CN1665204A - Method for real-time measurement of local service quality of computer network - Google Patents
Method for real-time measurement of local service quality of computer network Download PDFInfo
- Publication number
- CN1665204A CN1665204A CN2005100114446A CN200510011444A CN1665204A CN 1665204 A CN1665204 A CN 1665204A CN 2005100114446 A CN2005100114446 A CN 2005100114446A CN 200510011444 A CN200510011444 A CN 200510011444A CN 1665204 A CN1665204 A CN 1665204A
- Authority
- CN
- China
- Prior art keywords
- grouping
- measurement
- router
- packet
- loss rate
- 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.)
- Granted
Links
Images
Abstract
The invention discloses a method for real-time measuring local service quality state of a computer network, belonging to the field of computer network measuring technique, and characterized in that: it arranges a real-time measuring module on the top of each router, where the module can generate measuring packets and is provided with a calculating formula for parameters of service quality state and a data processing method for filtering temporary fluctuation in the measuring result; as the module operates, it calculates clock difference, pack transmission delay, usable bandwidth for oriented link and packet losing ratio by periodical timer and transmission and reception of measuring packet, and adds the calculated results to a sample space and makes data processing; the module can accurately measure the latest and stable parameters of service quality state, the measuring packets are a small burden on the link and the extendibility is better.
Description
Technical field
The method for real-time measurement of local service quality of computer network state belongs to the computer network field of measuring technique.
Background technology
The service quality route is a solution that is conceived to provide the service quality assurance in the Internet, and it is taken QoS parameter demands such as available bandwidth, transmission delay and Loss Rate into account when selecting route.It has been generally acknowledged that in the research of service quality route that each service quality router safeguards its local service quality state, and give other service quality router with this state propagation by the service quality Routing Protocol.Owing to be subjected to the influence of user behavior, link load and router state, the QoS state of network is difficult to estimate simply these parameters along with the time ceaselessly changes.Therefore, be necessary to carry out real-time network measure to obtain QoS state parameter accurately.
Comprise the extensive project of network at present among the Internet, as IEPM and Surveyor, it is very wide that it covers the region, and measurement result is very slowly to the feedback of tested network, can not be used for guide service quality route in real time." Low-cost precise QoS measurement tool " among the CESNETtechnical report adopts the mode of initiatively measuring to measure QoS performance end to end, by using the GPS technology, the measurement result of block transmission delay can be as accurate as several delicate." Integrating Active Methods and Flow Meters-an implementation usingNeTraMet " among the CAIDA is connected to a ntp server to carry out the synchronous of clock with each measurement point by extra transmission medium.If yet last two kinds of methods are deployed to each router in the territory, its cost may be excessive, and our scheme does not need to introduce third party device.In addition, although our packet loss rate measure with above-mentioned NTP scheme in flowmeter (trafficmeter) be similar, the NTP scheme is conceived to the monitoring system of independent operating, and the measuring system that our scheme is conceived to embed.
Summary of the invention
The object of the present invention is to provide the method for real-time measurement of a kind of local service quality of computer network state (being available bandwidth, block transmission delay and the packet loss rate on every link).
Method proposed by the invention is characterised in that: it contains following steps successively:
Step 1. real-time measurement module of every router deploy in network has defined following parameter at least in this module:
b
u: in the last bandwidth measurement cycle on average at dedicated bandwidth;
C: the capacity of tested local interface respective links, set point;
s
1: the moment of local router network interface card is left in clock difference measurements request grouping;
s
2: clock difference measurements request grouping arrives the moment of neighbor router network interface card;
s
3: clock difference measurements acknowledgment packet is left the moment of neighbor router network interface card;
s
4: clock difference measurements acknowledgment packet arrives the moment of local router network interface card;
t
1: the moment of the transmission measurement point of local router is left in the delay measurements grouping;
t
2: the delay measurements grouping arrives the moment of the reception measurement point of neighbor router;
p
1: at the last Loss Rate of neighbor router in measuring period, from the transmission measurement point of neighbor router to
The number of packet that local router sends;
p
2: successively receiving from neighbor router in two times between the adjacent Loss Rates measurement grouping,
The user data packets that the reception measurement point of local router receives from this neighbor router
Quantity;
Sequence number: be used to detect Loss Rate measurement grouping and self whether lose;
Measurement module can generate following grouping at least in real time:
Clock difference measurements request grouping: comprise parameter s at least
1With packet checks and;
Clock difference measurements acknowledgment packet: comprise parameter s at least
1, s
2, s
3With packet checks and;
Delay measurements grouping: comprise parametric t at least
1With packet checks and;
Loss Rate is measured grouping: comprise sequence number, parameter p at least
1With packet checks and;
At least be provided with following computing formula in real time in the measurement module:
The time clock correction: c=((s
1+ s
4)-(s
2+ s
3))/2;
Block transmission delay: d=t
2-t
1-c;
Oriented link available bandwidth: b
a=C-b
u
Packet loss rate: l=(1-p
2/ p
1) * 100%;
Comprise the data processing method that is used for filtering the measurement result transient swing in real time in the measurement module:
Median method is used for the data processing of block transmission delay, and its method is: sample space is sorted by numerical values recited, get median or two medians on average as statistics;
The index method of moving average is used for the data processing of oriented link available bandwidth and packet loss rate, and its computing formula is:
y
N＝a
N-1x
1+a
N-2(1-a)x
2+a
N-3(1-a)x
3+…+a(1-a)x
N-1+(1-a)x
N
Wherein, y
NBe N measurement result of having filtered transient swing, x
iBe i and do not cross the measurement result that filters transient swing, a is for reducing the tolerance of old sample weight in measurement result, 0＜a＜1;
Each measurement module of step 2. carries out the real-time measurement of local service quality of computer network state successively according to the following steps:
Step 2.1. reads in tested local interface number and the capacity of this interface respective links, the IP address of neighbor router;
Step 2.2. sets clock difference measurements cycle, bandwidth measurement cycle, delay measurements cycle, Loss Rate measuring period, and starts corresponding measurement timer;
Step 2.3. waits for that grouping arrives or timer expired;
If step 2.4. divides into groups to arrive, then carry out following steps successively:
Step 2.4.1. accumulative total block length and number of packet;
Step 2.4.2. is if be grouped into clock difference measurements acknowledgment packet, by the above-mentioned s that records
1, s
2, s
3, s
4Clock correction c forwards step 2.6 to during calculating;
Step 2.4.3. is if be grouped into the delay measurements grouping, by the above-mentioned t that records
1, t
2Calculate block transmission delay d, forward step 2.6 to;
Step 2.4.4. measures grouping if be grouped into Loss Rate, by the above-mentioned p that records
1, p
2Calculate packet loss rate l, forward step 2.6 to;
Step 2.4.5. to neighbor router tranmitting data register difference measurements acknowledgment packet, forwards step 2.3 if be grouped into clock difference measurements request grouping to;
Step 2.4.6. gives router and handles if be grouped into other grouping, forwards step 2.3 to;
If step 2.5. timer expiry is then carried out following steps successively:
Step 2.5.1. is if timer is the bandwidth measurement timer, by above-mentioned C, b
uCalculate oriented link available bandwidth b
a, forward step 2.6 to;
Step 2.5.2. to neighbor router tranmitting data register difference measurements request grouping, forwards step 2.3 if timer is a clock difference measurements timer to;
Step 2.5.3. measures grouping if timer is the delay measurements timer to the neighbor router transmission lag, forwards step 2.3 to;
Step 2.5.4. sends Loss Rate to neighbor router and measures grouping if timer is the Loss Rate measurement timer, forwards step 2.3 to;
Step 2.6. carries out data processing to result of calculation as stated above, forwards step 2.3 to;
Experimental results show that:
(1) local measurement can accurately be measured QoS state parameter (the bandwidth Kbps order of magnitude postpones 0.1 millisecond of order of magnitude, Loss Rate 0.001% order of magnitude), even the clock difference measurements that utilizes existing link also can operate as normal under the network congestion state;
(2) measurement result has reflected the network state of up-to-date (nearest 15 seconds);
(3) the most of transient swing of data processing in can screen obtains metastable measurement result;
(4) measure the burden very little (only take the bandwidth of 2Kbps) of grouping, large scale network is had good expandability link.
Description of drawings
Fig. 1. the method for real-time measurement flow chart of local service quality state;
Fig. 2. the graph model of having the right, Link State=(bandwidth postpones Loss Rate);
Fig. 3. clock correction during measurement;
Fig. 4. measure block transmission delay;
Fig. 5. measure grouping and user data packets;
Fig. 6. measure packet loss rate;
Fig. 7. the stratified random sampling;
Fig. 8. the relative pulse phase difference;
Fig. 9. the experiment topology;
Figure 10. block transmission delay measurement result (sample collection);
Figure 11. block transmission delay measurement result (data processing);
Figure 12. oriented link available bandwidth measurement result (sample collection);
Figure 13. oriented link available bandwidth measurement result (data processing);
Figure 14. packet loss rate measurement result (sample collection);
Figure 15. packet loss rate measurement result (data processing);
Figure 16. the clock level difference measurements.
Embodiment
Tested network is abstract to be the directed graph of having the right, and the state of router is included into the state of its each adjacent link.For fear of duplicate measurements, each router is only measured the link of neighbours to local direction.In Fig. 2, node A measures the state parameter of oriented link B → A and C → A.Real-time measurement in the service quality router should meet the following requirements:
Require one, accurately.Under existing technical conditions, the precision of local measurement should satisfy: the bandwidth Kbps order of magnitude postpones 0.1 millisecond of order of magnitude, Loss Rate 0.001% order of magnitude.
Require two, in real time.The path that the service quality router can satisfy user's QoS requirement according to known measurement of network state result selection.If the state parameter of network is too outmoded, the path of selection no longer satisfies the demands probably, thereby causes the decline of service quality.Therefore, the result of measurement should reflect the latest network state.
Require three, stable.Original measurement result has very strong fluctuation usually, and this selects route is disadvantageous.Need to obtain relative stability the variation tendency of reflection network through further handling to filter out the fluctuation of transient state.
Require four, hang down load and can expand.Measurement module continuous service each in network router, they should not produce excessive burden to router and link, should have autgmentability preferably to large scale network yet.
The available bandwidth of oriented link is meant a remaining transmittability of link.With the total length of bandwidth measurement all groupings that router receives in the cycle of accumulator statistics, calculate then on average at dedicated bandwidth, note is made b
uIf the known link capacity is C, available bandwidth b
aCan followingly calculate:
b
a＝C-b
u
Different with round-trip delay, the measurement of block transmission delay needs transmit leg and recipient that a timestamp respectively is provided.Because different router uses different clocks, must introduce certain mechanism with the relative time clock deviation of eliminating them (hereinafter to be referred as the time clock correction).In this programme, the time clock correction measurement be on the existing link between the neighboring router, to carry out.If the time that a certain moment router C1 and C2 show is respectively t
C1And t
C2, router C1 with respect to the time clock correction of C2 is:
c＝t
C1-t
C2
The time clock correction c method of measurement as follows: router C1 is at s
1Send one to C2 constantly and measure request, router C2 is at s
2Constantly receive the request grouping, and at s
3Send acknowledgment packet to router C1 constantly.Comprise time s in the acknowledgment packet
2And s
3Router C1 is at s
4Constantly receive acknowledgment packet, and clock correction (see figure 3) when calculating according to these four timestamps.Timestamp is defined as follows:
| Timestamp | Incident | Reference clock |
| ????s 1 | The network interface card of C1 is left in the request grouping | ????C1 |
| ????s 2 | The request grouping arrives the network interface card of C2 | ????C2 |
| ????s 3 | Acknowledgment packet is left the network interface card of C2 | ????C2 |
| ????s 4 | Acknowledgment packet arrives the network interface card of C1 | ????C1 |
Supposing that grouping is transferred to C2 and is transferred to the time that C1 consumed from C2 on transmission medium from C1 equates, is defined as medium transmission and postpones Δ s.The time clock correction c can followingly calculate:
Require emphasis timestamp s
1, s
2, s
3And s
4It is the moment that network interface card was left and arrived in grouping.These parameters also are not easy to measure, and this is a major reason that produces error.When realizing, should make measurement point as far as possible near transmission medium.Obviously, clock difference measurements agreement is based on data link layer.
On the basis of clock correction, the block transmission delay of link can be measured with the following method when known: router C2 sends one to C1 and measures grouping, has write down grouping in the grouping and has left the moment t that sends point
1Measure the receiving cache queue of grouping, arrive the reception measurement point (see figure 4) of C1 at last through transmission buffer queue, transmission medium and the C1 of C2.Timestamp is defined as follows:
| Timestamp | Incident | Reference clock |
| ????t 1 | Measure grouping and leave the transmission measurement point | ????C1 |
| ????t 2 | Measure grouping and arrive the reception measurement point | ????C2 |
If clock C1 is c with respect to the deviation of clock C2, then router C2 to the computing formula of the block transmission delay of C1 is:
d＝t
2-t
1-c
Because delay measurements is grouped on the user data path and transmits, the delay measurements agreement should realize on network layer.
The method of measurement of packet loss rate is as follows:
Router C2: setting Loss Rate measuring period is T.During each end cycle, C2 sends one to C1 and measures grouping, has write down in this grouping in this cycle from the transmission measurement point of the C2 quantity to the user data packets of C1 transmission, is designated as p
1(see figure 5).
Router C1: when receiving from the measurement of router C2 grouping, the C1 statistics receives measurement point from receiving the user data packets quantity p from C2 that last measurement grouping back successfully receives
2Because may be to the grouping that receives measurement point by the reception queue drop of the transmit queue of C2, transmission medium or C1, p from sending measurement point
2And p
1May unequal (see figure 6).
The computing formula of packet loss rate l from C2 to C1 is:
l＝(1-p
2/p
1)×100％
Be dropped if measure grouping itself, C1 can't calculate the Loss Rate of current period and next cycle, therefore should manage to allow router not abandon and measure grouping.Measurement grouping itself also should comprise a sequence number and detect losing of measurement grouping.In addition, if router is supported the burst or the reorganization of IP grouping, also it must be taken into account during measurement.
In passive measurement, measurement point is inserted into PHM packet handling module to gather the data sample of all processed groupings.Length in the bandwidth measurement adds up and the statistical packet quantity of Loss Rate in measuring all belongs to this class.
In initiatively measuring, sample objects is the measurement grouping that is injected in the network, the problem that exists initiatively sampling to initiate here when, and promptly when tranmitting data register difference measurements request grouping, delay measurements grouping and Loss Rate are measured grouping.Three kinds of method of samplings are arranged usually: periodic sampling (starting point at each time slot is sampled), stratified random sampling (constantly sampling at random in each time slot) and Poisson sampling (sampling instant is exponential distribution with speed λ).First method is the simplest, but possibly can't intactly measure the cyclic behaviour in the network, and measurement itself also may make network produce periodic swinging.The two kinds of methods in back have been avoided this cycle effect, but the randomness in sampling time makes frequency-domain analysis complicated more.In addition, the random time that obtains by the third method does not have the upper bound at interval.To sum up, initiatively measure the sampling of employing stratified random.Fig. 7 has described the Measuring Time in the stratified random sampling.
The precision of oriented link available bandwidth and packet loss rate depends primarily on the deployment of measurement point, and this internal structure with router is relevant.The precision of block transmission delay then not only depends on the deployment of measurement point, also is subjected to the influence of two factors: clock relative pulse phase difference (the single order inverse of clock jitter is seen Fig. 8) and clock refreshing frequency.In experiment, the clock relative pulse phase difference of two PCs is no more than 10 seconds/day, about 0.1 millisecond/second.The clock refreshing frequency is 1.0 milliseconds.Therefore, if with the time clock correction measuring period be made as 1 second, then by above-mentioned formula calculate the time clock correction worst error for (unit millisecond):
Worst error by the block transmission delay that is calculated by above-mentioned formula is (a unit millisecond):
Δ d=1.0 (refreshing frequency) * 4+0.1 (pulse phase difference)=4.1
If the clock refreshing frequency can be brought up to 10
4Hz, the error of block transmission delay will be reduced to (unit millisecond) so:
Δ d=0.1 (refreshing frequency) * 4+0.1 (pulse phase difference)=0.5
This shows that the measurement requirement clock of block transmission delay has low relative pulse phase difference and high refreshing frequency.
In order to filter the transient swing (requiring three) in the measurement result, reduce random error, need carry out data processing to the sample that collects.Because sample distributes more sparse (for meeting the demands four) in time, it is inappropriate carrying out data processing with arithmetic average.Several other methods are discussed below.
Same parameter is taken multiple measurements, and resulting one group of data have constituted sample space.To be dropped when new sample adds fashionable, the oldest sample, thereby make the size of sample space keep N constant.Each sample is by measurement time t
iWith numerical value x
iForm, so sample space can be expressed as:
The index rolling average is given new samples with bigger weight.As a new sample x
NewDuring arrival, statistics y
NewAt preceding once y as a result
OldThe basis on upgrade (recurrence formula):
y
new＝a·y
old+(1-a)·x
new?????(0＜a＜1)
Factor a has reflected the weight of old sample in statistics, and the value of a is big more, and old sample is just strong more to the influence of statistics.This formula is carried out (N-1) inferior iteration, obtains general formula:
y
N＝a
N-1·x
1+a
N-2(1-a)·x
2+a
N-3(1-a)·x
3
+…+a(1-a)·x
N-1+(1-a)·x
N
General formula has been given up too outmoded sample (sample before N) fully in statistics.Although recurrence formula reduces the weight of old sample in the result with factor a, the influence of old sample still can keep long time.Therefore, in handling in real time, use general formula.
Intermediate value sorts by numerical values recited to sample space, gets median (or two medians is average) as statistics.This method is applicable to the sample that accidental fluctuation is bigger, as block transmission delay.
We are applied to the local service quality state measurement in the service quality router prototype system.In experiment, main frame S injects the data flow of a bandwidth linear increase in time to network, and 0 second Time Bandwidth is that 0,1000 second Time Bandwidth is 100Mbps (being the capacity of link).Data flow is routed device A and B transmits, and finally arrives main frame T (Fig. 9).The measurement result of QoS state is (the requiring two) that calculates by the sample to nearest 15 seconds, and this numerical value equals to multiply by measuring period the size of sample space.Method of measurement sees the following form, and measurement result is seen Figure 10 ~ 16.In addition, we also test on two main frames and four routers, and obtain similar result.
| Sample collection | Data processing | ||||
| Cycle (second) | The result | Method | The sample space size | The result | |
| Block transmission delay | ????1 | Figure 10 | Intermediate value | ????15 | Figure 11 |
| Oriented link available bandwidth | ????3 | Figure 12 | The index rolling average | ????5 | Figure 13 |
| Packet loss rate | ????3 | Figure 14 | The index rolling average | ????5 | Figure 15 |
| The time clock correction | ????1 | Figure 16 | (nothing) | ????1 | (nothing) |
Compare Figure 10 and Figure 11, Figure 12 and Figure 13, Figure 14 and Figure 15, data processing has filtered out most transient swing.In Figure 16, even the clock difference measurements of the existing link of utilization also can operate as normal under the network congestion state.Increasing sample space (promptly reducing measuring period) can further level and smooth result, improve stability, but so the burden of network will increase the weight of.Therefore, need require three and require four between seek the compromise.When the PS/AS/BE flow point is not measured, the bandwidth that consumes 2Kbps altogether.Because the router quantity in this numerical value and the network is irrelevant, our measurement scheme has good expandability for large scale network.
By at the real-time measurement module of each router deploy, we can the distributed earth Measurement Network in the QoS state of every link, include to link available bandwidth, block transmission delay and packet loss rate, and without any need for miscellaneous equipment.Four kinds of requirements of measuring in real time all are met in our scheme: if 1. measurement point is rationally disposed, the precision of available bandwidth and Loss Rate can guarantee, and the precision of transmission delay also depends on the performance of clock; 2. statistics has reflected nearest 15 seconds network condition; 3. index rolling average and intermediate value are used to data processing to filter transient swing; 4. measure and be grouped in the bandwidth that only takies 2Kbps on every link, and irrelevant with the router quantity in the network.
This shows that the present invention has reached intended purposes.
Claims (1)
1. the method for real-time measurement of local service quality of computer network state is characterized in that, it contains following steps successively:
Step 1. real-time measurement module of every router deploy in network has defined following parameter at least in this module:
b
u: in the last bandwidth measurement cycle on average at dedicated bandwidth;
C: the capacity of tested local interface respective links, set point;
s
1: the moment of local router network interface card is left in clock difference measurements request grouping;
s
2: clock difference measurements request grouping arrives the moment of neighbor router network interface card;
s
3: clock difference measurements acknowledgment packet is left the moment of neighbor router network interface card;
s
4: clock difference measurements acknowledgment packet arrives the moment of local router network interface card;
t
1: the moment of the transmission measurement point of local router is left in the delay measurements grouping;
t
2: the delay measurements grouping arrives the moment of the reception measurement point of neighbor router;
p
1: at the last Loss Rate of neighbor router in measuring period, the number of packet that sends to local router from the transmission measurement point of neighbor router;
p
2: measure in times between the grouping the quantity that the reception measurement point of local router receives successively receiving two adjacent Loss Rates from neighbor router from the user data packets of this neighbor router;
Sequence number: be used to detect Loss Rate measurement grouping and self whether lose;
Measurement module can generate following grouping at least in real time:
Clock difference measurements request grouping: comprise parameter s at least
1With packet checks and;
Clock difference measurements acknowledgment packet: comprise parameter s at least
1, s
2, s
3With packet checks and;
Delay measurements grouping: comprise parametric t at least
1With packet checks and;
Loss Rate is measured grouping: comprise sequence number, parameter p at least
1With packet checks and;
At least be provided with following computing formula in real time in the measurement module:
The time clock correction: c=((s
1+ s
4)-(s
2+ s
3))/2;
Block transmission delay: d=t
2-t
1-c;
Oriented link available bandwidth: b
a=C-b
u
Packet loss rate: l=(1-p
2/ p
1) * 100%;
Comprise the data processing method that is used for filtering the measurement result transient swing in real time in the measurement module:
Median method is used for the data processing of block transmission delay, and its method is: sample space is sorted by numerical values recited, get median or two medians on average as statistics;
The index method of moving average is used for the data processing of oriented link available bandwidth and packet loss rate, and its computing formula is:
y
N＝a
N-1x
1+a
N-2(1-a)x
2+a
N-3(1-a)x
3+…+a(1-a)x
N-1+(1-a)x
N；
Wherein, y
NBe N measurement result of having filtered transient swing, x
iBe i and do not cross the measurement result that filters transient swing, a is for reducing the tolerance of old sample weight in measurement result, 0＜a＜1;
Each measurement module of step 2. carries out the real-time measurement of local service quality of computer network state successively according to the following steps:
Step 2.1. reads in tested local interface number and the capacity of this interface respective links, the IP address of neighbor router;
Step 2.2. sets clock difference measurements cycle, bandwidth measurement cycle, delay measurements cycle, Loss Rate measuring period, and starts corresponding measurement timer;
Step 2.3. waits for that grouping arrives or timer expired;
If step 2.4. divides into groups to arrive, then carry out following steps successively:
Step 2.4.1. accumulative total block length and number of packet;
Step 2.4.2. is if be grouped into clock difference measurements acknowledgment packet, by the above-mentioned s that records
1, s
2, s
3, s
4Clock correction c forwards step 2.6 to during calculating;
Step 2.4.3. is if be grouped into the delay measurements grouping, by the above-mentioned t that records
1, t
2Calculate block transmission delay d, forward step 2.6 to;
Step 2.4.4. measures grouping if be grouped into Loss Rate, by the above-mentioned p that records
1, p
2Calculate packet loss rate l, forward step 2.6 to;
Step 2.4.5. to neighbor router tranmitting data register difference measurements acknowledgment packet, forwards step 2.3 if be grouped into clock difference measurements request grouping to;
Step 2.4.6. gives router and handles if be grouped into other grouping, forwards step 2.3 to;
If step 2.5. timer expiry is then carried out following steps successively:
Step 2.5.1. is if timer is the bandwidth measurement timer, by above-mentioned C, b
uCalculate oriented link available bandwidth b
a, forward step 2.6 to;
Step 2.5.2. to neighbor router tranmitting data register difference measurements request grouping, forwards step 2.3 if timer is a clock difference measurements timer to;
Step 2.5.3. measures grouping if timer is the delay measurements timer to the neighbor router transmission lag, forwards step 2.3 to;
Step 2.5.4. sends Loss Rate to neighbor router and measures grouping if timer is the Loss Rate measurement timer, forwards step 2.3 to;
Step 2.6. carries out data processing to result of calculation as stated above, forwards step 2.3 to.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2005100114446A CN100364283C (en) | 2005-03-18 | 2005-03-18 | Method for real-time measurement of local service quality of computer network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2005100114446A CN100364283C (en) | 2005-03-18 | 2005-03-18 | Method for real-time measurement of local service quality of computer network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1665204A true CN1665204A (en) | 2005-09-07 |
| CN100364283C CN100364283C (en) | 2008-01-23 |
Family
ID=35036104
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2005100114446A Expired - Fee Related CN100364283C (en) | 2005-03-18 | 2005-03-18 | Method for real-time measurement of local service quality of computer network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100364283C (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100372320C (en) * | 2006-01-11 | 2008-02-27 | 华为技术有限公司 | Method and system for detecting service quality in next generation network |
| CN100459529C (en) * | 2005-09-12 | 2009-02-04 | 华为技术有限公司 | Time delay characteristic measuring method in data communication network |
| CN102137414A (en) * | 2010-06-25 | 2011-07-27 | 华为技术有限公司 | Time-delay-evaluating method and device for mobile video service |
| CN103155488A (en) * | 2010-10-04 | 2013-06-12 | 日本电气株式会社 | Delay measurement system and delay measurement method, as well as delay measurement device and delay measurement program |
| CN103297288A (en) * | 2012-02-24 | 2013-09-11 | 富士通株式会社 | Packet transfer delay measurement system |
| CN103840989A (en) * | 2014-03-26 | 2014-06-04 | 北京极科极客科技有限公司 | Method and device for testing network speed through router |
| CN112422367A (en) * | 2020-11-11 | 2021-02-26 | 成都渊数科技有限责任公司 | Method and system for realizing network delay detection by cloud control strategy and multi-protocol |
| CN112436981A (en) * | 2020-11-16 | 2021-03-02 | 成都渊数科技有限责任公司 | Method and system for measuring network flow by cloud control strategy and multi-protocol implementation |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6801939B1 (en) * | 1999-10-08 | 2004-10-05 | Board Of Trustees Of The Leland Stanford Junior University | Method for evaluating quality of service of a digital network connection |
| US6973622B1 (en) * | 2000-09-25 | 2005-12-06 | Wireless Valley Communications, Inc. | System and method for design, tracking, measurement, prediction and optimization of data communication networks |
| US7245584B2 (en) * | 2002-11-18 | 2007-07-17 | Avaya Technology Corp. | Method and apparatus for auditing service level agreements by test packet insertion |
| CN1302643C (en) * | 2003-05-21 | 2007-02-28 | 华为技术有限公司 | Apparatus for testing network equipment transmission performance and method thereof |
| CN100358296C (en) * | 2003-07-21 | 2007-12-26 | 华为技术有限公司 | Communication performance test method based on traffic |
-
2005
- 2005-03-18 CN CNB2005100114446A patent/CN100364283C/en not_active Expired - Fee Related
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100459529C (en) * | 2005-09-12 | 2009-02-04 | 华为技术有限公司 | Time delay characteristic measuring method in data communication network |
| CN100372320C (en) * | 2006-01-11 | 2008-02-27 | 华为技术有限公司 | Method and system for detecting service quality in next generation network |
| CN102137414B (en) * | 2010-06-25 | 2014-01-22 | 华为技术有限公司 | Time-delay-evaluating method and device for mobile video service |
| WO2011140901A1 (en) * | 2010-06-25 | 2011-11-17 | 华为技术有限公司 | Method and apparatus for evaluating time delay of mobile video service |
| CN102137414A (en) * | 2010-06-25 | 2011-07-27 | 华为技术有限公司 | Time-delay-evaluating method and device for mobile video service |
| CN103155488A (en) * | 2010-10-04 | 2013-06-12 | 日本电气株式会社 | Delay measurement system and delay measurement method, as well as delay measurement device and delay measurement program |
| CN103155488B (en) * | 2010-10-04 | 2016-01-20 | 日本电气株式会社 | Delay measurements system and delay measuring method and delay measurements equipment and delay measurements program |
| US9300558B2 (en) | 2010-10-04 | 2016-03-29 | Nec Corporation | Delay measurement system and delay measurement method, as well as delay measurement device and delay measurement program |
| CN103297288A (en) * | 2012-02-24 | 2013-09-11 | 富士通株式会社 | Packet transfer delay measurement system |
| CN103297288B (en) * | 2012-02-24 | 2016-03-30 | 富士通株式会社 | Grouping queries delay measurements system |
| CN103840989A (en) * | 2014-03-26 | 2014-06-04 | 北京极科极客科技有限公司 | Method and device for testing network speed through router |
| CN112422367A (en) * | 2020-11-11 | 2021-02-26 | 成都渊数科技有限责任公司 | Method and system for realizing network delay detection by cloud control strategy and multi-protocol |
| CN112422367B (en) * | 2020-11-11 | 2021-10-22 | 成都渊数科技有限责任公司 | Method and system for realizing network delay detection by cloud control strategy and multi-protocol |
| CN112436981A (en) * | 2020-11-16 | 2021-03-02 | 成都渊数科技有限责任公司 | Method and system for measuring network flow by cloud control strategy and multi-protocol implementation |
| CN112436981B (en) * | 2020-11-16 | 2021-10-15 | 成都渊数科技有限责任公司 | Method and system for measuring network flow by cloud control strategy and multi-protocol implementation |
Also Published As
| Publication number | Publication date |
|---|---|
| CN100364283C (en) | 2008-01-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1665204A (en) | Method for real-time measurement of local service quality of computer network | |
| Schulman et al. | Bartendr: a practical approach to energy-aware cellular data scheduling | |
| Liu et al. | Fluid models and solutions for large-scale IP networks | |
| Chaintreau et al. | Impact of human mobility on opportunistic forwarding algorithms | |
| CN1661968A (en) | Method for implementig mechanism of variable speed heartbeat based on network and load in host | |
| Solis et al. | In-network aggregation trade-offs for data collection in wireless sensor networks | |
| CN105406932B (en) | Device for computer network | |
| Guo et al. | Time-stepped hybrid simulation (TSHS) for large scale networks | |
| Jun et al. | A cost-efficient software based router and traffic generator for simulation and testing of IP network | |
| CN1878094A (en) | Method for testing maximum available bandwidth of network | |
| CN1770711A (en) | Method and system for measuring bottleneck bandwidth and remanent bandwidth | |
| Pham et al. | On data aggregation quality and energy efficiency of wireless sensor network protocols-extended summary | |
| CN111756656B (en) | Power communication network resource allocation method based on reliability and historical data | |
| US8879403B2 (en) | Link microbenchmarking with idle link correction | |
| Curtis et al. | Review of bandwidth estimation techniques | |
| CN101656653A (en) | Method and device for configuring receive buffer applied to multi-path transmission | |
| CN1921422A (en) | Method for estimating bag-losing ratio | |
| Sharma et al. | Dynamic data compression in multi-hop wireless networks | |
| Chatzipapas et al. | Modelling and real-trace-based evaluation of static and dynamic coalescing for energy efficient ethernet | |
| Yu et al. | Energy-efficient, qos-aware packet scheduling in high-speed networks | |
| Mohammad et al. | Delay analysis of wireless ad hoc networks: Single vs. multiple radio | |
| Hohn et al. | Capturing router congestion and delay | |
| Kleiberg et al. | DeSiNe: a flow-level QoS simulator of networks. | |
| Luo et al. | MEADOWS: modeling, emulation, and analysis of data of wireless sensor networks | |
| Maity et al. | Performance analysis of a discrete-time queue with versatile batch transmission rule under batch size sensitive policy |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20080123 Termination date: 20110318 |