EP0818119A1

EP0818119A1 - Method of operating a switching facility and an expandable digital switching system for carrying out the process

Info

Publication number: EP0818119A1
Application number: EP96911947A
Authority: EP
Inventors: Jessel Savory; Kenneth Butterfield; Cathal De Brun
Original assignee: Austel Licensing GmbH
Current assignee: Austel Licensing GmbH
Priority date: 1995-03-27
Filing date: 1996-03-27
Publication date: 1998-01-14
Also published as: WO1996031079A1; AU5496996A; CA2216530A1

Abstract

A method of operating a switching facility is based on a data base consisting of individual data sets which can be addressed via a key term. The data base is held on a mass store with relatively high access time. To reduce the data access time, a RAM is provided in which the data sets from the data base can be stored. Also provided is a second store area in which a field of pointers to such data sets can be stored, and access occurs in such a way that a field index is derived from the key term. By requesting a pointer to a data set in the first store area, the data set is taken from either the RAM or the mass store. In the latter case, the data set once read out is duplicated in the RAM and an appropriate pointer is entered. For that purpose, a preferably digital and expandable switching system is provided for switching the PCM signals; the said switching system comprises at least one host computer (60), at least one telephone server (70) and a plurality (50) of elementary telephone switching contacts (100, 101...) which are interconnected via a high-speed communications network (80).

Description

METHOD FOR OPERATING A SWITCHING SYSTEM AND EXTENSIBLE DIGITAL SWITCHING SYSTEM FOR IMPLEMENTING THE METHOD

The invention relates to methods according to the preamble of claim 1; in particular, it relates to a switching system consisting of a switching unit or switching stage and a guest or host computer and also relates to a method for operating the same.

Various tasks have to be mastered in a switching system. This includes setting up and transferring a call, which is referred to as "call management"; furthermore the so-called call services, which include inter alia call billing. For this purpose, for example, for each newly established call, that is, if a connected subscriber calls or is called himself, a data record is created in which the start and end times of the call, date information, the telephone numbers involved and the costs of the conversation and saved permanently. Finally, the operator services, which make up the more organizational part of the system, such as setting up and deleting before customers, configuring circuits or the billing service based on the data of the call, are mentioned Service.

Up until a few years ago, the switching stages for the remote connection were carried out with "intelligence", so that the entire handling of a call could be carried out in the switching stage itself. The programming was carried out essentially in a so-called assembler with non-standardized operating systems, which had to be specially developed for each switching stage. This required a lot of time for development and modifications were very expensive. Of course, services going beyond the elementary "call management" could hardly be implemented or not at all.

ORIGINAL DOCUMENTS A significant improvement results from the use of a standardized operating system, such as "Interactive UNIX", a higher programming language, such as "C", and a standardized, very robust database system, such as "Informix". These software modules are available for all hardware platforms, from the simplest PC to so-called "mamframes", which means that the software can be easily adapted to various powerful guest computers.

However, since the tasks for the guest computer are very diverse and also largely have to be carried out in real time, software development can be extremely complex. In addition, with different sizes of the respective switching unit, very different performance requirements are placed on the guest computer. In particular, accessing the hard disk during database operations places a great strain on the guest computer. Furthermore, different operating systems and database programs are generally required for different performance sizes, as a result of which the software creation and adaptation to different subscriber numbers is still very expensive. Therefore, several guest computers, which carry out the various tasks essentially in parallel, are often networked with one another and, in addition, the operations carried out in real time are reduced to a minimum.

The problem of hard disk access for extracting the data from the typically indexed databases remains. Therefore, in a first aspect the invention is based on the task of simplified and / or faster access to the data, and the solution is provided by the characterizing features of claim 1.

To this end, there are advantageous refinements in the claims

As already explained above, conventional switching systems consist of one or more host computers. Add to that the actual switching matrix and one or more telephone servers. Because of the hardly existing modularity, this architecture can only be expanded with great expense and labor. Modular architectures typically have small "sub-exchanges" formed by elementary telephone switching stages, to which are connected to one another and to the host computers and the telephone servers via a bus or a ring structure. This architecture is modular and therefore also easier to expand or adapt to changing requirements. The main problem lies on the one hand in the lack of reliability in bus-based systems and in general in the capacity limit of the transmission medium, whether a bus or a ring, which is soon reached in the case of extensions.

Therefore, in a second aspect, the object of the invention is to provide an architecture for a switching system which enables almost any expandability with a cost-effective, modular structure with increased reliability of the overall system.

This is achieved through the characteristics of Ar.sprucr.es.

By using the elementary Telepncr.scr.a_tεt .zer. ETS opens up the possibility of smallest Ar.la he. c_s z _ r.anez to build plants of unlimited size with a corresponding communication network at high speeds.

The advantage of a solution according to claim 6, in which the communication network is formed by at least one ATM switching stage, lies in the enormously high processable data rate (approximately 150Mbps) at each connection of the ATM switching stage, which results in a large increase in data throughput Entire system is possible and the system can be expanded almost indefinitely.

The problem with the use of an ATM switching stage in switching systems lies in the fact that the (pulse code modulated) PCM data rates are not large enough for small elementary telephone switching stages (ETS) and it is therefore only sensible, without further measures, to connect large elementary telephone switching stages (ETS) to one another via an ATM switching stage. This results in a rather confusing structure, in which the expansion is only possible in very large capacity steps, which contradicts the desired modular structure. Data concentrators (which are known to be used to reduce the channels in data transmission) in connection with ATM switching stages have so far not been used, probably because of the complexity that is difficult to control.

As already mentioned, these disadvantages are avoided by the features of claim 6. According to the invention, a large number of elementary telephone switching stages (ETS: small, simple, modular) are not connected directly to the connections of the ATM, but rather via concentrators; however, the ATM switching stage is therefore not used as a switching unit in the actual sense, but rather as a link between a plurality of partial switching stages formed by elementary telephone switching stages (ETS) and downstream concentrators. In this way, an ATM switching stage as a connecting element can advantageously bring out its extremely high data rate.

By using an ATM switching stage as a link, a significantly higher data transmission rate would be possible than before, whereby GBytes per second should be reachable. Here the connection of the two above-mentioned aspects of the invention is completed, in that not only is an expandable circuit obtained, but also the faster operation sought is made possible. In this respect, the embodiment shown under rubble represents an invention, also separate from the other features of claim 5, although the modular structure according to claim 5 is also achieved in a particularly favorable manner. In any case, this also means offering new transmission services in principle, which are very limited with the previous circuits Scope with low transmission quality were possible without changing the architecture.

A further advantageous embodiment of the system according to the invention can be found in claims 7.

Further details of the invention result from the following description of exemplary embodiments schematically shown in the drawing. Show it:

1 shows an exemplary arrangement for explaining the method according to the invention;

2 shows the process sequence for processing a conversation by the guest computer; the

Figures 3 and 4 show access to a first type of data; the

Figures 5 and 6 show access to a second type of data;

7 shows a block diagram of a conventional switching system with a bus structure;

Δ is a block diagram of a conventional one. Mediation system with ring structure;

9 shows a block diagram of an expandable digital switching system (SDSS) according to the invention;

10 is a block diagram of an extended SDSS;

11 shows a first example of the construction of an elementary telephone switching stage (ETS); and

12 shows an alternative structure of an ETS. In Fig. 1, a switching unit (1) is connected to other switching systems of a telecommunications network via groups of incoming lines (2) and outgoing lines (3). Furthermore, subscribers (8) are connected via lines (7). A guest or host computer (4) with at least one mass storage unit (6) controls the switching unit (1) via a data connection (5).

The switching unit (1) - controlled by the guest computer (4) - carries out the actual switching of the incoming and outgoing calls. For example, a commercially available "Excel Switch P1X512" unit can be used as the switching unit 1, the internal structure and the internal function of which are known per se and are therefore not described in detail here.

The guest computer (4) essentially performs the above-mentioned “call management”, “call services” and “operator services” tasks (on the basis of a program contained in a memory section or an external program).

The "call management" mainly concerns the establishment ["call set up"] and the transfer of a call ["call rcutmg"]. In this case, a channel bundle is selected for the issuing half of the call and a channel within the bundle is selected from various switching tables.

Call services include setting up and transferring a call, which - as already explained - is referred to as "call management"; furthermore the so-called call services, which mainly include call billing. For this purpose, for example, a record is created for each newly established call, that is, if a connected subscriber is either calling or being called, with the start and end times of the call, the date information, the telephone numbers involved and the The costs of the call are saved and saved permanently. The for Calculation of the data required for costs are taken from a customer database as a data record "ANI", which identifies the customer and also contains a reference "DNIS" in time and weekday specific tariff tables.

The operator services include customer management, which supports the establishment and deletion of customers, switching management for configuration of the switching device, and the call monitor for real-time display of the interior State of the switching device and a billing service, in which case the call data records stored in the billing period are evaluated for each customer, the accumulated call charges are summed up and finally settled.

2 shows in simplified form the process sequence for processing a call by the guest computer. 2 schematically shows the processing of a call in the form of a block diagram; 2a mean: collecting the customer data;

2b: detection of the area code, exchange of data; and 2c: further processing of the call. The data assigned to the customers are taken from an ANI database (20), the subscriber number typically serving as the key. For this, several hard disk accesses are necessary, which with one. typical PC system need about several milliseconds each. For example, the database itself can include 150,000 entries.

Furthermore, the switching information necessary for controlling the switching unit [1) (cf. FIG. 1) and billing the calls must be taken from a further database, the NPANXX database (21). Again, several hard disk accesses are necessary. The NPANXX database typically contains 60,000 entries. Since at least 3 to 4 hard disk accesses, each with about 10 milliseconds of access time, are usually required per database access, PC's as guest computers reach their performance limits even with a small amount of traffic; Artificial handles, such as the so-called "disk chachmg", reduce the access time to the individual data record to milliseconds, but this does not solve the problem.

A remedy for the time-consuming hard disk access is to keep the two databases in the memory of the guest computer. Access to a data record in the memory is about a factor 1000 to 10,000 faster than access to a data record on the hard disk.

This improvement in the access time means in particular that even higher-quality service features, such as the services "callback" and "ready-nng" explained below, are made possible with a single guest computer instead of a guest computer connected to one another by a network. With the "Callback" service, a subscriber calls the operator, i.e. practically the switching unit (1), with a fixed number, rings it and hangs up again; the switching system determines who called, calls the subscriber in question and provides him with a dialing request. The subscriber dials the number of his desired conversation partner and the exchange ensures the correct forwarding (routmg) of the call.

The "ready-nng" service, ie periodic polling of a subscriber line in order to determine whether the subscriber has picked up and then to send him a dialing request signal, requires particularly frequent accesses to the databases.

During the so-called "booting process", a first memory area for the NPANXX data sets used for the transfer and a second memory area for the customer set up the specific ANI data sets. At the start of operation, these two storage areas are still empty, but are already large enough to hold the entire database.

As an alternative or in a further development of the invention - especially if the databases are so large that they cannot be kept entirely in memory - the memory areas can be created in such a way that they can only hold the most frequently used data records. a special LRU algorithm can be used to create space in the event of overcrowding or to select external memories.

When the calls from the various participants arrive after the start of operation, the information relating to the users and the information about the dialed numbers are searched for in the corresponding databases on the disk, read out and stored in the pre-reserved memory areas. The strategy for accessing the data is to first search for the data in the memory and only - if the required data is not yet in the memory - to do so. reload the plate. In this way, the disk accesses are minimized, because after a certain time most of the data is already in the memory and can be read out from there. Experts .. Only when new users call or new destinations are called, new data records have to be loaded from the disk ¬ the.

The access to the NPANXX data for the switching information is described in more detail below with reference to FIGS. 3 and 4.

The database NPANXX contains location, coordinate and distance information of a telephone call. The possible destinations of a conversation are NPANXX-coded in a 6-digit number. Of the 1 million data records possible, only about 61,000 are actually used. It would not make sense 1 To reserve millions of data records, it is sufficient to reserve space for the 61,000 possible data records (30) in the memory during the boot process. At the same time, a field (31) is set up for 1 million pointers that can point to the data records. Another pointer freeptr (32) points to the first free, pre-reserved data record (33) in the memory.

If an NPANXX data record with a specific NPANXX number is now requested "Record the NPANXX number" (41), the first step (42) is to look in the pointer field VNHPTR [NPANXX] under the index NPANXX whether there is already an entry that points to a data record in the memory. If a, the data record can be read out from the memory under the pointer address (43). In this case, access to the data record was made possible by only two memory accesses without a single hard disk access. This completes this process (4a)

If there is no entry in the pointer field VNHPTR under the index NPANXX, the relevant data record must be read from the hard disk (44) and is entered in a free memory area reserved for data records in the memory (45). The address of this newly entered data record is entered in the index field VNHPTR under the index NPANXX (46,, wc - by which the next access to the relevant data record can also take place in memory. This completes this process (4b)

This type of storage organization could be referred to as a "hashing" function, this search function, which generates an index from the key term, in a known manner under which the data record can be found, the identity function. The key value itself specifies the index under which a pointer to the data record can be found. This access method is not possible for the user data from the ANI database, especially since the key term in the ANI database is formed by a 10-digit number. In the following, access to the ANI data is therefore described in more detail with reference to FIGS. 5 and 6.

According to the invention, the hash value is formed here by the last four digits of the key, namely the ANI number. Accordingly, a pointer field hptr (51) with 9999 elements is set up, each element being able to point to one of the pre-reserved ANI data sets (50). Another pointer afreeptr (52) points to the first free pre-reserved ANI data record (53) in the memory.

The above-mentioned type of allocation value naturally leads to multiple assignments of the same allocation value using several key terms in the ANI database.

According to the invention, the collisions which arise are resolved by the fact that each data record in the memory is expanded by a linking field (54) which, if it is not empty, points to further data records which give the same distribution value but different key terms.

If an ANI data record with a specific ANI number is now requested, the hash value hval is first formed (60 / and tested whether an entry exists in the pointer field hptr under the index hval (61).

If the entry is empty, the data record must be read from the hard disk (62) and is entered in the first free reserved memory area (63).

If the pointer field for the relevant index was free (64), the address of the newly stored data record can be entered directly in the pointer field hptr under the index hval (65), otherwise the chain of all data records linked with the entry hptr [hval] as the root must be rolled up (66), so that the address of the newly stored data record finally in the link field of the last data record of the chain, which a must be empty, can be entered (67).

If an address entry already existed when testing the pointer field hptr [hval] (61), this is buffered. This address entry addresses a data record, the ANI field of which is tested for agreement in the next step. If the ANI entry of the data record matches the requested ANI number, the correct data record has been found and can be read out (72). If not, a data record with the correct allocation value was found, but one with an incorrect ANI number.

If the link field of this data record is empty (70), the requested data record must be read from the hard disk, which method has already been described (62-67).

If the link field of the examined data record is not empty, the new data record addressed by the link field is examined (71), which process has already been described (69-72).

6a, 6b and 6c represent the completion of the respective machining process.

FIG. ^N now shows an outline representation of a switching system with elementary telephone switching stages (ETS): a plurality of ETS (11, 12 ...), each of which represents small, relatively independent switching systems and which PCM multiplex lines (for example T1 truncs) (2) connected to their inputs are connected via a central bus (3). Furthermore, a plurality of telephone servers (21, 22, ...) for handling the switching activity beyond an ETS and the host computer (4) are connected to the bus (30). closed. The host computer (4) enables not only the operation and maintenance of the system but also the billing of the telephone calls. A switching system with the architecture shown in FIG. 7 is probably modular in the sense that the elementary telephone switching stages (ETS) and the telephone server represent individual modules, so it is also easily expandable. However, it has two serious disadvantages: the central data exchange element of the system is formed by the bus (30). Should this fail due to a fault, the entire system would become inoperable. If the system is to be expanded by supplementing it with additional ETS and telephone server modules, the bus finally becomes both mechanical (because of the limited number of slots) and electrical (because of the limited driver power) represent a hard limit for expansion in terms of data transmission speed.

The disadvantages mentioned are reduced by an architecture according to FIG. 8. Here the elementary telephone switching stages (ETS) (31, 32, ...), the telephone servers (41, 42, ...) and the host computer (40) are linked together via a data ring (300) connected. Since the ring is preferably designed as a double ring in accordance with the prior art, failure of one ring sector does not immediately result in total failure of the overall system. However, the expandability is also limited by the ring due to the limited data throughput.

9 shows a block diagram of an expandable digital switching system (SDSS) according to the invention. The SDSS is formed from a set (50) of elementary telephone switching stages (ETS (100, 101, ...), a host computer (60), a set (70) of communication systems (80), which consist of concentrators (140, 141) and at least one ATM -Switching stage (160) east.

Each of the elementary telephone switching stages (ETS) (100, 101, ...) is connected to a bundle (110, 111, ...) of PCM-MPX lines (120,121, ...) connected. Each elementary telephone switching stage (ETS) is typically connected to 20 T1-MPX lines corresponding to a capacity of 20 × 24 = 480 PCM connections. The multiplex connections (130, 131, ...) of possibly five such elementary telephone switching stages (ETS) each are connected to a concentrator (140, 141), which is connected with its multiplex connection (150, 151) to a data connection of an ATM switching stage (160) . One (or more) host computer (60) and a plurality of telephone servers (70, 71, ...) are connected to further connections of the ATM switching stage.

Each of the elementary telephone switching stages (ETS) can have or operate a limited number of connections and suitably has an ATM interface which can be integrated with other ETS using the concentrator and the ATM switching stage. The concentrator (140, 141) is responsible for demultiplexing the incoming bandwidth from each ETS m the ATM switching stage. The ATM switching stage forwards all incoming data streams into outgoing streams. The total number of connections to be provided can be determined by the size of the matrix of the ATM switching stage, which can easily be expanded without changes to the hardware in the rest of the system.

In the basic configuration shown in FIG. 9, two concentrators (140, 141) with a total of 10 elementary telephone switching stages (ETS) and a 2x2 ATM switching stage are provided, which results in a switching capacity of 4800 ports. The division of the exchange into elementary telephone switching stage (ETS) on the one hand and into the high-speed communication system with concentrator and ATM switching stage on the other hand allows the ETS to be kept relatively small and therefore flexible and yet the ATM to offer the required high data rate (approx. 150 Mbit / s) at the data connections. The ATM switching stage itself is used as a link for the High-speed connection used for the data streams generated in the ETS and merged in the concentrators.

The particularly great flexibility and expandability of the architecture shown in FIG. 9 for the most varied of requirements is to be explained in the following. 10 shows the extension of the architecture described in FIG. 9 to the switching of twice the number of connections:

A total of four concentrators (201, 204), each with five elementary telephone switching stages (ETS) (210, 211, ... 299), are connected to a 4x4 ATM switching stage (200). The further modules connected to the respective ATM switching stage, such as host computers and telephone servers (cf. FIG. 1), have been omitted to simplify the illustration in this figure. The expandability of the architecture is easy to see. Similar ETS, only doubled in number, are terminated via identical, also twice as many, concentrators to an ATM switching stage with double capacity and result in a switching system with a total doubled capacity (in this case for a total of 9600 connections).

The expandability of the arrangement is further made clear by the fact that, in order to achieve an even higher number of connections, several ATM switching stages form an ATM network. can be interconnected in a known manner. In this way, systems with up to 1 million connections should be possible.

The elementary telephone switching stage (ETS) is a non-blocking switching stage with only a small to minimal capacity, which can supply approximately 480 connections (20 T1 lines). The basic idea behind the design of the ETS lies in an open modular architecture, so that the user can change the configuration simply by installing additional modules in the switching stage. As a result, the switching stage can be expanded so that it also other than Teiephondienste can only perform basic services, for example by adding ISDN interfaces or data communication services etc.

11 and 12 show the internal structure of an ETS. FIG. 11 shows an exemplary embodiment with VME bus modules, FIG. 12 shows an embodiment based on a personal computer (PC) with corresponding plug-in cards.

The elementary telephone switching stage (ETS) in FIG. 11 consists of five module types (each module preferably being formed by one or more printed circuit boards), namely a main control module (MCM) (301), a switching matrix module (SMM) (302), and one T1 interface module (TIM) (303), an interface module for the digital signal processor (DIM) (304) and an interface module for the ATM concentrators (ACIM) (305). In the ETS, each unit, with the exception of the main control module (MCM), is connected to two standard interface units: the standard VME bus (307) and the unit for the telephone path (308). The standard VME bus allows the use of so-called shelf technology, gives the design of the system a certain flexibility, reduces costs and increases reliability, since additional modules can easily be added to the basic system. The ETS exchanges information. to control its modules and the telephone path unit (308) over the VME bus. The data exchange itself takes place through a memory divided into input and output areas I and 0 with a two-port read-write memory (RAM) with free access, which represents a shared, shared memory. Compared to alternative methods, such as an HDLC connection (a data link control at a high level or a bit-oriented data transmission protocol standardized by ISO), this method brings significant speed advantages. The unit for the telephone path (308) consists of ordinary PCM multiplex lines and is used to switch voice and data with digitized PCM format. The maximum bandwidth of the telephone path is determined by the maximum number of non-blocking connections of an ETS.

The main control module (MCM) (301) of the ETS must be able to control and maintain the operation of the ETS, such as switching, control of the telephone path and generation of statistics. Therefore the MCM has a strong micro control, e.g. based on an MC68302, which is specially designed for communications applications; VME plates based on an MC68302 are available on the market from stock. To simplify control and data management in the ETS, the MCM can be equipped with a data interface (for example a SCSI interface) for connecting mass memories, such as a hard disk.

The SMM (302) provides a reciprocal connection of the digital telecommunication data (i.e. both voice and data) via the PCM-Muit piex lines (308) within the entire system. If the maximum switching size of an SM is> 480x480, an ETS up to 48C r.icr. -cloc- operate connections. This means that a ΞTΞ is unable to support the connection of up to 480 lines simultaneously. The main component of the SMM is a switching matrix, like Mitel MT8980.

The T1 interface module (TIM) (303) forms an interface for the digital T1 line data to the digital one. Data streams from the processing system for telecommunications. For this purpose, one or more standardized T1 PCM multiplex lines (309) with a data rate of, for example, 1.55 Mbps can be connected to the module (303). For example, four T1 lines are provided per printed circuit board, the T1 interface module (TIM) also having the functions of signaling and checking in addition to the function as the signal interface and monitoring of the connected PCM lines. Each elementary telephone switching stage (ETS) is equipped with a number of T1 interface modules (TIM) (303) corresponding to the number of T1 lines (309) to be supplied. If an ETS telephone line with special properties is to be able to switch, for example to offer ISDN services, the T1 interface modules (TIM) of the ETS only have to have the corresponding features.

The DSP interface module (DIM) (304) preferably comprises one or more digital signal processors DSP; on the one hand, tones with multiple frequencies are generated and detected for processing certain classes of conversations; on the other hand, many different announcements can be stored in a read-only memory (ROM) and the system can be fed in via the DSP. Finally, all tone signals used in the system (conitiuous tone, cadence tone and shot tone) are generated in the DIM, the parameters of which are also read from a read-only memory and evaluated by the DSP.

Finally, the ETS also has an interface module for the ATM concentrators (ACIM) (305); it serves to forward the data in the individual ETS in order to then have the higher-level switching tasks performed by the ATM switching stage which is used for the linkage. For this purpose, the ACIM has a PCM data link (310) for connection to a concentrator.

Fig. ¹ 2 shows an alternative embodiment of an elementary. Telephone switching stage (ETS) as a system based on a PC. The main control module (MCM) (351) is formed, for example, by a main plate of a PC equipped with at least one 486 or Pentium processor, to which the other modules are connected via one of the available PC buses (356), such as ISA or EISA of the ETS are connected. Some commercially available circuit boards for T1 interfaces (353), which are the interface for a PC System with ISA or EISA and the MVIP bus can already have switching matrix devices (354). By using this type of circuit board, the SMM becomes superfluous. The DSP interface module (DIM) (355) is formed by a DSP card which is available for PC applications. In this version of the ETS, too, an interface module for the ATM concentrators (ACIM) (357) is provided. The ETS can be connected to a concentrator via a PCM data connection (360) originating from the ACIM. The PCM data is transported between the individual modules of the ETS via PCM multiplex lines (358), which are preferably formed by an MVIP bus.

Claims

PATENT CLAIMS

1. A method for operating a switching system with a database which consists of individual data sets which can be addressed using a key term and which is created on a mass storage device with a relatively long access time, characterized in that m a read-write memory with free access ( RAM) a sufficiently large first memory area is set up for storing data records from the database, furthermore a second memory area is set up for storing a field of pointers to such data records and that access to a data record comprises the following method steps:

1: Derive a field index from the key term.

2: Query whether a pointer to a data record in the first memory area of the memory with free access (RAM) is entered in the pointer field of the second memory area under the index derived after step 1.

3a: If there is an entry, the data record searched for will be saved from the first memory. of the memory with free access (RAM) is read out with addressing by the pointer entered in the pointer field, with which the access method has been successfully completed.

3b: If there is no entry in the pointer field, the data record can be read from the mass storage using the key term.

4: Duplication of the data record read from the mass memory into the first memory area m a free data record area.

5: Entry of a pointer to the just duplicated data record in the pointer field at the index derived in step 1, whereby a possible next subsequent Gender access to the same data record is only possible through free memory access.

2. The method according to claim 1, characterized in that the key term is an integer and that the field index and key term are identical.

3. The method according to claim 1, characterized in that the key term is an integer and the field index is derived from the key term so that different Schlüsselbegπffe can lead to the same field index.

4. The method according to claim 3, characterized in that the data records stored in the first memory area are expanded in comparison to those on the mass storage by a pointer field, and steps 1 and 2 are not changed and steps 3 to 5 are changed as follows:

3.1a: If there is an entry, the data record addressed by the pointer entered in the pointer field is checked to determine whether it is actually the data record that matches the key term;

4.1a: If the key term matches and the data record searched has already been read out, the access procedure has been successfully completed;

4.1b: If the key term does not match and there is an entry in the pointer field of the data record, this entry is used as the new pointer and the method is continued with step 2;

4.1c: If the key term does not match and there is no entry in the pointer field of the data record, the data record is removed from the mass storage device using the key term. read and duplicated in the first memory area in a free data record area; the pointer field of this data record remains without an entry, as a result of which the data record is identified as the last one belonging to the relevant index;

5.1: Entry of a pointer to the just duplicated data record in the pointer field of the data record processed in step 4.1, whereby a possible subsequent access to the same data record is only possible through free memory accesses;

3.1b: If there is no entry in the pointer field, the data record can be read from the mass memory using the key term;

whereupon the method is continued with steps 4 and 5 and the pointer field of the data record remains without entry, with which the data record is identified as the last one belonging to the relevant index.

5. Expandable digital switching system for switching PCM signals, in particular for carrying out the method according to one of the preceding claims, characterized by at least one host computer (60), at least one telephone server (70) and a plurality (50) of elementary telephone switching stages (ETS) (100, 101, ..), which are connected to one another via a communication network (80) at high speed.

6. Expandable digital switching system according to claim 5, characterized in that the communication network (80) high speed comprises one or more ATM switching stages (160) serving as a link with at least two upstream concentrators (150, 151).

7. Expandable digital switching system according to claim 5 or 6, with elementary telephone switching stages (ETS) formed from individual modules, characterized in that at least two modules of an ETS are equipped with a common memory or a memory with at least two connections and the data exchange between these modules takes place via the common memory.