CA2130356C - A method and a device for monitoring channel split data packet transmission - Google Patents
A method and a device for monitoring channel split data packet transmission Download PDFInfo
- Publication number
- CA2130356C CA2130356C CA002130356A CA2130356A CA2130356C CA 2130356 C CA2130356 C CA 2130356C CA 002130356 A CA002130356 A CA 002130356A CA 2130356 A CA2130356 A CA 2130356A CA 2130356 C CA2130356 C CA 2130356C
- Authority
- CA
- Canada
- Prior art keywords
- value
- channel
- arriving
- changed
- channels
- 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.)
- Expired - Fee Related
Links
Classifications
-
- 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
- H04L12/5602—Bandwidth control in ATM Networks, e.g. leaky bucket
-
- 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
- H04L12/5601—Transfer mode dependent, e.g. ATM
- H04L2012/5629—Admission control
- H04L2012/5631—Resource management and allocation
- H04L2012/5636—Monitoring or policing, e.g. compliance with allocated rate, corrective actions
-
- 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
- H04L2012/5629—Admission control
- H04L2012/5631—Resource management and allocation
- H04L2012/5636—Monitoring or policing, e.g. compliance with allocated rate, corrective actions
- H04L2012/5637—Leaky Buckets
Abstract
The transfer of data packets are monitored in a line having various logical channels. For each logical channel there is a count value (L i) and at the arrival of a data packet or in each time slot, when the transmission on said line is time multiplexed, i.e. for each tame, when a packet could have arrived in a time multiplexed system, only said count value (L m) for a single channel (m) is decremented.
Then it is determined whether said received data packed, if there is one, is to be forwarded or is to be lost, by the comparison of a calculated value (F n) to a threshold value (T n) belonging to the channel (n), to which said packet belongs. In the calculation of said discrimination value (F n) the number of those channels is used, which have not been decremented after decrementing the channel (n), to which the received packet belongs. In this way the calculated value (F n) can be made to resemble the calculated value or level value, which is used in prior methods for handling loss of packets in a network. For each received packet this method requires only a decrementing of one count value (L n), implying a significant saving of time and a simplification of the circuits used for carrying out the method.
Then it is determined whether said received data packed, if there is one, is to be forwarded or is to be lost, by the comparison of a calculated value (F n) to a threshold value (T n) belonging to the channel (n), to which said packet belongs. In the calculation of said discrimination value (F n) the number of those channels is used, which have not been decremented after decrementing the channel (n), to which the received packet belongs. In this way the calculated value (F n) can be made to resemble the calculated value or level value, which is used in prior methods for handling loss of packets in a network. For each received packet this method requires only a decrementing of one count value (L n), implying a significant saving of time and a simplification of the circuits used for carrying out the method.
Description
I
~1 method and a Device _for_Mon,itc7n~p~ ~:tia,nnel _:~t Data Packet Transmission FIELD OF THE INVENTION
The present invention is related to a method and a device for monitoring the flow of data packets on a transmission line having several channels and in particular for prohibiting that data packets for each channel are t:ransm.itted with an intensity, which is too high and not allowed, what could also be expressed in the way that for time divided data packet traffic having several channels on a line the band width for each channel should always be preserved and not be exceeded.
PRIOR ART
Leaky Bucket (LBj is a prior art method for monitoring t:he band width for an incoming stream of data packets or cells, which for instance are forwarded in a networl~. of the ATM type (Asynchronous Transfer Modej , and i.t i.s desr..°.ribed i.n the US
patent 5,014, 260, Other prior art relevant to the invention appears from the patent documents EP~A1~0 381 2'75, EP-Al-0 2°~5 678, WO-A1-90/05416, US-A-4 896 316, US~A~4 993 024,,. EP-A1-0 310 17:3, EP-A1-0 387 958, EP-A2-0 384 758, EP-Al-0 333 ~~60.
In the prior algorithm counters are provided for each channel.
All these counters are decreased regularly in time with ;~ value, which is predetermined for each channel» When a data packet arrives, the identity or number' of the channel is determined, to which the packet belongs, and then the counter value of this channel is checked or inspected. T.f the s.:ounter value is larger than a threshold value, which is predete~-mi.r~ed far said channel, the data packet is discarded or lost and is thus not forwarded on the transmission line. Other.°wise the <:ounter value for this channel is incremented with a value part:icul.ar for said channel and the data packet is forwarded on the ~k:ransmission line.
By means of this prior process one parameter for the channel rate or band width can be monitored, e.g. trre peak rate or the mean rate. With a small threshold value the peak rate may be W~ 93/i8600 ! ~. ~ ~ y '°< :; PCT/S1E93/~0207 monitored and with a larger threshold value an upper limit of the mean rate rosy be monitored. 8y combining two processes of this kind, providing a so called Dual leaky Pucker, the peak and mean rate for a channel can be monitored. ~Lf any one of the respective threshold values is exceeded, the present data packer is discarded or lost.
In this priar method the whole process is performed in real time without any buffering of several data packets taking place.
Instead there is only one register or memory unit containing one data packet each time. 6dhen the data packet has arrived to this register, it is immediately decided by the comparison described above, if the data packet is to be forwarded. When the number of channels being used on the transmission line is large, there may be difficulties of having rime for the various calculations which are required in the above method. Particularly time consuming is the step of decrementing, in certain short time intervals, the count values for the various channels by a decrement value depending on each channel. If the number of channels for instance is so large as several hundreds, it may be impossible to have time for all these calculations. A
possibility would then be to provide individual counter circuits for each channel, i.e. one hardware counter for each channel.
For cost reasons this solution may be impossible.
SCARY OF TFiE IIdVEIdTIt~Id In order to obviate the problem indicated above a method and a device are proposed according to the invention, the characteristics of which appear from the appended claims, Thus there is, like the prior art, a count value for each channel on the transmission line. On each occasion, when otherwise a1:9. these count values should be decremented, only ~a small portion is decremented, preferably a small number of such count values, e.g. only one value, The decrement:ing of the count values of the channels is performed in.such a way, that said decrementing will pass the channels cyclically. Suitably there is stored in a memory unit, information identifying the channel, for which this value was latest changed.
Where a data packet arrives, i.rm>te~.~ci a c~~.~ca Kra:lrae is calculated, the size caf whic.rr w>rrc>:>pc,n;,i~> t ca the count vali:re which exists in the krnowrr mr~thoct. 't,)n_s calculated value is then compar_E=-:;~ to a t~:rc~:~r~c:l.d ~~r~.lr.zr~ for the charnel, in which the packet: J ~, t:r~uT-tsf ~~~:r:~~~3, ,~r~d if the calc::ulated value is Larger t:haru st.r i c t Imvv_.W rc:> l ct value, the data packet may be discarded anc~ t::hr.rs n~~~:d rsot be forwarded. Otherwise the p~.3a.~;ut i~, a7.w:r~~5 fc~rward.ed and there also said count ~Talue f oz: a frf ;.:viRf~ar if. t hr:~reof as modified. In certa in cases ar ~~rr;:vk~ tv ,rn~ ,' t: r~:::rwarded, for instance if the trarns.miss_i on ~~ct:l.vri t:y i~; srn~~l 1 on the line, even if the packet should have- been dis:~a:r~:iesca according to t:he comparison above. In snack. r:: ~:~:~:.s ~:~ l.at rnay be set in a signal field of the pracket_ I'r~~virrg t. rue rru~~airlirng that the packet can be last or di.scardE:~d. 3.n t n' f,:yl lr~wing on:Ly the case is treated where the pac~.k=~t: ..=r d~ :m.,ac-~,~C direr;tly, but it :is understood, that: the irrvant i.c:~n ~~~~_ua.l ey well can be r.rseci for the case where <.rll.y ~:r_>me inc:ls ;~t.i_~>rr is set that the packet is allowE~d t ca ~>e d~ ~;,:a r ~~ec ,o o- en 3 r; ,~ lower ~>ric~rity than other packets be l.~_.~nyimd :. -> t'm-' same channel.
For instance a loss px:iority i La~~ mr.r~J r~~=~ r~tzt or chang~.d in the header of the data packet, :>ai.d fl~:~~:~ irrctic:ating that the packet has a low l..oss pri<:~ri.t~yt arid :;are be lost or discarded before othE~r pac:k~~tw.: Iae:i.ruca a 1;3.r:ger loss priority, when the tv.;ransrr~:.i.ssic;~rr l~~rd i~ inic ti somewhere in the network.
More specifically, t: he presF=nt i.nYverutio!n pt::oxides, in a packet data communication cyst:~:rn o::ornp~:i..,:irr~x a transmission lines on which data pa<ket~s ~xvcv ~ r~~rc:-~rril~ T-,_ad c:>n ~, plurality of channels, wherein mach char~rnc> 1 r <:s~ ~~ ~r::i ~,T~a l y has a predetermined maxima l packet: t r ~ar7 rr~ r ~.:; r .>rr r-,~ ctuency, ,r first value, and a second valuf~, a rrrr:tto:act .,f' c::ontrolling transmission such that: data pac:ket:.~, are rrc~t transmitted 1,a with a frequency ex~.~eeding a a:h~arzr~m:l.'s pre~t~~t~m:mined maximal packet t:ransrri:i.ss i cn F_ .r c:c:~ut r~c:y c:;~.~rnpr.L; i.rug t he steps of changing the first. value c~t~~ r:caxt: ;:rye: ~:~f trie channels by a first step, wherein t_he ( ir-sr: s-c~.i ur:> area changed at each of successive first tune iraterv<:r_i.; ~:~nr~1 iru ~i c:ycl:ic:
order, for each data packet a:~ x~i.virrc,~ c>n tl-r~ trar~smis.;ion line, determining a r_~hanne.l. nr.rmt,er ~>i ~:~:-, aerri.vir~.g channel carrying the arriving data pac:kFt , arc=t ~:r:mi.7~::i. rod, fc>r Eeach arriving channel, a third va.lr:e::~ bra=e~r~ c:;n t.~e arriving channel's first val~_re and c>n i tm~ ,r~r:Lim :~f >raF: c:>f (a) a i~otal number of dat<~ packets l.rmt t:~a.v~e ,:~rrr it-E~d c.~n t:hEe transmission line sin~~~e t:rne F:i r: st: value ilf t he arriving channel was last chanced, and ~~;~) wli~~rr ~Lac~ c ~:~~,ive data packets arrive during suco:ess:iv~::~ F:c~c;n~ t: i:n~> irotervals, a total number of secon<.~ t~irne ir;t c,~ru~a l:> =i.:~l:~ >e>c:~ since the first value of the arrrivirug ~,:l;~.vrarr~:1 w~~~~ Lc:~ ~t: r::hanc~e~d, wherein the third values c::>f d:i.f:Eez:ert:. arr:iu~inc~ c:hann~els are based on the same one ot: (a) Gar-r;t ~;b) , rr=asps-~c:t ively comparing each third value to eac:r~~ ._~:r_ri 4~ir: ~ c°k~aranel' ~:
second v<~lue and, based c>n 1. h<~ c;c~nap<~r.i s,on, L:~cer forrninq one of the steps of perrn:i..tti ng t..:: ar5s4n i >~,~ :~srv o t t tn a re~spect.ive arriving data packet, dit>c;ard:i.ng t:lm_~ r~e:=ape::t p.ve arriving data packet, and marking the rerap~yrt iva ar:rivz.ng data packet.
The present invention also prc>vidc=.;~ a df=viE:e for controlling transmi;~sion o.f dat~.~ ~>ac~:ket.-, orx a transmission line having a plurality c7f ch~:ur:~rrei.:su~t:l~c tt-rat: data packets are not transmitted with r~ frE::~c~raerrc,y ctrk=wate~xm t han each channel's respective rnaxirna~. f ~~~_W :t: t.m~~rr:~m ~~;:ic>n frec(uency, wherein each channe.l_ has a re:~~~~~e.cti.ve. ~~I~ran_leO num)v>er, comprising means for st::oring ~,~ f: i.r.~t v~.~Lut~ arid a second value for each channe:.t., means f:c:>r <:r~ang i raid t'h~e f first value of i3 channel by a f:i rst st ep, t: he f. i r~>t~ va.iues c:f 3b succ;essive next ones of the c:lo~~or~f'1.:.; k>c=i.r~~g c::k~«Zr~ged during successive first time intervals crick i.n :3 cv;-c_:1 ic; order, means for determining, for each d~~ta p3~~ke~. arriving on the transmission line, a c:knannel rauzni'ac-~r c:af <~ra arriving channel carrying the arriving dat~r:a c3acx:c-at: a~cnd L :a:r clc~k.c::~rrr~.ining, for each arriving channel, a third vuTa:k~.~e ~a;i ee-i c>n the respective arriving channel's first v,~ l acs arid on the ~.~alue of one of (a) a total number :.;f axata l:o~~c~ke=~ta that haYJe arrived on the t.rant;m:ission l.i.t~e~ ~::sir;c.:e ';.r~~:; f::i_rst va:Luc:~
of the arriving channel. wa~~ 7. a:~t: c:lr;:~rug~d, rad (k: when successive data packets arrive during ::>~.~c:cc.-ssive second time intervals, a tot;jl rzurnber of seca.mu,i t.:nre intervals elapsed since the ~~i.r;:~t vaL;.ze c, lt:h~~ <~nr: i..v:~nc~ channel w<is last: changed, wherein the third v~:a:l_ue:~ ~;~f: ct:i..fferent arriving channels are based on ths~ samEa c::~nF:~ of (a) and (b) , a cc>mparator for comparing a thi. ~-c.z va LuE.r tc: t:he respective arriving channel's Eec;orud v<~lu~a; a:.~nG~ zr;F~ ~~s, k~~rse~.~l ors the?
com~aarison, for pai.~fo.x:m:iruc:l c>na c>t: t:it~~ k!~ra<:vti.ons ~:~f permitting transmi;~sic~n of t=he r~aK>pe~~tt ~re a;r.ri.vi.ng dat:.a packet discarding the respective a.rriv7rrg data packet, and marking the respective arr wing cl<lt.a p,.~:::1=st:;t .
The present invention also prow .ekes ~~ mE:thc:d of contrc~>ll.ing data packet transmission on a transrn:iss.ion lizie comprising the steps of providirug a plurality of channels, each channel having a maximal prac:ket t,.r:rrl:~rna ~ ~~ei.c:n f :r~:quc=n<:i~~, a channel number, a c;o~~r~t_ value, ar~.c~ ~~ t:Iuz E:~~>rlc:;~l.c'1 v~:~luc;, changing the count value c.~f eack: of tkue channels in succession, one count value be:izzc~ ~~;tnan«E~~d curing each successive first time intc~rv~L~1~ ~:ac::t~:~z-m:ir:~.i.r;g, for a<~<.h data packet arriving on then tr~~zznt;rni:_~:~:i.c;rl l.ir~e, thc~ channel number of an arriving channel. which i..r: c°arrying t:he arriving data packet, generating, Eor each: :~r.r_ivinq channel, a frequency vfalue i::as~:r on t':Y.cv~ s:rrriv-ing ctnanr;e7.'s court va:l_ue and channel rn.~ml~e:r ~,r~c~ can t';'~~e ~~hannel rrurr~ber corresponding to the count= val.rrF~ tnc:~~st x:wce:lt:l y c:harEgE:d in the changing step; arid cc}rnpar.ing, ~'cox: =~,ach ~;rwi~irrc~
channel, the frequency value ~~rith true r~-~sp~;ct_ive acrriving channel's threshold value and,, k:>asc.c:~ ors t:h+.~ c°c>mpari.son, per:Lorming one o.f the steps o:f: pc:'r:rr~it t i.ng t xvrrrsmi.ssion of the .respective arriving data ~~<.rr..:kc:~t, cW ::~c~<~:~c: sruq the respect ive arriving data ~?ac: kc.:t: ; ~~.r~d r7ia r k:i ;tcr t: he rest>ect ivc:~
arriving data packer .
BRIEF DESCRIPTION OF ':E'HE L)fZAW:Ir~C.
The invention will now be descr~iir~ :ct w:~ t i~ r ~tc-~r-ence to the accompanying drawings, ors whic to -Fig. 1 schematical_Ly she>ws wtr~e:r:e ,~ cif:v i ~:e ,3c:eording i=a the invention may be loc~at ed, -Fic:~. 2 schematical Ly shawl a device f r,~,z c~~rzying out the invention, -Figures 3 and 4 schernatical:I~~ ~~.h~::>w two ca l c-~;ul.ating steps according to the invention, -Figures 5-7 show at3 example rnade fc>r _.~~~x.t::rir~ channel in twhe method according to ttae in~rc~nt:::i_on, when-c. the latter two E,igures show the variation of t:hne c.:our3t vai.iac: and the variation of the ca_Lc~.rlated v~~7~.ze rc.:~;~»~ vt:infely.
WC) 93/18600 ~ .~ '~ &~ ~ .~~ ,~ f~f/S1E93/00207 s,.. a. ~~ ;: _, ; .
In Fig. 1 is schematically shown how a number of data packets, the channel numbers of which are indicated in the drawing symbols of the data packets, arrive to a device 2. according to the invention. In the device according to the invention certain data packets or cells are discarded (lostj or marked, as is illustrated for the packet Craving the channel number n in the Figure. The mark indicates, that the packet can be discarded.
The device according to the invention has the purpose of prohibiting that a data network is overloaded, for instance in particular that a switch unit 3 is flooded by arriving cells.
In Fig. 2 is illustrated in more detail how the device according to the invention is constructed and operates. For a transmission line 5 a buffer memory or register 7 is provided, which can contain maximally one data packet or one cell. When a data packet arrives to the buffer 7, information is fetched from said, packet comprising the channel number n of the packet. This channel number n is then processed by logical circuits 9 in a monitoring unit 11. 6dhen a data packet arrives to the buffer 7, thus a calculation loop is started inside these logical circuits 9.
The monitoring unit 11 also contains several memories or registers for various items used by the logical circuits 9. In these memories or registers thus can be found the stored number n of the channel of the present data packet and a channel number m indicating the number of a channel, the first count value Lm of which was latest changed. Further, information is provided on the tatal number (Mj of channels, which the monitoring unit 11 is capable of treating. For each channel also a threshold value Ti is recorded, a modification value or step value Di, an increment or addition value Ii, and the cou~at value or the first value Li. i here means the channel number and runs from ~. up to and including M. The parameters Ti, Di, and Ii are :individual or particular gor each channel in the general case.
b~aen a data packet arrives to the buffer 7, a signal thereof and information on the channel number n of the packet are sent to dV~ 93/18600 PCf/SE93/00207 the logical circuits 9. These circuits then start the following calculation sequence. In the block 13 a value Fn for the channel is calculated. This value depends on the channel number and also on the number of channels, which are left and the values of which are to be changed in steps until the value for exactly this channel shall be changed or decremented. This calculated value Fn is then compared in the block 15 to a threshold value Tn for the channel n and if the threshold value is smaller than this ~:alculated value Fn, the packet is discarded and a suitable signal is in the block 16 sent to the buffer 7, this signal having the meaning, that the buffer shall not forward the packet. If the comparison in the block 15 instead receives the answer No, the packet is accepted and is to be forwarded and in tine block 17 a signal thereof is sent to the buffer 7. Then the count value Ln for this channel is changed in the block 1~.
The transmission of packets on the line 5 is supposed to be regularly divided in time in such a way, that packets are sent or arrive on fixed, recyularly repeated times or in the corresponding periods between said times, that the transmission takes always place in definite time slots. The time period between these repeated times can also be termed a cell time or cell period. With a small load on the network packets do not arrive in each time slot. Further, for each time slot or cell period, in the block 21 the value of a variable m is changed, this variable indicating the number of that channel, for which the count value Lm was latest decremented. The change can be performed for instance by incrementing m by one. when this is performed for this next channel in turn having number m, the first count value Lm belonging thereto is modified in the block 23. When these steps have been performed, the process will await the beginning of the next cell period. For a full load on 'the line 5 this condition is the same as the arrival of a new data packet to the buffer 7.
In Fig. 3 is illustrated in detail the calculation of Fn in the block 13. In a bloc~.k 25 the present number m, which indicates the number of the channel, the count value Lm of which was latest changed, is thus compared to the channel number n of the WO 93/18600 PC'f/SE93/00207 s -y - .~ c. a .J ,,' ..
received data packet and if 'this latter value n is less than or equal to the number m, in the block 27 Fn is calculated as Ln Dn'(m - n). If instead the comparison receives the result N~, in the block 29 the calculation of Fn is performed by means of the e~cpression Ln - Dn' (rI + m - n) , These two cases are necessary, since the channel number m for the channel, for which the count value has been latest decremented, for instance cyclically runs through the values U, 1, 2, ..., t~i - 1. The expression inside the brackets in the blocks 27 and 29 will thereby indicate the number of channels, the count values of which remain to be decremented, before the count value of the channel having the number n belonging to the present cell is to be decremented.
Tn Fig. 4 is illustrated in detail how the change of the count value Ln in block 19 is performed. F3ere first in a block 31 it is compared, if the calculated value Fn is larger than or equal tn zero. If the answer from the comparison is X1ES, in the block 33 the count value Ln is decremented with the step In, which is stored in the monitoring unit r1. If the answer from the comparison is IdO, in a block 35 the channel number m for the channel, the count value Lm of which has been latest decremented, is compared to the channel number n of the present cell. If the former channel number m is larger than the channel number of the present cell n, 'the block 37 is performed, where a new count value is provided by the ea~pression In + Dy (r - n), while in the opposite case the count value is given the value 7Cn + Dn' (P~ + m - n) in the block 39.
In the latter case, when the calculated value Fn thus is smaller than zero, the new values of the count value Ln axe set in such a way, that a new value Fn calculated by means of theses values is given the value zero.
Below is given, formulated in pseudo code, the process which is performed each tire, when a data pac~Cet arrives to the ronitoring unit 3.
P~fi'/S E93/00207 W~ 93/18600 N .,G t~ ~~ ;.J r.~ ..
Process begin =.
repeat wait (at) if(cell arrived) begin if m >~ n then Fn : Ln ° Dn~(m ° n)% < The real level value else . for channel n is Fn : Ln - Dn*(M + ~a - n)% . interpolated.
if Fn >= 0 then . If Fn < Tn the cell if Tn -~ Fn >= 0 then , will be accepted begin . and the level value Ln '-- Ln + zn% - is increased by the "Accept cell'° . value In.
end else "Discard .cell°' else . Tf Fr1 < 0, some begin . error is present, if m >= n then . since the real level Ln : In + Dn*(m - n)% . value cannot be else , below zero. Ln is Ln : In + Dn ~ (M -~ m , given a value ~rhich n % . causes that Fn ~Cu aeACCept Cell~ . given the Value end . zero.
end m : (m -~ 1)P~IOD Pi; . The variable m taDces Lm := Lm M*Dm; . care of the channel if L~ < 0 then . ~rhich is to be Lm : 0% . decremented. m is forever . incremented in the end . channel number . order.
wo v~ns~oo ~c°ros~y~eoo2o7 ~~,~;~~~~;c~, ~xAZ~L~
The difference between the count value Ln and the calculated level value Fn for a channel will be elucidated with an example illustrated in Figures 5 - 7. We will consider that 10 cells arrive sequentially all having the same channel number, say channel number Z, followed by 10 cells having some other channel number and thereafter further 12 cells all having the channel number 1, as appears from Fig. 5.
Figures 6 and 7 show how the count value hl, which also can be termed the fictive level value, and the calculated or "real' level value Fn respectively will vary, if the input signal on the line ~ is such as shown in Fig. 5. The numbers in the drawings are calculated for the following values of the variables belonging to channel 1:
The number of channels, which can be monitored by the device, is set to 32. In real cases the number of channels, which can be monitored by the device, will be significantly larger.
- DZ: The channel dependant decrement value D1 for channel 1 is set to ~.
' Il: The increment value I1 is set to 28600.
- T1: The threshold value is set to a very high value.
The following process will then be obtained and is first illustrated in Fig. ~, showing the variation of the count value L1, when cells arrive. The states arc numbered witty the channel number m for the count value Ln which has been latest decreased:
~t = 1, 2, ... 10: Since the threshold value T1 is very large, all of the 5.0 first cells will be accepted. The level count L1 is increased by the value I1 for each cell, which has the channel number 1. and which fs accepted. After the first 10 cells have arrived, Ll has thus obtained the value 10°I1 (=286000).
- m ~ 11, 12, ..,, 20: The following 10 cells have not channel number 1 and thus L1 is not influenced.
' m = 21, 22, ..., 31, 0: When the first cell having channel number 1 arrives (m = 21), it is discovered that the real level value Fl is smaller than zero (see Fig. 7). In order to correct it in such a way, that this value F1 actually will indicate a avo 9si~~soa ~~-'ri~~93i~z~7 g .~ . t: ~ :~ :_ value, which corresponds to the level value in the known algorithm, L1 is given a value (T1 -~ (m - n)°~31), which makes the real level value F1 equal to zero. After that Ll is increased by T1 for each cell, which arrives and has channel number 1. 6dhen the variable m again takes the value 1, channel 1 is decremented by the value M°D1.
- m = 1: 45hen m = 1, the count value L1 of channel 1 is to be decremented by M'D1.
Tn Fig. 7 is shown the variation of the real or the calculated level value F1 plotted against the number of arrived cells. The process is as follows:
- m = 1, 2, ..., l0: The first 10 cells are accepted by the process. For each accepted cell a value T1 is added to L1. At the same time the variable m is increased by one, ~rhich results in F1 being decreased by D1 for each arriving cell.
- m = 11, 12, ..., 20: These cells have not channel number 1 and thus L1 is not influenced. Tnstead F1 is continuously decreased by the leakage value D1.
- m = 21, 22, ..., 31, 0: When the first cell having channel number 2 arrives, the real value F1 is calculated. Then it is discovered, that F1 has a value smaller than zero (_ --3.07260).
~y giving L1 the value T1 ~- (m ° n)'D1 F1 is given the value zero.
- m = 1: Then m = 1 L1 is decremented by ~d'D1. This gives F1 =
Lx.
~1 method and a Device _for_Mon,itc7n~p~ ~:tia,nnel _:~t Data Packet Transmission FIELD OF THE INVENTION
The present invention is related to a method and a device for monitoring the flow of data packets on a transmission line having several channels and in particular for prohibiting that data packets for each channel are t:ransm.itted with an intensity, which is too high and not allowed, what could also be expressed in the way that for time divided data packet traffic having several channels on a line the band width for each channel should always be preserved and not be exceeded.
PRIOR ART
Leaky Bucket (LBj is a prior art method for monitoring t:he band width for an incoming stream of data packets or cells, which for instance are forwarded in a networl~. of the ATM type (Asynchronous Transfer Modej , and i.t i.s desr..°.ribed i.n the US
patent 5,014, 260, Other prior art relevant to the invention appears from the patent documents EP~A1~0 381 2'75, EP-Al-0 2°~5 678, WO-A1-90/05416, US-A-4 896 316, US~A~4 993 024,,. EP-A1-0 310 17:3, EP-A1-0 387 958, EP-A2-0 384 758, EP-Al-0 333 ~~60.
In the prior algorithm counters are provided for each channel.
All these counters are decreased regularly in time with ;~ value, which is predetermined for each channel» When a data packet arrives, the identity or number' of the channel is determined, to which the packet belongs, and then the counter value of this channel is checked or inspected. T.f the s.:ounter value is larger than a threshold value, which is predete~-mi.r~ed far said channel, the data packet is discarded or lost and is thus not forwarded on the transmission line. Other.°wise the <:ounter value for this channel is incremented with a value part:icul.ar for said channel and the data packet is forwarded on the ~k:ransmission line.
By means of this prior process one parameter for the channel rate or band width can be monitored, e.g. trre peak rate or the mean rate. With a small threshold value the peak rate may be W~ 93/i8600 ! ~. ~ ~ y '°< :; PCT/S1E93/~0207 monitored and with a larger threshold value an upper limit of the mean rate rosy be monitored. 8y combining two processes of this kind, providing a so called Dual leaky Pucker, the peak and mean rate for a channel can be monitored. ~Lf any one of the respective threshold values is exceeded, the present data packer is discarded or lost.
In this priar method the whole process is performed in real time without any buffering of several data packets taking place.
Instead there is only one register or memory unit containing one data packet each time. 6dhen the data packet has arrived to this register, it is immediately decided by the comparison described above, if the data packet is to be forwarded. When the number of channels being used on the transmission line is large, there may be difficulties of having rime for the various calculations which are required in the above method. Particularly time consuming is the step of decrementing, in certain short time intervals, the count values for the various channels by a decrement value depending on each channel. If the number of channels for instance is so large as several hundreds, it may be impossible to have time for all these calculations. A
possibility would then be to provide individual counter circuits for each channel, i.e. one hardware counter for each channel.
For cost reasons this solution may be impossible.
SCARY OF TFiE IIdVEIdTIt~Id In order to obviate the problem indicated above a method and a device are proposed according to the invention, the characteristics of which appear from the appended claims, Thus there is, like the prior art, a count value for each channel on the transmission line. On each occasion, when otherwise a1:9. these count values should be decremented, only ~a small portion is decremented, preferably a small number of such count values, e.g. only one value, The decrement:ing of the count values of the channels is performed in.such a way, that said decrementing will pass the channels cyclically. Suitably there is stored in a memory unit, information identifying the channel, for which this value was latest changed.
Where a data packet arrives, i.rm>te~.~ci a c~~.~ca Kra:lrae is calculated, the size caf whic.rr w>rrc>:>pc,n;,i~> t ca the count vali:re which exists in the krnowrr mr~thoct. 't,)n_s calculated value is then compar_E=-:;~ to a t~:rc~:~r~c:l.d ~~r~.lr.zr~ for the charnel, in which the packet: J ~, t:r~uT-tsf ~~~:r:~~~3, ,~r~d if the calc::ulated value is Larger t:haru st.r i c t Imvv_.W rc:> l ct value, the data packet may be discarded anc~ t::hr.rs n~~~:d rsot be forwarded. Otherwise the p~.3a.~;ut i~, a7.w:r~~5 fc~rward.ed and there also said count ~Talue f oz: a frf ;.:viRf~ar if. t hr:~reof as modified. In certa in cases ar ~~rr;:vk~ tv ,rn~ ,' t: r~:::rwarded, for instance if the trarns.miss_i on ~~ct:l.vri t:y i~; srn~~l 1 on the line, even if the packet should have- been dis:~a:r~:iesca according to t:he comparison above. In snack. r:: ~:~:~:.s ~:~ l.at rnay be set in a signal field of the pracket_ I'r~~virrg t. rue rru~~airlirng that the packet can be last or di.scardE:~d. 3.n t n' f,:yl lr~wing on:Ly the case is treated where the pac~.k=~t: ..=r d~ :m.,ac-~,~C direr;tly, but it :is understood, that: the irrvant i.c:~n ~~~~_ua.l ey well can be r.rseci for the case where <.rll.y ~:r_>me inc:ls ;~t.i_~>rr is set that the packet is allowE~d t ca ~>e d~ ~;,:a r ~~ec ,o o- en 3 r; ,~ lower ~>ric~rity than other packets be l.~_.~nyimd :. -> t'm-' same channel.
For instance a loss px:iority i La~~ mr.r~J r~~=~ r~tzt or chang~.d in the header of the data packet, :>ai.d fl~:~~:~ irrctic:ating that the packet has a low l..oss pri<:~ri.t~yt arid :;are be lost or discarded before othE~r pac:k~~tw.: Iae:i.ruca a 1;3.r:ger loss priority, when the tv.;ransrr~:.i.ssic;~rr l~~rd i~ inic ti somewhere in the network.
More specifically, t: he presF=nt i.nYverutio!n pt::oxides, in a packet data communication cyst:~:rn o::ornp~:i..,:irr~x a transmission lines on which data pa<ket~s ~xvcv ~ r~~rc:-~rril~ T-,_ad c:>n ~, plurality of channels, wherein mach char~rnc> 1 r <:s~ ~~ ~r::i ~,T~a l y has a predetermined maxima l packet: t r ~ar7 rr~ r ~.:; r .>rr r-,~ ctuency, ,r first value, and a second valuf~, a rrrr:tto:act .,f' c::ontrolling transmission such that: data pac:ket:.~, are rrc~t transmitted 1,a with a frequency ex~.~eeding a a:h~arzr~m:l.'s pre~t~~t~m:mined maximal packet t:ransrri:i.ss i cn F_ .r c:c:~ut r~c:y c:;~.~rnpr.L; i.rug t he steps of changing the first. value c~t~~ r:caxt: ;:rye: ~:~f trie channels by a first step, wherein t_he ( ir-sr: s-c~.i ur:> area changed at each of successive first tune iraterv<:r_i.; ~:~nr~1 iru ~i c:ycl:ic:
order, for each data packet a:~ x~i.virrc,~ c>n tl-r~ trar~smis.;ion line, determining a r_~hanne.l. nr.rmt,er ~>i ~:~:-, aerri.vir~.g channel carrying the arriving data pac:kFt , arc=t ~:r:mi.7~::i. rod, fc>r Eeach arriving channel, a third va.lr:e::~ bra=e~r~ c:;n t.~e arriving channel's first val~_re and c>n i tm~ ,r~r:Lim :~f >raF: c:>f (a) a i~otal number of dat<~ packets l.rmt t:~a.v~e ,:~rrr it-E~d c.~n t:hEe transmission line sin~~~e t:rne F:i r: st: value ilf t he arriving channel was last chanced, and ~~;~) wli~~rr ~Lac~ c ~:~~,ive data packets arrive during suco:ess:iv~::~ F:c~c;n~ t: i:n~> irotervals, a total number of secon<.~ t~irne ir;t c,~ru~a l:> =i.:~l:~ >e>c:~ since the first value of the arrrivirug ~,:l;~.vrarr~:1 w~~~~ Lc:~ ~t: r::hanc~e~d, wherein the third values c::>f d:i.f:Eez:ert:. arr:iu~inc~ c:hann~els are based on the same one ot: (a) Gar-r;t ~;b) , rr=asps-~c:t ively comparing each third value to eac:r~~ ._~:r_ri 4~ir: ~ c°k~aranel' ~:
second v<~lue and, based c>n 1. h<~ c;c~nap<~r.i s,on, L:~cer forrninq one of the steps of perrn:i..tti ng t..:: ar5s4n i >~,~ :~srv o t t tn a re~spect.ive arriving data packet, dit>c;ard:i.ng t:lm_~ r~e:=ape::t p.ve arriving data packet, and marking the rerap~yrt iva ar:rivz.ng data packet.
The present invention also prc>vidc=.;~ a df=viE:e for controlling transmi;~sion o.f dat~.~ ~>ac~:ket.-, orx a transmission line having a plurality c7f ch~:ur:~rrei.:su~t:l~c tt-rat: data packets are not transmitted with r~ frE::~c~raerrc,y ctrk=wate~xm t han each channel's respective rnaxirna~. f ~~~_W :t: t.m~~rr:~m ~~;:ic>n frec(uency, wherein each channe.l_ has a re:~~~~~e.cti.ve. ~~I~ran_leO num)v>er, comprising means for st::oring ~,~ f: i.r.~t v~.~Lut~ arid a second value for each channe:.t., means f:c:>r <:r~ang i raid t'h~e f first value of i3 channel by a f:i rst st ep, t: he f. i r~>t~ va.iues c:f 3b succ;essive next ones of the c:lo~~or~f'1.:.; k>c=i.r~~g c::k~«Zr~ged during successive first time intervals crick i.n :3 cv;-c_:1 ic; order, means for determining, for each d~~ta p3~~ke~. arriving on the transmission line, a c:knannel rauzni'ac-~r c:af <~ra arriving channel carrying the arriving dat~r:a c3acx:c-at: a~cnd L :a:r clc~k.c::~rrr~.ining, for each arriving channel, a third vuTa:k~.~e ~a;i ee-i c>n the respective arriving channel's first v,~ l acs arid on the ~.~alue of one of (a) a total number :.;f axata l:o~~c~ke=~ta that haYJe arrived on the t.rant;m:ission l.i.t~e~ ~::sir;c.:e ';.r~~:; f::i_rst va:Luc:~
of the arriving channel. wa~~ 7. a:~t: c:lr;:~rug~d, rad (k: when successive data packets arrive during ::>~.~c:cc.-ssive second time intervals, a tot;jl rzurnber of seca.mu,i t.:nre intervals elapsed since the ~~i.r;:~t vaL;.ze c, lt:h~~ <~nr: i..v:~nc~ channel w<is last: changed, wherein the third v~:a:l_ue:~ ~;~f: ct:i..fferent arriving channels are based on ths~ samEa c::~nF:~ of (a) and (b) , a cc>mparator for comparing a thi. ~-c.z va LuE.r tc: t:he respective arriving channel's Eec;orud v<~lu~a; a:.~nG~ zr;F~ ~~s, k~~rse~.~l ors the?
com~aarison, for pai.~fo.x:m:iruc:l c>na c>t: t:it~~ k!~ra<:vti.ons ~:~f permitting transmi;~sic~n of t=he r~aK>pe~~tt ~re a;r.ri.vi.ng dat:.a packet discarding the respective a.rriv7rrg data packet, and marking the respective arr wing cl<lt.a p,.~:::1=st:;t .
The present invention also prow .ekes ~~ mE:thc:d of contrc~>ll.ing data packet transmission on a transrn:iss.ion lizie comprising the steps of providirug a plurality of channels, each channel having a maximal prac:ket t,.r:rrl:~rna ~ ~~ei.c:n f :r~:quc=n<:i~~, a channel number, a c;o~~r~t_ value, ar~.c~ ~~ t:Iuz E:~~>rlc:;~l.c'1 v~:~luc;, changing the count value c.~f eack: of tkue channels in succession, one count value be:izzc~ ~~;tnan«E~~d curing each successive first time intc~rv~L~1~ ~:ac::t~:~z-m:ir:~.i.r;g, for a<~<.h data packet arriving on then tr~~zznt;rni:_~:~:i.c;rl l.ir~e, thc~ channel number of an arriving channel. which i..r: c°arrying t:he arriving data packet, generating, Eor each: :~r.r_ivinq channel, a frequency vfalue i::as~:r on t':Y.cv~ s:rrriv-ing ctnanr;e7.'s court va:l_ue and channel rn.~ml~e:r ~,r~c~ can t';'~~e ~~hannel rrurr~ber corresponding to the count= val.rrF~ tnc:~~st x:wce:lt:l y c:harEgE:d in the changing step; arid cc}rnpar.ing, ~'cox: =~,ach ~;rwi~irrc~
channel, the frequency value ~~rith true r~-~sp~;ct_ive acrriving channel's threshold value and,, k:>asc.c:~ ors t:h+.~ c°c>mpari.son, per:Lorming one o.f the steps o:f: pc:'r:rr~it t i.ng t xvrrrsmi.ssion of the .respective arriving data ~~<.rr..:kc:~t, cW ::~c~<~:~c: sruq the respect ive arriving data ~?ac: kc.:t: ; ~~.r~d r7ia r k:i ;tcr t: he rest>ect ivc:~
arriving data packer .
BRIEF DESCRIPTION OF ':E'HE L)fZAW:Ir~C.
The invention will now be descr~iir~ :ct w:~ t i~ r ~tc-~r-ence to the accompanying drawings, ors whic to -Fig. 1 schematical_Ly she>ws wtr~e:r:e ,~ cif:v i ~:e ,3c:eording i=a the invention may be loc~at ed, -Fic:~. 2 schematical Ly shawl a device f r,~,z c~~rzying out the invention, -Figures 3 and 4 schernatical:I~~ ~~.h~::>w two ca l c-~;ul.ating steps according to the invention, -Figures 5-7 show at3 example rnade fc>r _.~~~x.t::rir~ channel in twhe method according to ttae in~rc~nt:::i_on, when-c. the latter two E,igures show the variation of t:hne c.:our3t vai.iac: and the variation of the ca_Lc~.rlated v~~7~.ze rc.:~;~»~ vt:infely.
WC) 93/18600 ~ .~ '~ &~ ~ .~~ ,~ f~f/S1E93/00207 s,.. a. ~~ ;: _, ; .
In Fig. 1 is schematically shown how a number of data packets, the channel numbers of which are indicated in the drawing symbols of the data packets, arrive to a device 2. according to the invention. In the device according to the invention certain data packets or cells are discarded (lostj or marked, as is illustrated for the packet Craving the channel number n in the Figure. The mark indicates, that the packet can be discarded.
The device according to the invention has the purpose of prohibiting that a data network is overloaded, for instance in particular that a switch unit 3 is flooded by arriving cells.
In Fig. 2 is illustrated in more detail how the device according to the invention is constructed and operates. For a transmission line 5 a buffer memory or register 7 is provided, which can contain maximally one data packet or one cell. When a data packet arrives to the buffer 7, information is fetched from said, packet comprising the channel number n of the packet. This channel number n is then processed by logical circuits 9 in a monitoring unit 11. 6dhen a data packet arrives to the buffer 7, thus a calculation loop is started inside these logical circuits 9.
The monitoring unit 11 also contains several memories or registers for various items used by the logical circuits 9. In these memories or registers thus can be found the stored number n of the channel of the present data packet and a channel number m indicating the number of a channel, the first count value Lm of which was latest changed. Further, information is provided on the tatal number (Mj of channels, which the monitoring unit 11 is capable of treating. For each channel also a threshold value Ti is recorded, a modification value or step value Di, an increment or addition value Ii, and the cou~at value or the first value Li. i here means the channel number and runs from ~. up to and including M. The parameters Ti, Di, and Ii are :individual or particular gor each channel in the general case.
b~aen a data packet arrives to the buffer 7, a signal thereof and information on the channel number n of the packet are sent to dV~ 93/18600 PCf/SE93/00207 the logical circuits 9. These circuits then start the following calculation sequence. In the block 13 a value Fn for the channel is calculated. This value depends on the channel number and also on the number of channels, which are left and the values of which are to be changed in steps until the value for exactly this channel shall be changed or decremented. This calculated value Fn is then compared in the block 15 to a threshold value Tn for the channel n and if the threshold value is smaller than this ~:alculated value Fn, the packet is discarded and a suitable signal is in the block 16 sent to the buffer 7, this signal having the meaning, that the buffer shall not forward the packet. If the comparison in the block 15 instead receives the answer No, the packet is accepted and is to be forwarded and in tine block 17 a signal thereof is sent to the buffer 7. Then the count value Ln for this channel is changed in the block 1~.
The transmission of packets on the line 5 is supposed to be regularly divided in time in such a way, that packets are sent or arrive on fixed, recyularly repeated times or in the corresponding periods between said times, that the transmission takes always place in definite time slots. The time period between these repeated times can also be termed a cell time or cell period. With a small load on the network packets do not arrive in each time slot. Further, for each time slot or cell period, in the block 21 the value of a variable m is changed, this variable indicating the number of that channel, for which the count value Lm was latest decremented. The change can be performed for instance by incrementing m by one. when this is performed for this next channel in turn having number m, the first count value Lm belonging thereto is modified in the block 23. When these steps have been performed, the process will await the beginning of the next cell period. For a full load on 'the line 5 this condition is the same as the arrival of a new data packet to the buffer 7.
In Fig. 3 is illustrated in detail the calculation of Fn in the block 13. In a bloc~.k 25 the present number m, which indicates the number of the channel, the count value Lm of which was latest changed, is thus compared to the channel number n of the WO 93/18600 PC'f/SE93/00207 s -y - .~ c. a .J ,,' ..
received data packet and if 'this latter value n is less than or equal to the number m, in the block 27 Fn is calculated as Ln Dn'(m - n). If instead the comparison receives the result N~, in the block 29 the calculation of Fn is performed by means of the e~cpression Ln - Dn' (rI + m - n) , These two cases are necessary, since the channel number m for the channel, for which the count value has been latest decremented, for instance cyclically runs through the values U, 1, 2, ..., t~i - 1. The expression inside the brackets in the blocks 27 and 29 will thereby indicate the number of channels, the count values of which remain to be decremented, before the count value of the channel having the number n belonging to the present cell is to be decremented.
Tn Fig. 4 is illustrated in detail how the change of the count value Ln in block 19 is performed. F3ere first in a block 31 it is compared, if the calculated value Fn is larger than or equal tn zero. If the answer from the comparison is X1ES, in the block 33 the count value Ln is decremented with the step In, which is stored in the monitoring unit r1. If the answer from the comparison is IdO, in a block 35 the channel number m for the channel, the count value Lm of which has been latest decremented, is compared to the channel number n of the present cell. If the former channel number m is larger than the channel number of the present cell n, 'the block 37 is performed, where a new count value is provided by the ea~pression In + Dy (r - n), while in the opposite case the count value is given the value 7Cn + Dn' (P~ + m - n) in the block 39.
In the latter case, when the calculated value Fn thus is smaller than zero, the new values of the count value Ln axe set in such a way, that a new value Fn calculated by means of theses values is given the value zero.
Below is given, formulated in pseudo code, the process which is performed each tire, when a data pac~Cet arrives to the ronitoring unit 3.
P~fi'/S E93/00207 W~ 93/18600 N .,G t~ ~~ ;.J r.~ ..
Process begin =.
repeat wait (at) if(cell arrived) begin if m >~ n then Fn : Ln ° Dn~(m ° n)% < The real level value else . for channel n is Fn : Ln - Dn*(M + ~a - n)% . interpolated.
if Fn >= 0 then . If Fn < Tn the cell if Tn -~ Fn >= 0 then , will be accepted begin . and the level value Ln '-- Ln + zn% - is increased by the "Accept cell'° . value In.
end else "Discard .cell°' else . Tf Fr1 < 0, some begin . error is present, if m >= n then . since the real level Ln : In + Dn*(m - n)% . value cannot be else , below zero. Ln is Ln : In + Dn ~ (M -~ m , given a value ~rhich n % . causes that Fn ~Cu aeACCept Cell~ . given the Value end . zero.
end m : (m -~ 1)P~IOD Pi; . The variable m taDces Lm := Lm M*Dm; . care of the channel if L~ < 0 then . ~rhich is to be Lm : 0% . decremented. m is forever . incremented in the end . channel number . order.
wo v~ns~oo ~c°ros~y~eoo2o7 ~~,~;~~~~;c~, ~xAZ~L~
The difference between the count value Ln and the calculated level value Fn for a channel will be elucidated with an example illustrated in Figures 5 - 7. We will consider that 10 cells arrive sequentially all having the same channel number, say channel number Z, followed by 10 cells having some other channel number and thereafter further 12 cells all having the channel number 1, as appears from Fig. 5.
Figures 6 and 7 show how the count value hl, which also can be termed the fictive level value, and the calculated or "real' level value Fn respectively will vary, if the input signal on the line ~ is such as shown in Fig. 5. The numbers in the drawings are calculated for the following values of the variables belonging to channel 1:
The number of channels, which can be monitored by the device, is set to 32. In real cases the number of channels, which can be monitored by the device, will be significantly larger.
- DZ: The channel dependant decrement value D1 for channel 1 is set to ~.
' Il: The increment value I1 is set to 28600.
- T1: The threshold value is set to a very high value.
The following process will then be obtained and is first illustrated in Fig. ~, showing the variation of the count value L1, when cells arrive. The states arc numbered witty the channel number m for the count value Ln which has been latest decreased:
~t = 1, 2, ... 10: Since the threshold value T1 is very large, all of the 5.0 first cells will be accepted. The level count L1 is increased by the value I1 for each cell, which has the channel number 1. and which fs accepted. After the first 10 cells have arrived, Ll has thus obtained the value 10°I1 (=286000).
- m ~ 11, 12, ..,, 20: The following 10 cells have not channel number 1 and thus L1 is not influenced.
' m = 21, 22, ..., 31, 0: When the first cell having channel number 1 arrives (m = 21), it is discovered that the real level value Fl is smaller than zero (see Fig. 7). In order to correct it in such a way, that this value F1 actually will indicate a avo 9si~~soa ~~-'ri~~93i~z~7 g .~ . t: ~ :~ :_ value, which corresponds to the level value in the known algorithm, L1 is given a value (T1 -~ (m - n)°~31), which makes the real level value F1 equal to zero. After that Ll is increased by T1 for each cell, which arrives and has channel number 1. 6dhen the variable m again takes the value 1, channel 1 is decremented by the value M°D1.
- m = 1: 45hen m = 1, the count value L1 of channel 1 is to be decremented by M'D1.
Tn Fig. 7 is shown the variation of the real or the calculated level value F1 plotted against the number of arrived cells. The process is as follows:
- m = 1, 2, ..., l0: The first 10 cells are accepted by the process. For each accepted cell a value T1 is added to L1. At the same time the variable m is increased by one, ~rhich results in F1 being decreased by D1 for each arriving cell.
- m = 11, 12, ..., 20: These cells have not channel number 1 and thus L1 is not influenced. Tnstead F1 is continuously decreased by the leakage value D1.
- m = 21, 22, ..., 31, 0: When the first cell having channel number 2 arrives, the real value F1 is calculated. Then it is discovered, that F1 has a value smaller than zero (_ --3.07260).
~y giving L1 the value T1 ~- (m ° n)'D1 F1 is given the value zero.
- m = 1: Then m = 1 L1 is decremented by ~d'D1. This gives F1 =
Lx.
Claims (33)
1. In a packet data communication system comprising a transmission line on which data packets are transmitted on a plurality of channels, wherein each channel respectively has a predetermined maximal packet transmission frequency, a first value, and a second value, a method of controlling transmission such that data packets are not transmitted with a frequency exceeding a channel's predetermined maximal packet transmission frequency comprising the steps of:
changing the first value of a next one of the channels by a first step, wherein the first values are changed at each of successive first time intervals and in a cyclic order;
for each data packet arriving on the transmission line, determining a channel number of an arriving channel carrying the arriving data packet;
determining, for each arriving channel, a third value based on the arriving channel's first value and on the value of one of:
(a) a total number of data packets that have arrived on the transmission line since the first value of the arriving channel was last changed, and (b) when successive data packets arrive during successive second time intervals, a total number of second time intervals elapsed since the first value of the arriving channel was last changed, wherein the third values of different arriving channels are based on the same one of (a) and (b);
respectively comparing each third value to each arriving channel's second value and, based on the comparison, performing one of the steps of:
permitting transmission of the respective arriving data packet, discarding the respective arriving data packet, and marking the respective arriving data packet.
changing the first value of a next one of the channels by a first step, wherein the first values are changed at each of successive first time intervals and in a cyclic order;
for each data packet arriving on the transmission line, determining a channel number of an arriving channel carrying the arriving data packet;
determining, for each arriving channel, a third value based on the arriving channel's first value and on the value of one of:
(a) a total number of data packets that have arrived on the transmission line since the first value of the arriving channel was last changed, and (b) when successive data packets arrive during successive second time intervals, a total number of second time intervals elapsed since the first value of the arriving channel was last changed, wherein the third values of different arriving channels are based on the same one of (a) and (b);
respectively comparing each third value to each arriving channel's second value and, based on the comparison, performing one of the steps of:
permitting transmission of the respective arriving data packet, discarding the respective arriving data packet, and marking the respective arriving data packet.
2. The method of claim 1, wherein the first time intervals are such that the first values are change upon the arrivals of the data packets.
3. The method of claim 1 or 2, wherein the first time intervals and the second time intervals are equal.
4. The method of claim 1, 2 or 3, wherein the third value of an arriving channel is a change of the first value of the arriving channel is a change of the first value of the arriving channel having a sign that is different from a sign of the first step.
5. The method of any one of claims 1 to 4, wherein each channel has a respective first step.
6. The method of any one of claims 1 to 5, wherein only one first value is changed at each first time interval and the first step is based on a product of the total number of channels and a fourth value.
7. The method of any one of claims 1 to 6, wherein, when a changed first value has a sign that is different from its sign before being changed, the changed first value is changed to zero.
8. The method of claim 6, wherein each channel has a respective fourth value.
9. The method of any one of claims 1 to 7, wherein the determination of the third value is also based on a fourth value.
10. The method of claim 9, wherein each channel has a respective fourth value.
11. The method of any of claims 1 to 7, wherein the third value is one of a sum of and a difference of the first value of the arriving channel and a value derived from a predetermined function of one of (a) the total number of data packets that have arrived on the transmission line since the first value of the arriving channel was last changed and, (b) when successive data packets arrive during successive second intervals, the total number of second time intervals elapsed since the first value of the arriving channel was last changed; and the value derived from the predetermined function for different arriving channels is based on the same one of (a) and (b).
12. The method of claim 11, wherein the value derived from the predetermined function is a product of a fourth value and one of (a) the total number of data packets that have arrived on the transmission line since the first value of the arriving channel was last changed and, (b) when successive data packets arrive during successive second time intervals, the total number of second time intervals elapsed since the first value of the arriving channel was last changed; and the value derived from the predetermined function for different arriving channels is based on the same one of (a) and (b).
13. The method of claim 12, wherein each channel has a respective fourth value.
14. The method of any one of claims 1 to 13, wherein the first value of an arriving channel is changed by a second step when transmission of the respective arriving data packet is permitted.
15. The method of claim 14, wherein each channel has a respective second step.
16. The method of claim 14, wherein the first step and the second step have different signs.
17. The method of any one of claims 1 to 16, wherein, when a third value has a sign that is different from a sign of the respective arriving channel's first value, that first value is changed such that a third value determined based on the changed first value is zero.
18. The method of any one of claims 1 to 17, wherein, when the channels are organized in groups, the first values of the channels in a next one of the groups are changed by the first step at each of successive first time intervals, the groups being changed in a cyclic order.
19. A device for controlling transmission of data packets on a transmission line having a plurality of channels such that data packets are not transmitted with a frequency greater than each channel's respective maximal packet transmission frequency, wherein each channel has a respective channel number, comprising:
means for storing a first value and a second value for each channel;
means for changing the first value of a channel by a first step, the first values of successive next ones of the channels being changed during successive first time intervals and in a cyclic order;
means for determining, for each data packet arriving on the transmission line, a channel number of an arriving channel carrying the arriving data packet and for determining, for each arriving channel, a third value based on the respective arriving channel's first value and on the value of one of:
(a) a total number of data packets that have arrived on the transmission line since the first value of the arriving channel was last changed, and (b) when successive data packets arrive during successive second time intervals, a total number of second time intervals elapsed since the first value of the arriving channel was last changed, wherein the third values of different arriving channels are based on the same one of (a) and (b);
a comparator for comparing a third value to the respective arriving channel's second value; and means, based on the comparison, for performing one of the functions of permitting transmission of the respective arriving data packet discarding the respective arriving data packet, and marking the respective arriving data packet.
means for storing a first value and a second value for each channel;
means for changing the first value of a channel by a first step, the first values of successive next ones of the channels being changed during successive first time intervals and in a cyclic order;
means for determining, for each data packet arriving on the transmission line, a channel number of an arriving channel carrying the arriving data packet and for determining, for each arriving channel, a third value based on the respective arriving channel's first value and on the value of one of:
(a) a total number of data packets that have arrived on the transmission line since the first value of the arriving channel was last changed, and (b) when successive data packets arrive during successive second time intervals, a total number of second time intervals elapsed since the first value of the arriving channel was last changed, wherein the third values of different arriving channels are based on the same one of (a) and (b);
a comparator for comparing a third value to the respective arriving channel's second value; and means, based on the comparison, for performing one of the functions of permitting transmission of the respective arriving data packet discarding the respective arriving data packet, and marking the respective arriving data packet.
20. The device of claim 19, wherein the first values are changed upon the arrivals of the data packets.
21. The device of claim 19 or 20, wherein the first time intervals and the second time intervals are equal.
22. The device of claim 19, 20 or 21, wherein the third value of an arriving channel is change of the first value of the arriving channel having a sign that is different from a sign of the first step.
23. The device of any one of claims 19 to 22, wherein the storing means also stores a respective first step for each channel.
24. The device of any one of claims 19 to 23, wherein the storing means stores a fourth value for each channel and the changing means changes only one first value at each first time interval and the first step is based on a product of the total number of channels and the fourth value.
25. The device of any one of claims 19 to 24, wherein when a changed first value has a sign that is different from its sign before being changed, the changing means changes the changed first value to zero.
26. The device of any one of claims 19 to 25, wherein the storing means also stores a fourth value for each channel and the determining means determines the third value also on the basis of the fourth value.
27. The device of any one of claims 19 to 26, wherein the determining means determines one of a sum of and a difference of a first value of the arriving channel and a value derived from a predetermined function of one of (a) the total number of data packets that have arrived on the transmission line since the first value of the arriving channel was last changed and, (b) when successive data packets arrive during successive second time intervals, the total number of second time intervals elapsed since the first value of the arriving channel was last changed; and the value derived from the predetermined function for different arriving channels is based on the same one of (a) and (b).
28. The device of claim 27, wherein the storing means also stores respective fourth values for the channels and the value derived from the predetermined function is a product of the respective fourth value and one of (a) the total number of data packets that have arrived on the transmission line since the respective first value of the arriving channel was last changed and, (b) when successive data packets arrive during successive second time intervals, the total number of second time intervals elapsed since the respective first value of the arriving channel was last changed; and the valve derived from the predetermined function for different arriving channels is based on the same one of (a) and (b).
29. The device of any of claims 19 to 28, wherein the first value of an arriving channel is changed by a second step when transmission of the respective arriving data packet is permitted.
30. The device of claim 29, wherein the first step and the second step have different signs.
31. The device of any one of claims 19 to 30, wherein the changing means, when a third value has a sign that is different from a sign of the respective arriving channel's first value, changes that first value such that a third value determined based on the changed first value is zero.
32 The device of any one of claims 19 to 31, wherein, when the channels are organized in groups, the changing means changes the first values of the channels in successive next ones of the groups during successive first time intervals, the groups being changed in a cyclic order.
33. A method of controlling data packet transmission on a transmission line comprising the steps of:
providing a plurality or channels, each in channel having a maximal packet transmission frequency, a channel number, a count value, and a threshold value;
changing the count value of each of the channels in succession, one count value being changed during each successive first time interval;
determining, for each data packet arriving on the transmission line, the channel number of an arriving channel which is carrying the arriving data packet;
generating, for each arriving channel, a frequency value based on the arriving channel's count value and channel number and on the channel number corresponding to the count value most recently changed in the changing step; and comparing, for each arriving channel, the frequency value with the respective arriving channel's threshold value and, based on the comparison, performing one of the steps of:
permitting transmission of the respective arriving data packet;
discarding the respective arriving data packet; and marking the respective arriving data packet.
providing a plurality or channels, each in channel having a maximal packet transmission frequency, a channel number, a count value, and a threshold value;
changing the count value of each of the channels in succession, one count value being changed during each successive first time interval;
determining, for each data packet arriving on the transmission line, the channel number of an arriving channel which is carrying the arriving data packet;
generating, for each arriving channel, a frequency value based on the arriving channel's count value and channel number and on the channel number corresponding to the count value most recently changed in the changing step; and comparing, for each arriving channel, the frequency value with the respective arriving channel's threshold value and, based on the comparison, performing one of the steps of:
permitting transmission of the respective arriving data packet;
discarding the respective arriving data packet; and marking the respective arriving data packet.
Applications Claiming Priority (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| SE9200775A SE470002B (en) | 1992-03-13 | 1992-03-13 | A method for preventing the transmission of data packets with a higher intensity than a predetermined value for the channel and a device for carrying out the method on one of a number of channels on a common transmission line |
| SE9200775-6 | 1992-03-13 | ||
| PCT/SE1993/000207 WO1993018600A1 (en) | 1992-03-13 | 1993-03-10 | A method and a device for monitoring channel split data packet transmission |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CA2130356A1 CA2130356A1 (en) | 1993-09-14 |
| CA2130356C true CA2130356C (en) | 2004-02-17 |
Family
ID=20385609
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CA002130356A Expired - Fee Related CA2130356C (en) | 1992-03-13 | 1993-03-10 | A method and a device for monitoring channel split data packet transmission |
Country Status (17)
| Country | Link |
|---|---|
| US (1) | US5361252A (en) |
| EP (1) | EP0630541B1 (en) |
| JP (1) | JP3294847B2 (en) |
| KR (1) | KR100228500B1 (en) |
| CN (1) | CN1057416C (en) |
| AU (1) | AU682820B2 (en) |
| BR (1) | BR9306049A (en) |
| CA (1) | CA2130356C (en) |
| DE (1) | DE69326830T2 (en) |
| DK (1) | DK0630541T3 (en) |
| ES (1) | ES2139653T3 (en) |
| FI (1) | FI944188A0 (en) |
| GR (1) | GR3032206T3 (en) |
| MX (1) | MX9301324A (en) |
| NO (1) | NO943378D0 (en) |
| SE (1) | SE470002B (en) |
| WO (1) | WO1993018600A1 (en) |
Families Citing this family (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| SE515178C2 (en) * | 1992-03-20 | 2001-06-25 | Ericsson Telefon Ab L M | Procedures and devices for prioritizing buffer management in packet networks |
| SE503589C2 (en) * | 1994-02-10 | 1996-07-15 | Ericsson Telefon Ab L M | Method and apparatus for monitoring a memory |
| GB9405406D0 (en) * | 1994-03-18 | 1994-05-04 | Netcomm Ltd | Atm cell switch |
| JP3309020B2 (en) * | 1994-06-20 | 2002-07-29 | 富士通株式会社 | Transmission throughput control apparatus and method |
| US5594729A (en) * | 1994-09-23 | 1997-01-14 | Lucent Technologies Inc. | System and method for employing single-bit feedback control within a variable bit rate data transmission network |
| US5513174A (en) * | 1994-12-07 | 1996-04-30 | At&T Corp. | Telecommunication system with detection and control of packet collisions |
| NL9500585A (en) * | 1995-03-27 | 1996-11-01 | Nederland Ptt | ATM policing body. |
| US5751964A (en) * | 1995-09-12 | 1998-05-12 | International Business Machines Corporation | System and method for automatic determination of thresholds in network management |
| FI100155B (en) | 1995-11-09 | 1997-09-30 | Nokia Telecommunications Oy | Traffic measurement in a data communication system |
| FI100158B (en) * | 1995-11-09 | 1997-09-30 | Nokia Telecommunications Oy | Traffic control in a telecommunication system |
| US5963542A (en) * | 1996-09-03 | 1999-10-05 | The United States Of America As Represented By The Secretary Of The Navy | Asynchronous transfer mode cell loss estimator |
| KR100250659B1 (en) * | 1997-04-03 | 2000-04-01 | 강병호 | Apparatus for selecting and displaying channel during no.7 signalling monitor |
| US6108736A (en) * | 1997-09-22 | 2000-08-22 | Intel Corporation | System and method of flow control for a high speed bus |
| US7107371B1 (en) | 1997-09-22 | 2006-09-12 | Intel Corporation | Method and apparatus for providing and embedding control information in a bus system |
| US6088370A (en) | 1997-09-22 | 2000-07-11 | Intel Corporation | Fast 16 bit, split transaction I/O bus |
| US6256308B1 (en) | 1998-01-20 | 2001-07-03 | Telefonaktiebolaget Lm Ericsson | Multi-service circuit for telecommunications |
| JP2004364217A (en) * | 2003-06-09 | 2004-12-24 | Matsushita Electric Ind Co Ltd | Packet communication apparatus |
| CN107682423A (en) * | 2017-09-22 | 2018-02-09 | 深圳市盛路物联通讯技术有限公司 | A kind of data forwarding method and device |
Family Cites Families (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4734907A (en) * | 1985-09-06 | 1988-03-29 | Washington University | Broadcast packet switching network |
| US4769810A (en) * | 1986-12-31 | 1988-09-06 | American Telephone And Telegraph Company, At&T Bell Laboratories | Packet switching system arranged for congestion control through bandwidth management |
| FR2616025B1 (en) * | 1987-05-26 | 1989-07-21 | Lespagnol Albert | METHOD AND SYSTEM FOR PACKET FLOW CONTROL |
| FR2616024B1 (en) * | 1987-05-26 | 1989-07-21 | Quinquis Jean Paul | SYSTEM AND METHOD FOR PACKET FLOW CONTROL |
| DE3732937A1 (en) * | 1987-09-30 | 1989-04-20 | Philips Patentverwaltung | CIRCUIT ARRANGEMENT FOR AVOIDING OVERLOAD IN A BROADBAND SWITCHING SYSTEM |
| SE462360B (en) * | 1988-10-28 | 1990-06-11 | Ellemtel Utvecklings Ab | PROCEDURE AND DEVICE TO PREVENT IT TO BE SENDED ON A COMMON TRANSFER LINK HIGH-INTENSITY DATA PACKAGE WITH A PREDICTED VALUE DETERMINED AT A PREDICTED VALUE |
| AU622424B2 (en) * | 1988-11-10 | 1992-04-09 | Alcatel N.V. | Statistical measurement equipment and telecommunication system using same |
| NL8900269A (en) * | 1989-02-03 | 1990-09-03 | Nederland Ptt | METHOD FOR TRANSMISSION OF TIME-DISTRIBUTED TRANSMISSION CHANNELS OVER A STAGE OF DATA CELLS THROUGH A MULTIPLE OF ASYNCHRONOUS Maintaining a counter reading per transmission channel, which depends on the number of data cells per time. |
| FR2643532B1 (en) * | 1989-02-17 | 1991-05-10 | France Etat | METHOD FOR RESERVING RATES AND TIME SWITCHES OF ASYNCHRONOUS PACKETS |
| JPH02220531A (en) * | 1989-02-22 | 1990-09-03 | Toshiba Corp | Call connection control system and flow monitor system |
| NL8900640A (en) * | 1989-03-16 | 1990-10-16 | At & T & Philips Telecomm | METHOD FOR TRANSMITTING DATA PACKAGES IN ATD (ASYNCHRONOUS TIME DIVISION) AND AN APPARATUS FOR APPLYING THIS METHOD |
| FR2648649B1 (en) * | 1989-06-20 | 1991-08-23 | Cit Alcatel | METHOD AND DEVICE FOR QUANTIFIED EVALUATION OF THE FLOW RATE OF VIRTUAL CIRCUITS EMPLOYING AN ASYNCHRONOUS TIME MULTIPLEXING TRANSMISSION CHANNEL |
| US5115429A (en) * | 1990-08-02 | 1992-05-19 | Codex Corporation | Dynamic encoding rate control minimizes traffic congestion in a packet network |
| US5216670A (en) * | 1991-07-03 | 1993-06-01 | International Business Machines Corporation | Message stripping protocol for a communication network |
-
1992
- 1992-03-13 SE SE9200775A patent/SE470002B/en not_active IP Right Cessation
-
1993
- 1993-03-10 MX MX9301324A patent/MX9301324A/en not_active IP Right Cessation
- 1993-03-10 EP EP93906932A patent/EP0630541B1/en not_active Expired - Lifetime
- 1993-03-10 CA CA002130356A patent/CA2130356C/en not_active Expired - Fee Related
- 1993-03-10 JP JP51559893A patent/JP3294847B2/en not_active Expired - Fee Related
- 1993-03-10 KR KR1019940703202A patent/KR100228500B1/en not_active IP Right Cessation
- 1993-03-10 AU AU37716/93A patent/AU682820B2/en not_active Ceased
- 1993-03-10 BR BR9306049A patent/BR9306049A/en not_active Application Discontinuation
- 1993-03-10 DK DK93906932T patent/DK0630541T3/en active
- 1993-03-10 ES ES93906932T patent/ES2139653T3/en not_active Expired - Lifetime
- 1993-03-10 WO PCT/SE1993/000207 patent/WO1993018600A1/en active IP Right Grant
- 1993-03-10 DE DE69326830T patent/DE69326830T2/en not_active Expired - Lifetime
- 1993-03-12 US US08/030,724 patent/US5361252A/en not_active Expired - Lifetime
- 1993-03-13 CN CN93104492A patent/CN1057416C/en not_active Expired - Fee Related
-
1994
- 1994-09-12 FI FI944188A patent/FI944188A0/en unknown
- 1994-09-12 NO NO943378A patent/NO943378D0/en not_active Application Discontinuation
-
1999
- 1999-12-21 GR GR990403293T patent/GR3032206T3/en unknown
Also Published As
| Publication number | Publication date |
|---|---|
| DK0630541T3 (en) | 2000-04-17 |
| WO1993018600A1 (en) | 1993-09-16 |
| DE69326830T2 (en) | 2000-04-13 |
| FI944188A (en) | 1994-09-12 |
| CN1082286A (en) | 1994-02-16 |
| US5361252A (en) | 1994-11-01 |
| SE9200775L (en) | 1993-09-14 |
| DE69326830D1 (en) | 1999-11-25 |
| SE470002B (en) | 1993-10-18 |
| KR950701170A (en) | 1995-02-20 |
| FI944188A0 (en) | 1994-09-12 |
| AU3771693A (en) | 1993-10-05 |
| GR3032206T3 (en) | 2000-04-27 |
| SE9200775D0 (en) | 1992-03-13 |
| JP3294847B2 (en) | 2002-06-24 |
| BR9306049A (en) | 1997-11-18 |
| CN1057416C (en) | 2000-10-11 |
| JPH07504547A (en) | 1995-05-18 |
| KR100228500B1 (en) | 1999-11-01 |
| CA2130356A1 (en) | 1993-09-14 |
| EP0630541A1 (en) | 1994-12-28 |
| MX9301324A (en) | 1993-09-01 |
| NO943378L (en) | 1994-09-12 |
| NO943378D0 (en) | 1994-09-12 |
| ES2139653T3 (en) | 2000-02-16 |
| AU682820B2 (en) | 1997-10-23 |
| EP0630541B1 (en) | 1999-10-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CA2130356C (en) | A method and a device for monitoring channel split data packet transmission | |
| EP0381275B1 (en) | Method for transmitting, via a plurality of asynchronously time-divided transmission channels, a flow of data cells, the state of a counter for each transmission channel being kept up to date in accordance with the number of data cells per unit of time | |
| US8126010B2 (en) | Methods for dynamic bandwidth allocation and queue management in Ethernet passive optical networks | |
| EP1151556B1 (en) | Method of inverse multiplexing for atm | |
| US4901277A (en) | Network data flow control technique | |
| EP0577269A2 (en) | Arrangement for bounding jitter in a priority-based switching system | |
| WO2003049350A3 (en) | Optical network communication system | |
| JP3505118B2 (en) | Traffic shaper | |
| EP0772325B1 (en) | ATM exchange system and priority control method | |
| CA2094561A1 (en) | Method and circuit arrangement for checking the observance of prescribed transmission bit rates in an atm switching equipment | |
| EP0716557A3 (en) | Telecommunication system with detection and control of packet collisions | |
| JPH05219093A (en) | Method and circuit for polishing | |
| CN1170487A (en) | Traffic control in telecommunications system | |
| GB2181926A (en) | Network data flow control | |
| US7110365B2 (en) | Apparatus for analyzing performance of traffic in asynchronous transfer mode (ATM) switch and method thereof, and ATM switching system employing the same | |
| EP0909061A3 (en) | Transmission equipment and a load-distribution transmitting method in the transmission equipment | |
| JP3141880B2 (en) | Packet control device | |
| JPH04162847A (en) | Cell traffic monitor | |
| EP2416509B1 (en) | Method for multiplexing and demultiplexing digital data from a plurality of sources with an assigned nominal capacity. | |
| JPH0630026A (en) | Cell traffic monitoring device | |
| Bhardwaj et al. | ISDN frame relay and its congestion control | |
| Reid et al. | An analysis of space priority queueing in ATM switches | |
| Chang et al. | Analysis of an ATM multiplexer with correlated periodic real-time and independent batch Poisson nonreal-time traffic | |
| Durinovic-Johri et al. | A new threshold-based traffic characterization and policing mechanism for communication networks | |
| Wei et al. | A Compressed and Dynamic-range-based Expression of Timestamp and Period for Timestamp-based Schedulers |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| EEER | Examination request | ||
| MKLA | Lapsed |