WO2002033908A1 - Skalierbares zeiteinteilungsverfahren mit insbesondere abstands- und neueinteilungsfunktion beim versand von atm-zellen sowie zeiteinteilungsvorrichtung dafür - Google Patents
Skalierbares zeiteinteilungsverfahren mit insbesondere abstands- und neueinteilungsfunktion beim versand von atm-zellen sowie zeiteinteilungsvorrichtung dafür Download PDFInfo
- Publication number
- WO2002033908A1 WO2002033908A1 PCT/DE2001/003756 DE0103756W WO0233908A1 WO 2002033908 A1 WO2002033908 A1 WO 2002033908A1 DE 0103756 W DE0103756 W DE 0103756W WO 0233908 A1 WO0233908 A1 WO 0233908A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- time
- data
- vcc
- cell
- atm
- Prior art date
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04Q—SELECTING
- H04Q11/00—Selecting arrangements for multiplex systems
- H04Q11/04—Selecting arrangements for multiplex systems for time-division multiplexing
- H04Q11/0428—Integrated services digital network, i.e. systems for transmission of different types of digitised signals, e.g. speech, data, telecentral, television signals
- H04Q11/0478—Provisions for broadband connections
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/54—Store-and-forward switching systems
- H04L12/56—Packet switching systems
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/54—Store-and-forward switching systems
- H04L12/56—Packet switching systems
- H04L12/5601—Transfer mode dependent, e.g. ATM
- H04L2012/5638—Services, e.g. multimedia, GOS, QOS
- H04L2012/5646—Cell characteristics, e.g. loss, delay, jitter, sequence integrity
- H04L2012/5652—Cell construction, e.g. including header, packetisation, depacketisation, assembly, reassembly
- H04L2012/5653—Cell construction, e.g. including header, packetisation, depacketisation, assembly, reassembly using the ATM adaptation layer [AAL]
- H04L2012/5656—Cell construction, e.g. including header, packetisation, depacketisation, assembly, reassembly using the ATM adaptation layer [AAL] using the AAL2
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/54—Store-and-forward switching systems
- H04L12/56—Packet switching systems
- H04L12/5601—Transfer mode dependent, e.g. ATM
- H04L2012/5678—Traffic aspects, e.g. arbitration, load balancing, smoothing, buffer management
- H04L2012/5679—Arbitration or scheduling
Definitions
- the invention relates to a time division method for the dispatch of ATM cells with the preamble features of claim 1 and a time division device therefor.
- ATM Asynchronous Transfer Mode
- ATM Adaptation Layer Type 2 was (AAL2) protocol standardized as a multiplexing protocol that, for example, the case of voice calls using a fes ⁇ th ATM payload of 48 bytes when using uses of short subscriber packets.
- connection device The currently customary implementation of a connection device required for this is outlined with reference to FIG. 1.
- a first physical input line 1 leads into a connecting device 2.
- Data that are fed into the connecting device 2 via the input line 1 are, in this connection device in particular, virtually changed over or re-divided and then output via a physical output line 3.
- the input line 1 and the output line 3 are currently usually designed as lines with 155 Mbps.
- incoming ATM cells are demultiplexed by an ATM demultiplexer 4 and sent to an AAL2
- Demultiplexer 5 forwarded.
- the data are divided over a large number of virtual connections and finally rearranged in an AAL2 switching device 6.
- the newly ordered data is fed to an AAL2 multiplexer 7 and finally to an ATM multiplexer 8, which collect and prepare the cells for forwarding via the physical output line 3.
- a large number of virtual connections thus arrive at the switching device 6, the data of which must be correspondingly distributed over suitable virtual output connections. Buffering is usually provided so that a certain amount of data arriving and to be processed at the same time can be buffered in buffer devices.
- the Virtual connection takes into account so-called traffic contracts. These system-specific traffic contracts ensure the required quality of service (QoS) for a virtual connection. In particular, these contracts ensure that only as much data traffic is admitted into the switching device 6 as from this and in particular also the physical output line 3 can be managed.
- QoS quality of service
- CBR Constant Bit Rate
- PCR Peak Cell Rate
- CDVT Cell Delay Variation Tolerance
- An entity located in the connection device 2 for monitoring the traffic contract marks or deletes a cell if it does not correspond to the generic line rate algorithm GCRA (I / PCR, CDVT) (GCRA: Generic Cell Rate Algorithm). Marked cells can preferably be deleted from another connection device at a later time if necessary.
- the object of the present invention is to propose a time division method for the dispatch of ATM cells and a suitable time division device therefor.
- Allocation time in a scheduler with a defined number of allocation times an allocation time by means of a function depending on a specific allocation time of the number of allocation times.
- the number of assignment times can be limited to a maximum number in that the function maps the actually progressing time to this specified maximum number of assignment times. For the entry of an event or a reference to such a memory area of the scheduler corresponding to the allocation time, the memory space of the scheduler no longer has to be searched up to this point.
- the search for registered events is easy, since a searched time is directly with a simple function, in particular a hash function, such as the modulo function.
- the result of the calculation then refers directly to the memory space allocated to this time.
- data can be assigned or retrieved at a specific point in time with very little, clearly defined time. Only in the case of several entries in one of the storage spaces must one then search, and this only with regard to the entries therein.
- Times to be assigned outside of the currently assignable times or related events would be especially in the case of Use of a modulo function can be assigned a "wrong" assignment time. In cases where this cannot be ruled out due to the system or due to a sufficient number of assignment times, such times or the associated data can be ignored and / or temporarily registered in a separate mapping area.
- Assigning entry areas in a memory space to the number of assignment times offers the possibility to directly enter data and / or reference information on data, programs or events into the entry areas. It is particularly advantageous to enter pointers in a data storage space, so that the storage space of the scheduler and the storage space for data can be managed separately in terms of system technology and / or construction.
- the data or reference information relating to several events can be stored particularly easily as a set, in particular a list, in the same entry area.
- pointers as elements of a double chain enables simple reintegration or removal or also the reorganization of individual entries in such a list.
- the method for determining a transmission time for a data cell via an output line of a packet-oriented transmission system can be carried out simply and quickly by entering a specific transmission time as the allocated time for the data cell in a scheduler.
- reference information in particular a pointer
- a data cell and / or its storage space referencing enables events to be entered quickly in the scheduler, since the reference information to be managed can be handled with little effort.
- a sufficient spacing of the data to be sent is observed on the output line, a specific traffic pattern can be generated on the output line, so that in communication systems with monitoring devices, the amount of incoming and processed data is used to sort out data for the Case can be prevented in which too much data arrives.
- the monitoring devices check the origin of the incoming data and only accept a certain amount of data of the same origin if the data density is too high, because with the present method, data from different data sources can e.g. mixed in conjunction with a multiplexer and spaced accordingly or distributed over time.
- the method thus advantageously provides a scalable time planner with time allocation functions in connection with the multiplexing of several ATM connections, taking into account time constraints.
- Such an algorithm can advantageously be implemented in a few hardware cycles.
- By the corresponding ⁇ de spacing a desired traffic pattern is formed on each of the output lines.
- This is expedient because otherwise policing data that do not conform to the traffic contract could otherwise be separated from later monitoring devices.
- This leads to optimized fairness when allocating resources for the affected connections.
- the mechanism is scalable, can be implemented in a few cell cycles and does not require dynamic memory allocation. In particular, there is also the possibility of a new timing (re-timing), so that cells can be moved to another time of transmission.
- connection-individual shaping or shaping is combined with the ATM multiplexing of several connections by the spacing before the
- Scheduling is carried out.
- the spacing is possible because only one entry per connection is made in the scheduler.
- double-linked lists with access addresses are advantageously kept.
- other quantity representations are also possible for mapping storage space assignments, in which individual elements or references (pointers) can be extracted.
- a calendar-like table that serves as a data structure for ATM cells for which a specific transmission request that is still in the future has been entered.
- the finished schedule is ultimately in a send queue.
- Chained list structures allow a new timing for cells. Carrying spacing variables for individual virtual connections or a path, ie a group of virtual connections gene, also enables the transmission request to be postponed to a point in time that is permissible in accordance with the corresponding traffic contract.
- AAL2 / ATM supported can perform the following tasks accordingly.
- Data of an ATM line can be demultiplexed into individual virtual ATM channel connections (VCCs) and further demultiplexed into individual virtual AAL2 connections.
- VCCs virtual ATM channel connections
- AAL2 multiplexing is carried out.
- the data are delayed for a maximum predetermined period of time (TCU) or until ATM cell completion.
- TCU maximum predetermined period of time
- transmission devices can be provided to meet the current traffic contract requirements, which can space the cell stream or its data of a virtual ATM channel connection in order to avoid cell losses due to a sorting out by the monitoring device (policer) of a subsequent connection device or the like.
- the ATM output is filled using ATM multiplexing.
- real-time data for AAL2 multiplexes can be placed in queues in switching nodes and the like.
- Multiplexing the time of the cell spacing and an ATM multiplexing delay in addition to the simple timer interval TCU can be used for AAL2 multiplexing in a single control device.
- a modular scheduling algorithm is also advantageously made possible in this way.
- a scalable time division system and a separate spacing algorithm lead to a time planner, which is also scalable with regard to a large number of virtual channel connections, for data cells an uncomplicated assignment or reassignment of specific ones Broadcast times enabled and a cell spacing for secure data forwarding guaranteed.
- AAL2 paths are implemented in a common hardware facility in a scalable manner. This enables the delay due to cell spacing and ATM multiplexing to also be used for AAL2 multiplexing.
- the modular construction of the algorithms makes it easy to adapt the spacing function for AAL2 paths with CBR characteristics to any other, e.g. implement more tolerant path characteristics.
- the architecture of the scheduler forms a basis, in particular for other ATM applications.
- An extension can lead, for example, in the direction of priority planning or spacing based on virtual ATM path connections, with the favorable runtime properties of the algorithms being retained.
- a transfer of the scalable scheduling mechanism to other technologies e.g. the area of the Internet protocol IP is possible, for example to support RTP / UDP / IP multiplexing there.
- a multiplex method with time constraints is first used when the AAL2 / ATM
- 1 shows a switching device in a physical line with basic devices for switching data packets or data cells; 2 shows a scheduler with a large number of queues for virtual connections, a send queue and a calendar-like table; F Fiigg .. 3 3 a double-linked VCC list;
- Fig. 4 shows a flowchart for a generic cell rate algorithm for monitoring the cell current
- Fig. 5 shows a flowchart for an algorithm for determining a theoretical transmission time
- F Fiigg .. 6 6 is a flowchart for an algorithm for updating a theoretical transmission time
- Fig. 7 shows a section from an extended scheduler for the AAL2 standard
- FIG. 8 shows a flowchart for an algorithm for inputting data packets into a multiplexer of this time planner
- FIG. 9 shows a flowchart for an algorithm for outputting data packets from this time planner.
- data is transmitted from stations to other stations.
- the data of a sending station are broken down into individual data packets or cells, sent in this way and reassembled in the receiving station.
- These data from different connections can be unpacked from their original cells on the transmission path and packed into new cells together with data from other connections, in order to optimize the utilization of the physical transmission paths. correspond Accordingly, this data is removed from such cells at another location and ultimately used in cells or data packets that are transferred to the receiving station.
- connection devices 2 are provided, for example switching devices (switches) known in data networks.
- At least one first physical input line 1 leads into a connecting device 2.
- Data which are fed into the connecting device 2 via the input line 1 are converted or newly divided in this connecting device and then output via a physical output line 3.
- the input line 1 and the output line 3 are currently usually, but not necessarily, configured as lines with 155 Mbps.
- incoming ATM cells are demultiplexed by an ATM demultiplexer 4 and forwarded to an AAL2 demultiplexer 5.
- the data are divided over a large number of virtual connections and finally rearranged in an AAL2 switching device 6.
- the newly ordered data is supplied from the switching device 6 to an AAL2 multiplexer 7 and finally to an ATM multiplexer 8, which assemble and provide appropriate cells for forwarding via the physical output line 3.
- AAL2 multiplexer 7 and finally to an ATM multiplexer 8, which assemble and provide appropriate cells for forwarding via the physical output line 3.
- ATM multiplexer 8 which assemble and provide appropriate cells for forwarding via the physical output line 3.
- all steps are carried out, e.g. the de / multiplexing in a control device is purely virtual.
- a method is described below, for example for AAL2, in which the cells on such a virtual channel connection VCC are not assigned to a fixed point in time or time slot as in, for example, a time division multiplexing system (TDM: Time Division Multiplexing), but are processed further as they arrive can be.
- TDM Time Division Multiplexing
- the virtual split-off data expediently contain Assigned a separate header section as required.
- cells of different virtual channel connections VCCs can contribute to a multiplexed ATM cell stream in the ATM output line.
- more than one ATM cell for the ATM output can be completed at the same time, of which one is then successively forwarded via the output line 3.
- connection device 2 is connected or equipped with a scheduler 9.
- an advantageous time planner has a send queue (Send Queue 10), a table constructed in the manner of a calendar, hereinafter referred to as Calendar 11, and a large number of queues for ATM cells for assignment for individual virtual connections, hereinafter VCC Queues 12 on.
- VCC Queues 12 individual cells for sending the respective VCC connections or VCC lines 1, 2,... N are shown.
- the VCC queue 1 has three cells 13 which are to be output at later times via the virtual connection 1 in the direction of the multiplexers 7, 8.
- the VCC queue 12 for the second virtual connection has no data cells to be output at the time shown. Since the cells remain in memory until the end, they are preferably logical multiplexers 7, 8.
- the VCC queue 12 for the third virtual connection in turn has two cells to be output, etc.
- the connection device is connected via the input line 1 2 incoming data is split into a plurality of VCC queues 12 for later output. Incoming data packets or cells can be forwarded unchanged directly into one of the VCC queues 12 or temporarily stored in a memory location corresponding to this VCC queue 12.
- the connection direction 2 incoming data are taken out of the original cells and redistributed, these data then being assigned to corresponding cells of one or more of the VCC queues 12.
- VCC queues 12 or the memory locations assigned to them in the connection device 2 as in the case of AAL2, for example, either previously assembled or transferred whole cells or individual data can be inserted, with individual data then forming cells within the VCC queues 12 13 are put together.
- the listed devices are advantageously logical devices and / or functions.
- the send queue 10 and the calendar 11 are used for inserting data into the VCC queues 12 or for emptying or releasing cells 13 from these VCC queues 12.
- the system determines at which future point in time it should be sent again.
- data that are to be sent on via a specific virtual connection are jointly provided as cell 13 in one of the VCC queues 12.
- An entry is made in the calendar 11 from which the time of transmission can be at least indirectly derived.
- VCC queue 15 is shown in FIG. 3.
- VCC list 15 is shown in FIG. 3.
- the corresponding VCC element 16 with the value 2 did not exist at all or is at least not arranged in the send queue 10, since no data from the second VCC queue 12 are to be output over the second virtual connection in the next time.
- VCC list elements 16 are classified that belong to VCC queues 12, in which
- each of these calendar queues 17 stands for a desired output time or a desired output period, the zeroth calendar queue 17 arranged first being taken into account first, while those in the first calendar queue 17 are processed in a later period.
- the corresponding VCC list 15 is moved from the zeroth calendar queue 17 into the send Transfer queue 10.
- the VCC list 15 of the first calendar queue 17 is transmitted accordingly into the now empty zero calendar queue 17, etc.
- a coherent arrangement of queues can also be provided, each with VCC lists 15 contains.
- a pointer points to one of the queues, which is then handled as send queue 10. After it has been processed, the pointer is then set to the next queue in this comprehensive queue arrangement, so that it then functions as a send queue.
- the assignment of newly arriving data or cells or cells to be reorganized, which is described later, must also be restructured accordingly.
- the VCC list elements 16 in the present case the list element 16 with the value 10 are transmitted into the send queue 10, so that the first cell is subsequently read out of the tenth VCC queue 12.
- the VCC list 15 is restored from the zeroth calendar queue 17 into the send queue 10 or subsequently, the VCC list 15 of the first calendar queue 17 becomes the zeroth
- Calendar queue 17 re-stored, so that after the processing of the VCC list element 16 with the value 10 from the send queue 10, the VCC list elements 16 with the values 7 and 3 then in the zero calendar queue 17 with the values 7 and 3 are Queue to be transferred. Accordingly, the first cell from the seventh VCC queue 12 is output in a third cell cycle in a first cell cycle and then the first cell of the third VCC queue 12 is output via the corresponding virtual connection in a second cell cycle of this third time range.
- VCC list elements 16 are assigned to each of the VCC queues 12.
- the temporal process sequence and the insertion and / or implementation of VCC lists 15 or their VCC list elements 16 into the send queue 10 and the calendar 11 or its calendar queues 17 are carried out with Help with time variables and the modulo function.
- modulo function enables the concept described below to be implemented as a so-called hash function or hash tables.
- a certain memory location is assigned to a large number of incoming cells using a hash function, here the modulo function.
- this allows a dynamic allocation of the storage space, on the other hand, the entire storage space does not have to be searched for this cell in order to find a particular cell later, since it can be found quickly with the hash function, at least with regard to the calendar queue.
- the time is measured in cell cycles that are made available as transport units with a data volume of 53 bytes.
- a cell cycle with the exemplary transmission rate of 155 Mbps shows duration of 2.7 ⁇ sec.
- the start time can be selected as the time at which the operating system was started.
- the scheduler 9 is formed by a field with K columns, which are numbered from 0 to K-1. The oldest date in the schedule is then ⁇ Tomorrow> and the most recent date is then ⁇ Today> + K
- VCC Virtual Channel Connection
- a generic cell rate algorithm which is shown in FIG. 4, is used to monitor the cell flow.
- This algorithm is called every time a cell arrives in the connection device 2.
- the time of arrival the cell always corresponds to the system time, ie the date variable ⁇ no ⁇ v>, which stands for the time since the last system start.
- a static variable ⁇ TAT> is the theoretical arrival time ⁇ TAT> (TAT: Theoretical Arrival Time) for the next cell.
- Each virtual ATM channel connection VCC is assigned a variable ⁇ TAT>. If the next cell does not arrive earlier than ⁇ TAT>, taking into account a certain tolerance (limit), the cell is treated as conforming to the generic line rate algorithm GCRA (J, L), otherwise not.
- an unsuitable cell is marked with a bit for cell loss priority (CLP: Cell Lost Priority), in order to avoid this cell in the event of forwarding problems in the connection device or later in the network in buffers or to be able to delete them entirely.
- CBR Constant Bit Rate
- PCR Peak Cell Rate
- CDVT Cell Delay Variation Tolerance
- GCRA Generic Cell Rate Algorithm
- SCR Sustainable Bit Rate
- BT burst tolerance
- the goal of the spacing algorithm described below is to adapt the cell flow to the traffic contract.
- next possible send time ⁇ NS> must be calculated depending on a desired send target time ⁇ TT> (TT: Target Time) for this cell.
- TT Target Time
- the algorithm “next spacing” or “spacing text” described in FIG. 5 can be used.
- a theoretical send time ⁇ TST> similar to the variable ⁇ TAT> in the cell rate algorithm GCRA is used in this algorithm.
- the next transmission time ⁇ NS> is set to the next theoretical transmission time ⁇ TST> minus the tolerance limit L. Otherwise, if the cell wants to be sent late enough, the next send time ⁇ NS> is set to the target time ⁇ TT>. Finally, the next broadcast time ⁇ NS> is output.
- the "closest spacing" algorithm described with reference to FIG. 5 is called to find out when a cell can be sent, that is, before scheduling the cell.
- the spacing update algorithm is called every time a cell has been sent, i.e. when the planning result is fulfilled.
- variable ⁇ TST> per circuit for virtual connections VCC is stored.
- a highly filled VCC queue means a long delay for the ATM cells, which cannot be tolerated in real time. Therefore, a threshold can be provided for the application purposes for the individual VCC queues 12 which determines a maximum number of ATM cells 13. Above the thresholds, the ATM cells 13 of the corresponding VCC queues 12 are no longer taken into account.
- VCC list elements 16 for the virtual channel connections or their VCC queues 12 are advantageously provided as double-linked lists, as can also be seen from FIG. 3.
- a VCC list element 16 in particular exactly one VCC list element 16, is provided for each VCC queue 12.
- Each of these VCC list elements 16 has a virtual connection identifier (VCI: Virtual Connection Identifier), which is essentially a constant pointer to the VCC queue 12 or to its storage location.
- VCC list element 16 is assigned a first pointer 17 (next) and a previous pointer 18 (previous), these referring to the adjacent VCC list elements 16.
- this structure can be stored in a particularly memory-efficient manner.
- the head pointer 19 refers to any VCC list element 16, in particular to the one that is first referenced within a VCC list 15.
- VCC list element 16 to be edited. All but the last VCC list element 16 in the list point with their pointer 18 to the VCC list element 16, which is to be processed below.
- the last list element in the list refers to the constant zero that terminates the VCC list 15.
- VCC list element 16 ' from the VCC list 15 works as follows: the first pointer 17 (Next) or, in the case of the first VCC list element 16', the head pointer 19 (Head), the VCC to be removed on this List element 16 shows must be directed anew to the VCC list element 16 ′′ pointed to by the first pointer 17 (Next) of the VCC list element 16 ′ to be removed. The same applies to the hands in the opposite direction. In this way, the removal of an individual VCC list element 16 'can be carried out without great effort in a short and known period of time.
- a particular advantage of such a double-linked list over a single-linked list is that in the case of a single-linked list, the pointers or contents of many list elements can be compared.
- the preferred scheduler 9 essentially consists of two different units.
- the calendar 11 holds future send requests from or for virtual channel connections in the form of the addresses of the VCC queue 12, which are assigned to these virtual channel connections and contain cells 13 to be sent.
- the send queue 10 holds send requests for the cases of positive ATM multiplex delay, or send requests for those VCC queues 12, the contents of which are to be transmitted in the very next cell cycles.
- a specific cell cycle can be selected or reserved as the next transmission time ⁇ NS> for each virtual channel connection VCC or for the correspondingly assigned VCC queues 12, in which the next ATM cell 13 from the VCC queue via this virtual channel connection 12 should be sent.
- the order of the cell cycles reserved for reading out the first cell of a VCC queue 12 is entered in the calendar 11.
- the term today or ⁇ Today> for the current cell cycle and the term tomorrow or ⁇ Tomorrow> for the subsequent cell cycle are used for illustration.
- the individual days thus correspond to the individual cell cycles, which in the present exemplary embodiment actually have a duration of 2.7 ⁇ sec.
- a module mapping ⁇ NS> mod K would be the column number or calendar queue number in the Calendar 11 set.
- a modulo mapping two numbers are divided, i.e. a first number divided by a second, floating point numbers are rounded to whole numbers, and the result is a remaining remainder of this division step, so that, for example, 19 mod 6.7 gives 5 as the result , Under this assumption, a field with K columns is sufficient to display all cell cycles that can be referenced.
- Calendar 11 thus comprises columns for the days “tomorrow * ⁇ Tomorrow>) to additionally K days ( ⁇ Today> + K-l).
- the calendar 11 has a field with L areas, in each of which not only a single VCC list element 16 but also a VCC list 15 is stored can be, as can be seen from Fig. 2.
- the double-linked VCC lists 15 in turn contain a large number of VCC list elements 16, each of which represents a VCC queue 12.
- two cases are considered for the assignment of a desired transmission time for a cell 13 which is arranged in a VCC queue 12 as the next cell to be transmitted.
- the desired send time is in the past or for the present line or send cycle.
- the VCC list element 16 assigned to the corresponding VCC queue 12 is inserted directly into the send queue 10.
- a desired transmission time is requested for a future point in time or transmission cycle.
- the VCC list element assigned to this VCC queue 12 is entered in one of the columns of the calendar 11.
- the above-mentioned modulo mapping is preferably used for the assignment of a column. If there is already a VCC list element 16 in the corresponding queues, that is to say the send queue 10 or the corresponding calendar queue 17, this added VCC list element 16 is duplicated with the VCC list element (s) 16 in the manner described above concatenated.
- Such an arrangement of the VCC queues 12, the transmit queue 10 and the calendar 11 with the calendar queues 17 enables a functional interaction such that the spacing functions and the ATM multiplexing can be carried out in a very short time.
- ⁇ rr d ⁇ P d ⁇ P- o F- ⁇ ⁇ P- ⁇ ⁇ P. o to CSI P ⁇ -3 dds Co dj tr dj d II o ⁇ dd F- sQ d tr rt ⁇ F- d ⁇ rr 0) ⁇ tr ⁇ F- P- ⁇ ⁇ P rr P d O
- VCC with a high data rate per connection PCR can therefore be prioritized over virtual ATM channel connections VCC with a low data rate per connection PCR, since these are more influenced by the same absolute ATM multiplexing jitter , Accordingly, the appending of VCC
- the VCC list elements 16, which are registered for the next cell cycle or the next time period, that is to say for “tomorrow * ⁇ Tomorrow>, are transferred from the first calendar queue 17 of the calendar 11 into the send queue 10 ,
- the double-linked VCC list 15 is removed from the calendar 11 at the location ⁇ To orrow + l mod L> and is appended to the end of the double-linked VCC list 15 of the transmission queue 10.
- VCC queue 12 If the VCC queue 12 is not yet empty, as would be the case, for example, for the first VCC queue 12 after reading out the bottom cell 13, then the VCC list element 16 becomes this VCC queue 12 or one of the calendar queues 17, by ordering a cell cycle for the next cell 13 in the VCC queue 12.
- This system advantageously also allows a VCC list element 16 to be rearranged.
- An existing send request can be removed from calendar 11 or send queue 10 by simply removing the corresponding VCC list element 16. Due to the arrangement with double-linked or double-linked VCC lists 15, this is done in a very quick manner using the method described above, with no distinction between VCC list elements 16 in the send queue 10 or a calendar. Queue 17 of calendar 11 is to be made.
- the change of an existing send request for a VCC queue 12 can be implemented accordingly by simply removing its VCC list element 16 and reinserting this VCC list element 16 for the new target time.
- Such a method can be carried out, for example, using an algorithm ReRegistriere ( ⁇ TT>). According to the register algorithm ( ⁇ TT>), such an algorithm removes a VCC list element 16 from its structure and o co ro ro 1- 't- 1
- ACE AAL2 Connecting element
- ATM asynchronous transmission mode
- cells In ATM networks with the asynchronous transmission mode, cells have a fixed useful size of 48 bytes and an information header or information header with a size of 5 bytes. This is a connection-oriented
- VCC virtual channel connection
- the bandwidth is shared among the different VCCs by asynchronously multiplexing the ATM cells belonging to different VCCs.
- VCC queues or the virtual channel connections VCC assigned to them are not assigned to a specific time slot, as in a system with a time division multiple access system (TDMA) system, and these cells are sent as soon as they are sent arrive. Therefore, it is possible that more than one VCC to send a cell at the same time should only be allowed to manage one VCC queue 12.
- the other VCC queues 12 have to wait for the transmission of one cell 13 at a time, whereby only one cell 13 can be transmitted.
- the cells 13 of the other VCCs have to be delayed, which leads to an ATM multiplexing delay.
- This task is carried out by a scheduler 9, which carries out transmission instructions to the virtual channel connections VCC.
- variable ⁇ Tout> is updated as the sum of the current time value or cell cycle ⁇ now> and the time TCU required for multiplexing, whereupon the ATM cell is registered in the calendar 11 for the time ⁇ Tout>.
- the variable ⁇ Tout> is greater than the current time value ⁇ now>.
- the list element of the ATM cell is removed from the calendar 11 and registered for the next cell cycle ⁇ Tomorrow> with the function Registri ere taking into account the spacing, so that it is in the next cell cycle could be sent. If the last ATM cell is not sufficient or does not provide enough space, a new cell must be made available and the variable for the desired transmission time ⁇ Tout> is updated accordingly.
- the function or the subroutine Registri ere ( ⁇ To orrow>) is called, so that the next, "tomorrow" * cell for this VCC queue - the cycle is ordered. Otherwise, the cell is entered in the calendar 11 with a desired or calculated starting time for sending ⁇ Tout>. For this purpose, a comparison is again carried out as to whether the variable value of the starting time ⁇ Tout> is greater than tomorrow's time value ⁇ Tomorrow> or not. Depending on this, the function Register ( ⁇ Tout>) or the function Register ( ⁇ Tomorrow>) is called, after which the function Send ATM cell is ended.
- AAL2 packets enter the scheduler via one or more AAL2 packet input streams, whereby several AAL2 paths can be used.
- AAL2 packet input streams On the output side, one or more ATM cell streams are preferably placed on an output line.
- AAL2 multiplexing, spacing and ATM multiplexing are integrated.
- ATM multiplexing is carried out here by the time planner described above, but AAL2 multiplexing takes place in the queues of the virtual channel connections. Multiplexing of several AAL2 packets into one ATM cell takes place, whereby a timer limits the maximum multiplex time.
- ATM cells can be ready to send in several VCC queues, but only one virtual channel connection or its queue sends an ATM cell at the same time.
- spacing is carried out, taking care that the traffic contract is observed for each virtual channel connection, in particular with a view to PCR and CDVT.
- a concept can be implemented for the allocation of broadcast times, in which a fair distribution is ensured. Overall, this procedure offers a throughput maximization of the entire module or the entire connection device. A particular advantage is shown in real-time processing or almost real-time processing.
- spacing is carried out since other network nodes carry out monitoring or policing at a later point in time and data that are too close to one another could be deleted there.
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
Description
Claims
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
AU2002213825A AU2002213825A1 (en) | 2000-10-16 | 2001-10-01 | Scalable scheduling method having, in particular, a spacing and retiming function during the dispatching from atm cells and a corresponding scheduling device therefor |
US10/399,173 US20040047362A1 (en) | 2000-10-16 | 2001-10-01 | Scalable scheduling method having, in particular, a spacing and retiming function during the dispatching from atm cells and a correspponding scheduling device therefor |
EP01982171A EP1327340A1 (de) | 2000-10-16 | 2001-10-01 | Skalierbares zeiteinteilungsverfahren mit insbesondere abstands- und neueinteilungsfunktion beim versand von atm-zellen sowie zeiteinteilungsvorrichtung dafür |
JP2002536784A JP2004512738A (ja) | 2000-10-16 | 2001-10-01 | Atmセルの送信時にとりわけスペーシング機能および再配分機能を有するスケーラブルな時間配分方法、並びにそのための時間配分装置 |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
DE10051146.5 | 2000-10-16 | ||
DE10051146A DE10051146A1 (de) | 2000-10-16 | 2000-10-16 | Skalierbares Zeiteinteilungsverfahren mit insbesondere Abstands- und Neueinteilungsfunktion beim Versand von ATM-Zellen sowie Zeiteinteilungsvorrichtung dafür |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2002033908A1 true WO2002033908A1 (de) | 2002-04-25 |
Family
ID=7659905
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/DE2001/003756 WO2002033908A1 (de) | 2000-10-16 | 2001-10-01 | Skalierbares zeiteinteilungsverfahren mit insbesondere abstands- und neueinteilungsfunktion beim versand von atm-zellen sowie zeiteinteilungsvorrichtung dafür |
Country Status (7)
Country | Link |
---|---|
US (1) | US20040047362A1 (de) |
EP (1) | EP1327340A1 (de) |
JP (1) | JP2004512738A (de) |
CN (1) | CN100414922C (de) |
AU (1) | AU2002213825A1 (de) |
DE (1) | DE10051146A1 (de) |
WO (1) | WO2002033908A1 (de) |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20060176882A1 (en) * | 2005-02-10 | 2006-08-10 | Beceem Communications Inc. | Method and system of early indication for multi-user wireless communication systems |
US8495309B2 (en) * | 2010-10-22 | 2013-07-23 | Business Objects Software Limited | Variable length data protected by seqlock |
US9960999B2 (en) * | 2015-08-10 | 2018-05-01 | Futurewei Technologies, Inc. | Balanced load execution with locally distributed forwarding information base in information centric networks |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
GB2293720A (en) * | 1994-09-30 | 1996-04-03 | Roke Manor Research | ATM queuing and scheduling apparatus |
US5956340A (en) * | 1997-08-05 | 1999-09-21 | Ramot University Authority For Applied Research And Industrial Development Ltd. | Space efficient fair queuing by stochastic Memory multiplexing |
Family Cites Families (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE19507569C2 (de) * | 1995-03-03 | 1997-02-13 | Siemens Ag | Schaltungsanordnung zur Aufnahme und Weiterleitung von Nachrichtenzellen durch eine ATM-Kommunikationseinrichtung |
US5940397A (en) * | 1997-04-30 | 1999-08-17 | Adaptec, Inc. | Methods and apparatus for scheduling ATM cells |
US5987031A (en) * | 1997-05-22 | 1999-11-16 | Integrated Device Technology, Inc. | Method for fair dynamic scheduling of available bandwidth rate (ABR) service under asynchronous transfer mode (ATM) |
US6226262B1 (en) * | 1997-06-19 | 2001-05-01 | Alcatel Canada Inc. | Correction of calendar based ATM cell scheduling for downstream cell insertion |
US6167059A (en) * | 1998-01-26 | 2000-12-26 | Motorola Inc. | Apparatus and method for transmitting data |
US6526062B1 (en) * | 1998-10-13 | 2003-02-25 | Verizon Corporate Services Group Inc. | System and method for scheduling and rescheduling the transmission of cell objects of different traffic types |
EP1065907B1 (de) * | 1999-06-30 | 2003-07-30 | Alcatel | Verfahren zur Erzeugung von ATM Zellen für Anwendungen mit niedriger Bitrate |
-
2000
- 2000-10-16 DE DE10051146A patent/DE10051146A1/de not_active Withdrawn
-
2001
- 2001-10-01 WO PCT/DE2001/003756 patent/WO2002033908A1/de active Application Filing
- 2001-10-01 EP EP01982171A patent/EP1327340A1/de not_active Withdrawn
- 2001-10-01 JP JP2002536784A patent/JP2004512738A/ja active Pending
- 2001-10-01 CN CNB018203612A patent/CN100414922C/zh not_active Expired - Fee Related
- 2001-10-01 AU AU2002213825A patent/AU2002213825A1/en not_active Abandoned
- 2001-10-01 US US10/399,173 patent/US20040047362A1/en not_active Abandoned
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
GB2293720A (en) * | 1994-09-30 | 1996-04-03 | Roke Manor Research | ATM queuing and scheduling apparatus |
US5956340A (en) * | 1997-08-05 | 1999-09-21 | Ramot University Authority For Applied Research And Industrial Development Ltd. | Space efficient fair queuing by stochastic Memory multiplexing |
Also Published As
Publication number | Publication date |
---|---|
EP1327340A1 (de) | 2003-07-16 |
AU2002213825A1 (en) | 2002-04-29 |
US20040047362A1 (en) | 2004-03-11 |
CN100414922C (zh) | 2008-08-27 |
CN1479989A (zh) | 2004-03-03 |
DE10051146A1 (de) | 2002-04-25 |
JP2004512738A (ja) | 2004-04-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
DE69832205T2 (de) | Paketübertragungsverfahren und -vorrichtung | |
DE69717455T2 (de) | Verfahren und anlage zur steuerung von quellengeschwindigkeit in einem atm netzwerk | |
DE69927808T2 (de) | Adaptive zuweisungsanlage und-verfahren zur bedienung von mehrstufigen qos in der aal-2 schicht | |
DE60036682T2 (de) | Maschine zur gewichteten ringförmigen Ablaufsteuerung | |
DE69827349T2 (de) | Gigabit-Ethernet Schnittstelle für ein synchrones Netz (SONET) | |
DE69937862T2 (de) | Bandbreitensteuerung mit zwei Komponenten, zur Anwendung in digitalen Kommunikationssystemen mit mehreren Klassen | |
DE19726076B4 (de) | Verfahren und Knoten zum Transport und zur Wiedergewinnung von Hintergrunddaten in einem existierenden Paketvermittlungsnetz | |
DE69924732T2 (de) | Quellknoten fuer ein breitbandnetzwerk mit atm zellen | |
DE69931302T2 (de) | Zeitbasierte Ablaufsteuerungsarchitektur und Verfahren für ATM Netzwerke | |
DE69839334T2 (de) | Verfahren zur Zuweisung von Aufwartszeitschlitzen zu einer Netzwerkendeinrichtung , sowie Netzwerkendeinrichtung und Zugriffssteuerung zur Durchführung eines solchen Verfahrens | |
DE60038600T2 (de) | Netzwerk-Datenübertragungs-Zuteilungsverfahren und -vorrichtungen zum Bestimmen einer Paketübertragungspriorität zwichen einer Vielzahl von Datenströmen | |
DE19531749A1 (de) | Verkehrsgestaltungseinrichtung und Paket-Kommunikationsgerät | |
DE29825153U1 (de) | Schalteraufbau eines ATM-Schalters mit großer Kapazität und Mehrklassenkern | |
DE69725718T2 (de) | Verfahren zur strukturierung digitaler daten, die in beiden richtungen eines passiven optischen netzwerks eines pon-tdma-systems übertragbar sind | |
DE2922246A1 (de) | Integriertes erd/satelliten-nachrichtenuebertragungsnetz | |
DE60217685T2 (de) | System und verfahren zum vermitteln von daten unter verwendung eines gemeinsamen koppelfeldes | |
EP0730362A2 (de) | Verfahren und Schaltungsanordnung zum Weiterleiten von über eine ATM-Kommunikationseinrichtung übertragenen Nachrichtenzellen an eine Abnehmerleitung | |
DE19757964A1 (de) | Verfahren zum Organisieren der Übertragung von Datenzellen in einem ATM-Netz | |
EP0730361A2 (de) | Schaltungsanordnung zur Aufnahme und Weiterleitung von Nachrichtenzellen durch eine ATM-Kommunikationseinrichtung | |
DE10023037A1 (de) | Koppelfeld für ein Fernmeldenetz und Verfahren zur Vermittlung in einem Koppelfeld | |
DE60035846T2 (de) | System für Ablaufsteurung zur Übertragung von ATM-Zellen | |
WO2002033908A1 (de) | Skalierbares zeiteinteilungsverfahren mit insbesondere abstands- und neueinteilungsfunktion beim versand von atm-zellen sowie zeiteinteilungsvorrichtung dafür | |
DE60216914T2 (de) | Verfahren zur übertragung eines informationstromes, sowie übertragungsystem, sender und computer produkt zur ausführung davon | |
DE3743685A1 (de) | Koppelfeld und koppelfeldsteuerung fuer ein vermittlungssystem | |
DE60012233T2 (de) | System zum verbesserten ressourcenmanagement für atm-netz-verbindungen |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AK | Designated states |
Kind code of ref document: A1 Designated state(s): AU CN CZ HU JP MX PL US ZA |
|
AL | Designated countries for regional patents |
Kind code of ref document: A1 Designated state(s): AT BE CH CY DE DK ES FI FR GB GR IE IT LU MC NL PT SE TR |
|
DFPE | Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101) | ||
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
WWE | Wipo information: entry into national phase |
Ref document number: 2001982171 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2002536784 Country of ref document: JP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 018203612 Country of ref document: CN |
|
WWP | Wipo information: published in national office |
Ref document number: 2001982171 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 10399173 Country of ref document: US |