number or type of data records to be loaded.
9. A load control unit for a loading unit for loading different data records in a multi-computer system via a communication network into a number of computer module, where the load control unit is provided with a memory unit for sioring data records and with a send unit for transmitting data records wherein the load control unit is provided with a grouping unit for allocating data records to n groups of data records and with a control unit which is so designed that, in a first phase, it causes each group to be sent to one of n loader slaves and, in a second phase, it causes control instructions to be sent to the n loader slaves, where the control instructions are so selected that they cause the n loader slaves to distribute the data records simultaneously to the computer modules. A program module for a load control unit for loading different data records in a multi-computer system via a communication network into a number of computer .fl.
S modules, where the program module contains a set of control instructions wherein the set of control instructions is so structured that when it is executed, it controls the operating sequence in the load control unit in such a way that, during a first phase, the data records are allocated to n groups of data records and each group is sent to one of n loader slaves, and that during a second phase control instructions are sent to the n loader slaves which are so chosen that they cause the loader slaves to distribute the stored data records simultaneously via the communication network to the :0 computer modules.
11. A multi-computer system with a loading unit, with a number of computer modules and with a communication network by which the loading unit is connected to the computer modules wherein the loading unit is provided with a load control unit and n loader slaves, that the load control unit is provided with a send unit for transmitting data records, a memory unit to store the data records, a grouping unit for allocating the data records to n groups of data records and a control unit which is so designed that, in a first phase, it causes each group to be sent to one of the loader slaves, and in a second phase it causes control instructions to be sent to the n loader slaves, where the control instructions are so chosen that they cause the n loader slaves to distribute the data records simultaneously to the computer modules, and that the n loader slaves are each provided with a receiver unit to receive a group of data records, a memory unit for storing the group, a send unit for distributing the data records of the group to the computer modules via the communication network, and a control unit which is so designed that it causes distribution of the data records in response to a control instruction from the load control unit.
12. A switching system with a loading unit, with a number of computer modules to carry out switching functions, and with a communication network by which the loading unit is connected with the computer modules wherein the loading unit is provided with a load control unit and with n loader slaves, that the load control unit is provided with a send unit for transmitting data records, a memory unit for storing the data records, a grouping unit for allocating the data records to n groups of data records, and a control unit which is so designed that, in a first phase, it causes each group to be sent to one of the loader slaves, and in a second phase it causes control Sinstructions to be sent to the n loader slaves, where the control instructions are so chosen that they cause the n loader slaves to distribute the data records simultaneously to the computer modules, and that the n loader slaves are each provided with a receiver unit to receive a group of data records, a memory unit for storing the group, a send unit for distributing the data records of the group to the computer modules via the communication network, and a control unit which is so designed that it causes distribution of the data records in response to a control 2'D instruction from the load control unit.
13. A method substantially as herein described with reference to Figures 1 2 of the accompanying drawings.
14. A switching system substantially as herein described with reference to Figures 1 2 of the accompanying drawings. DATED THIS THIRTEENTH DAY OF OCTOBER 1995 ALCATEL N.V I ABSTRACT A loading method for a multi-computer system which provides fast loading of data records into a number of computer modules. According to the invention a load preparation phase, the data records are distributed to and stored in several loader slaves (LS1 to LS3). In a load phase, all of these loader slaves (LS1 to LS3) then distribute the data records simuiianeously to the computer modules (M1 to M9). During a software change, the data records (software for the individual computer modules) are distributed to a plurality of loader slaves (LS1 to LS3), where they are compressed and stored. Then, the system is put in the boot condition, and all loader slaves (LS1 to LS3) are unloaded at the same time. *go S. *o f S *.see S *SSo S «a f
AU34287/95A 1994-10-29 1995-10-17 Loading method for multi computer system Ceased AU702953B2 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
DE4438697A DE4438697A1 (en) 1994-10-29 1994-10-29 Method for loading multi-computer systems
DEP4438697 1994-10-29

Publications (2)

Publication Number Publication Date
AU3428795A AU3428795A (en) 1996-05-09
AU702953B2 true AU702953B2 (en) 1999-03-11

Family

ID=6532011

Family Applications (1)

Application Number Title Priority Date Filing Date
AU34287/95A Ceased AU702953B2 (en) 1994-10-29 1995-10-17 Loading method for multi computer system

Country Status (2)

Country Link
AU (1) AU702953B2 (en)
DE (1) DE4438697A1 (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE19741870A1 (en) 1997-09-23 1999-03-25 Cit Alcatel Method for distributing data packets of operating software
DE19810843B4 (en) 1998-03-12 2004-11-25 Telefonaktiebolaget Lm Ericsson (Publ) Method and access device for determining the storage address of a data value in a storage device

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0162970A1 (en) * 1984-06-01 1985-12-04 BELL TELEPHONE MANUFACTURING COMPANY Naamloze Vennootschap Multiple memory loading system
US5619655A (en) * 1993-02-26 1997-04-08 Siemens Aktiengesellschaft System for administration and management of network from remote or central station through an operator interface which prepares and stores the management orders for execution

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH06103481B2 (en) * 1985-11-15 1994-12-14 株式会社日立製作所 Program loading method
CA1333418C (en) * 1988-08-02 1994-12-06 Richard Anthony Hatherill Digital key system architecture
DE4134207C1 (en) * 1991-10-16 1993-04-01 Ant Nachrichtentechnik Gmbh, 7150 Backnang, De Loading double-computer standby system - preparing passive computer for loading and taking new software from data source for entering into memory of active computer
FR2682786B1 (en) * 1991-10-17 1993-12-10 Bull Sa DOWNLOADING AN OPERATING SYSTEM THROUGH A NETWORK.

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0162970A1 (en) * 1984-06-01 1985-12-04 BELL TELEPHONE MANUFACTURING COMPANY Naamloze Vennootschap Multiple memory loading system
US5619655A (en) * 1993-02-26 1997-04-08 Siemens Aktiengesellschaft System for administration and management of network from remote or central station through an operator interface which prepares and stores the management orders for execution

Also Published As

Publication number Publication date
AU3428795A (en) 1996-05-09
DE4438697A1 (en) 1996-05-02

Similar Documents

Publication Publication Date Title
CA1114934A (en) Dialed number to function translator for telecommunications switching system control complex
US5751574A (en) Method for loading software in communication systems with non-redundant, decentralized equipment
US4956768A (en) Wideband server, in particular for transmitting music or images
US4993017A (en) Modularly structured ISDN communication system
US4967288A (en) Facsimile storage and exchange system
US5058133A (en) Method and apparatus for digital communication
CN101136932B (en) Cluster type stream media networking system and its content issue and service method
EP0653885A1 (en) Video server
JP2009104603A (en) Method of distributing data packet of system software
DE69133547T2 (en) Packet switching system with self-routing switching networks
US5822422A (en) Method and apparatus for controlling an exchange
AU702953B2 (en) Loading method for multi computer system
CA2051817A1 (en) An arrangement and method of downloading data to a plurality of destinations in a digital telephone system
EP0383931B1 (en) Method of automatically editing data for controlling a processor that carries out distributed control and a system using the same
US6047384A (en) Rapid recovery and start-up system for peripheral systems
US4879714A (en) Channel assignment system for loop network
EP0823186B1 (en) A method of controlling connections in a telecommunication system
AU701633B2 (en) Multiple computer loading method
US6865180B1 (en) Method for bundling connections
JPS63151240A (en) Video program distribution system
JPS60253396A (en) Routing control system
JP2002524889A (en) Resource interface unit for telecommunication switch nodes
JPH0635871A (en) Multiprocessor system
JP3237617B2 (en) Delivery method and device for digital communication file
JPH0662007A (en) Set line restoring device for line setting system

Legal Events

Date Code Title Description
MK14 Patent ceased section 143(a) (annual fees not paid) or expired