CN111200567B - Dispatching method and dispatching system applied to member port of switch - Google Patents
Dispatching method and dispatching system applied to member port of switch Download PDFInfo
- Publication number
- CN111200567B CN111200567B CN201911412706.8A CN201911412706A CN111200567B CN 111200567 B CN111200567 B CN 111200567B CN 201911412706 A CN201911412706 A CN 201911412706A CN 111200567 B CN111200567 B CN 111200567B
- Authority
- CN
- China
- Prior art keywords
- mmu
- member port
- port
- utilization rate
- average
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 28
- 239000000725 suspension Substances 0.000 claims abstract description 34
- 230000002776 aggregation Effects 0.000 description 2
- 238000004220 aggregation Methods 0.000 description 2
- 239000000470 constituent Substances 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L49/00—Packet switching elements
- H04L49/25—Routing or path finding in a switch fabric
- H04L49/253—Routing or path finding in a switch fabric using establishment or release of connections between ports
- H04L49/254—Centralised controller, i.e. arbitration or scheduling
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention relates to a dispatching method and a dispatching system applied to a member port of a switch, wherein the method comprises the following processes: establishing a resource scheduling pool and adding all member ports; assigning a fixed mmu to each member port; and in the resource scheduling pool, detecting the utilization rate of the mmu in real time, and suspending a member port corresponding to the mmu when the utilization rate of the mmu is higher than a set upper limit. The system comprises a resource scheduling pool for accommodating all member ports and an allocation module for allocating a fixed mmu to each member port; and the scheduling module is used for calculating the suspension probability of each member port, suspending the member port when the suspension probability is 100 percent, and removing the suspension state of the corresponding member port when the mmu utilization rate is lower than the lower limit. By adopting the dynamic scheduling method of the member ports of the port channel ports, the mmu utilization rate of each member port of the port channel ports can be dynamically monitored, and the phenomenon of packet loss is avoided.
Description
Technical Field
The present invention relates to the field of network technologies, and in particular, to a scheduling method and a scheduling system applied to a member port of a switch.
Background
In the process of switch application, the use of ports is involved, wherein a port group is a physical port group on a configuration level, and a physical port configured in the port group can participate in link aggregation and becomes a certain member port in a port channel. The port channel is an aggregation of a group of physical ports, and is logically regarded as one physical port, and for a user, the port channel can be completely regarded as one port for use, so that not only can the bandwidth of a network be increased, but also a backup function of a link and load balancing can be provided.
In a port channel, the port channel may include a plurality of members, and when forwarding traffic, the port channel generally forwards traffic between member ports based on a source IP, a destination IP, a source MAC, and a destination MAC, for example, based on the source IP, then IP1 may forward on member port 1, and IP2 may forward on member port 2, and since the size of traffic corresponding to each IP is different, it may happen that subsequent traffic still goes to member port 1 when the bandwidth of member port 1 runs out, thereby resulting in packet loss.
Disclosure of Invention
The embodiment of the invention provides a scheduling method and a scheduling system applied to a member port of a switch, which are used for solving the problem of packet loss of the member port in the prior art.
In order to solve the technical problem, the embodiment of the invention discloses the following technical scheme:
the invention provides a dispatching method applied to a member port of a switch, which comprises the following processes:
establishing a resource scheduling pool and adding all member ports;
allocating a fixed mmu (cache management unit) to each member port;
and in the resource scheduling pool, detecting the utilization rate of the mmu in real time, and suspending a member port corresponding to the mmu when the utilization rate of the mmu is higher than a set upper limit.
Furthermore, for the suspended member port, when the utilization rate of the mmu corresponding to the suspended member port is lower than the set lower limit, the suspended state of the member port is cancelled, and the suspended member port is recovered to be in a normal state.
Preferably, when a member port is in the suspend state, the traffic before the member port is suspended remains unchanged, and the member port no longer receives new traffic ingress.
Furthermore, the specific implementation principle of calculating the suspension probability and suspending is as follows:
presetting a lower limit and an upper limit;
calculating the average mmu utilization rate of each member port in the current period;
judging whether the average mmu utilization rate of the member port exceeds a lower limit, if so, entering the next step;
calculating the suspension probability by using the ratio of the average mmu utilization rate to the upper limit;
and judging whether the suspension probability is 100%, if so, suspending the member port.
Preferably, the calculation formula of the average mmu utilization rate is as follows: average mmu utilization (mmu utilization of previous cycle x (1- (1/2)n) ) + (mmu utilization of current cycle x (1/2)n),0≤n≤15。
Based on the above method, an embodiment of the present invention further provides a scheduling system applied to a member port of a switch, where the system includes:
a resource scheduling pool accommodating all member ports, an
The distribution module is used for distributing a fixed mmu to each member port; and
and the scheduling module is used for calculating the suspension probability of each member port, suspending the member port when the suspension probability is 100%, and releasing the suspension state of the corresponding member port when the mmu utilization rate is lower than the lower limit.
Further, the scheduling module includes:
the first setting unit is used for setting an upper limit, a lower limit and cycle time; and
the first calculation unit is used for calculating the average mmu utilization rate of each member port in the current period;
the judging unit is used for judging whether the average mmu utilization rate of the member port exceeds the lower limit or not, and if so, the second calculating unit is started;
a second calculation unit for calculating a hang-up probability by using a ratio of the average mmu utilization rate to an upper limit;
and the scheduling unit is used for judging whether the suspension probability is 100%, if so, suspending the corresponding member port, and releasing the suspension state of the corresponding member port when the mmu utilization rate is lower than the lower limit.
The beneficial effects of the above technical scheme are that: the traffic forwarding strategy in the port channel port forwards traffic according to source IP, but the traffic sizes corresponding to each IP are different, so that the situation that subsequent traffic still goes to an Ethernet1 port when the bandwidth of the Ethernet1 port is used up may occur, and packet loss is caused.
In addition, a scheduling strategy based on the average mmu utilization rate calculates a suspension probability based on the average mmu utilization rate, and the suspension state is visually displayed by using the suspension probability, so that on one hand, the situation that the member port frequently enters the suspension state due to instantaneous burst flow is avoided, the mmu can be presented in a stable state, and the influence of the burst flow on the flow scheduling of the port channel port is reduced; on the other hand, because the MMU utilization is a gradual process, when the MMU utilization is 20%, the port may be congested only a little, the port is set to the suspend state with a small probability, when the MMU utilization reaches 50%, indicating that congestion has been compared, the port is set to the suspend state with a medium probability, and when the MMU utilization reaches 80%, the port is set to the suspend state with a percentage probability, and the probability value suspension can guarantee the port utilization and adapt to various changes of traffic to a greater extent.
Drawings
In order to illustrate the embodiments of the present invention or the technical solutions in the prior art more clearly, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious for a person skilled in the art to obtain other drawings without creative efforts.
FIG. 1 is a flow chart of a method provided by an embodiment of the present invention;
FIG. 2 is a schematic diagram of a system provided by an embodiment of the present invention;
fig. 3 is a schematic block diagram of a scheduling module in the system of the present invention.
Detailed Description
In order to make those skilled in the art better understand the technical solution of the present invention, the technical solution in the embodiment of the present invention will be clearly and completely described below with reference to the drawings in the embodiment of the present invention, and it is obvious that the described embodiment is only a part of the embodiment of the present invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, a scheduling method applied to a member port of a switch according to an embodiment of the present invention includes the following steps:
first, a resource scheduling pool is established and all member ports are added.
For the implementation of this process, an actual structure is taken as an example for explanation, and the structure is continued in the subsequent steps, the structure includes two switches, which are Switch 1 or Switch 2, respectively, three sets of physical ports for connecting Ethernet, which are Ethernet1, Ethernet2, Ethernet 3, and three sets of physical ports Ethernet1, Ethernet2, Ethernet 3, form a port channel 1.
Based on the actual architecture, the system establishes a resource scheduling pool, and adds Ethernet1, Ethernet2, and Ethernet 3 to the resource scheduling pool for management.
Secondly, a fixed mmu (cache management unit) is allocated to each member port, and the implementation principle of the process is mainly to set a mmu allocation strategy on Switch 1 and allocate the fixed mmu to each port.
After the two steps are completed, the subsequent steps can be utilized to realize resource scheduling on the port, the strategy of the resource scheduling is preset, and the realization principle is as follows:
1) a lower limit and an upper limit are preset.
For the preset upper limit, the upper limit is self-defined by Switch, and the default is 80%, for example, when the average MMU utilization of the port channel member port Ethernet1 port reaches the value (i.e. is greater than or equal to the value), the Ethernet1 port is set to be in a suspended state in the port scheduling resource pool of the port channel1 at this time. For example, when the detection mmu utilization rate of the Ethernet1 port is greater than or equal to the upper limit, the state in the port channel1 port scheduling resource pool is a suspended state, and after the state of the Ethernet1 port in the port channel1 port scheduling resource pool is the suspended state, the original forwarding traffic remains unchanged.
For the preset lower limit, the lower limit is self-defined by Switch, and default is 20%, for example, when the mmu utilization rate of the port channel member port Ethernet1 port is lower than or equal to the value, the port Ethernet1 port is set to be in a normal state in the port scheduling resource pool of the port channel1 port at this time.
2) And calculating the average mmu utilization rate of each member port in the current period.
The average mmu utilization is calculated as: average mmu utilization (mmu utilization of previous cycle x (1- (1/2)n) ) + (mmu utilization of current cycle x (1/2)n) N is more than or equal to 0 and less than or equal to 15. Generally, n is an integer, the main purpose of the coefficient is to prevent the impact of the instant traffic by adding a reserved buffer space to the traffic of the port, and the average mmu utilization refers to the average utilization of the port in the whole period on the premise of reserving the buffer space.
3) And judging whether the average mmu utilization rate of the member port exceeds a lower limit, if so, entering the next step. It should be noted that the subsequent scheduling process is not started until the average mmu utilization is greater than or equal to the lower limit, for example, 20%, otherwise, the process is repeated until the requirement for entering the next step is met.
4) When the average mmu utilization rate is larger than the lower limit and smaller than the upper limit, the hang probability (average mmu utilization rate/upper limit) of each member port in the current period is calculated, the mmu probability value of each port is dynamically calculated, and the port is set to be in a hang state probabilistically.
It is noted that, according to the above-described calculation formula, when the average mmu utilization is greater than the lower limit, the probability of hang thereof is already over 20%, and if it is calculated with the upper limit of 80%, when the lower limit is reached, the probability of hang thereof is initially 25%.
5) And judging whether the suspension probability is 100% or not based on the calculation result of the last step, and if so, suspending the member port.
And for the suspended member port, when the utilization rate of the mmu corresponding to the suspended member port is lower than the set lower limit, canceling the suspended state of the member port and recovering the suspended member port to be in a normal state. For example, after the state of the Ethernet1 port in the port channel1 scheduling resource pool is set to the suspend state, the suspend state is cancelled when the next cycle detects that the mmu utilization rate of the Ethernet1 port is less than or equal to the lower limit.
Steps 1) -5) are a repeated loop process, which is a loop step executed separately for each port, and the loop step continues as long as any one of all the constituent ports in the port channel is not suspended.
Referring to fig. 2, based on the foregoing method, an embodiment of the present invention further provides a scheduling system applied to a member port of a switch, where the system includes:
and the resource scheduling pool is used for accommodating all member ports, and the member ports refer to all constituent ports in the port channel.
And the allocation module is used for allocating a fixed mmu to each member port after the member ports are added into the resource scheduling pool.
And the scheduling module is used for calculating the suspension probability of each member port, suspending the member port when the suspension probability is 100%, and releasing the suspension state of the corresponding member port when the mmu utilization rate is lower than the lower limit.
For the function realization of the scheduling module, based on a plurality of processing units therein, the specific composition and function are explained as follows:
the first setting unit is used for setting an upper limit, a lower limit and cycle time. For the preset upper limit, the upper limit is self-defined by Switch, and default to 80%, and for the preset lower limit, the lower limit is self-defined by Switch, and default to 20%. The period time may be selected as a default value or may be self-defined.
In an initial state, the first computing unit starts to work, and continuously computes the average mmu utilization rate of each member port in the current period, wherein the computing formula is as follows: average mmu utilization (mmu utilization of previous cycle x (1- (1/2)n) ) + (mmu utilization of current cycle x (1/2)n),0≤n≤15。
And starting the second calculating unit when the judging unit judges that the average mmu utilization rate of the member port exceeds the lower limit based on the calculation result of the first calculating unit, otherwise, entering a continuous judging process, and judging at least once in each period.
When the average mmu utilization rate exceeds the lower limit, the second calculating unit calculates the suspension probability by using the ratio of the average mmu utilization rate to the upper limit, and the suspension probability is used for displaying the suspension state of the member port and is used as a judgment standard for whether the member port is suspended.
Based on the calculated probability value, when the scheduling unit judges that the suspension probability is 100%, the corresponding member port is suspended, the flow of the port in the suspension state is kept unchanged, and only new flow is not added to the member port any more.
It should be noted that the specific functions of each module and the units constituting the module of the system correspond to the principles of the method described in the technical solution, and for the parts of each module or unit whose functions are not described in detail, the working principles in the steps of the method of the present application shall be subject to.
The foregoing are merely exemplary embodiments of the present invention, which enable those skilled in the art to understand or practice the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (4)
1. A scheduling method applied to a member port of a switch is characterized by comprising the following processes:
establishing a resource scheduling pool and adding all member ports;
assigning a fixed mmu to each member port;
in a resource scheduling pool, calculating the suspension probability of each member port in real time, and suspending the member port when the suspension probability is 100%;
the specific implementation principle of calculating the suspension probability and suspending is as follows:
presetting a lower limit and an upper limit;
calculating the average mmu utilization rate of each member port in the current period;
judging whether the average mmu utilization rate of the member port exceeds a lower limit, if so, entering the next step;
calculating the suspension probability by using the ratio of the average mmu utilization rate to the upper limit;
and judging whether the suspension probability is 100%, if so, suspending the member port.
2. The method according to claim 1, wherein for a suspended member port, when the utilization of its corresponding mmu is lower than a preset lower limit, the suspended state of the member port is cancelled and the member port is returned to the normal state.
3. The method according to claim 1, wherein the average mmu utilization is calculated by the following formula: average mmu utilization (mmu utilization of previous cycle x (1- (1/2)n) ) + (mmu utilization of current cycle x (1/2)n),0≤n≤15。
4. A dispatch system for a member port of a switch, the system comprising:
a resource scheduling pool accommodating all member ports, an
The distribution module is used for distributing a fixed mmu to each member port; and
the scheduling module is used for calculating the suspension probability of each member port, suspending the member port when the suspension probability is 100%, and removing the suspension state of the corresponding member port when the mmu utilization rate is lower than the lower limit;
the scheduling module comprises:
the first setting unit is used for setting an upper limit, a lower limit and cycle time; and
the first calculation unit is used for calculating the average mmu utilization rate of each member port in the current period;
the judging unit is used for judging whether the average mmu utilization rate of the member port exceeds the lower limit or not, and if so, the second calculating unit is started;
a second calculation unit for calculating a hang-up probability by using a ratio of the average mmu utilization rate to an upper limit;
the scheduling unit is used for judging whether the suspension probability is 100%, and if so, suspending the corresponding member port; and when the mmu utilization rate is lower than the lower limit, the suspended state of the corresponding member port is released.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911412706.8A CN111200567B (en) | 2019-12-31 | 2019-12-31 | Dispatching method and dispatching system applied to member port of switch |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911412706.8A CN111200567B (en) | 2019-12-31 | 2019-12-31 | Dispatching method and dispatching system applied to member port of switch |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111200567A CN111200567A (en) | 2020-05-26 |
CN111200567B true CN111200567B (en) | 2021-09-10 |
Family
ID=70747363
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201911412706.8A Active CN111200567B (en) | 2019-12-31 | 2019-12-31 | Dispatching method and dispatching system applied to member port of switch |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111200567B (en) |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5983278A (en) * | 1996-04-19 | 1999-11-09 | Lucent Technologies Inc. | Low-loss, fair bandwidth allocation flow control in a packet switch |
CN1571354A (en) * | 2003-07-12 | 2005-01-26 | 华为技术有限公司 | A method for implementing link aggregation |
CN102045253A (en) * | 2010-12-06 | 2011-05-04 | 杭州华三通信技术有限公司 | Method and device for sending video surveillance service messages in link aggregation network |
US8018851B1 (en) * | 2004-06-30 | 2011-09-13 | Marvell Israel (Misl) Ltd. | Flow control for multiport PHY |
CN103391582A (en) * | 2012-05-08 | 2013-11-13 | 普天信息技术研究院有限公司 | Combined dispatching method and device based on buffering state |
CN105245826A (en) * | 2015-08-27 | 2016-01-13 | 浙江宇视科技有限公司 | Method and device for controlling transmission of monitoring video stream |
CN106533964A (en) * | 2015-09-09 | 2017-03-22 | 中兴通讯股份有限公司 | Method and device for managing packet loss of link aggregation member ports |
Family Cites Families (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
GB2461132B (en) * | 2008-06-27 | 2013-02-13 | Gnodal Ltd | Method of data delivery across a network |
US20100054127A1 (en) * | 2008-08-26 | 2010-03-04 | Broadcom Corporation | Aggregate congestion detection and management |
US8238250B2 (en) * | 2009-10-16 | 2012-08-07 | Hei Tao Fung | QoS-aware flow-based dynamic load balancing for link aggregation |
JP5593517B2 (en) * | 2011-07-15 | 2014-09-24 | 株式会社日立製作所 | Network apparatus and transmission frame control method |
US8995277B2 (en) * | 2012-10-30 | 2015-03-31 | Telefonaktiebolaget L M Ericsson (Publ) | Method for dynamic load balancing of network flows on LAG interfaces |
US10917349B2 (en) * | 2017-08-11 | 2021-02-09 | Marvell Israel (M.I.S.L) Ltd. | Congestion avoidance in a network switch device |
-
2019
- 2019-12-31 CN CN201911412706.8A patent/CN111200567B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5983278A (en) * | 1996-04-19 | 1999-11-09 | Lucent Technologies Inc. | Low-loss, fair bandwidth allocation flow control in a packet switch |
CN1571354A (en) * | 2003-07-12 | 2005-01-26 | 华为技术有限公司 | A method for implementing link aggregation |
US8018851B1 (en) * | 2004-06-30 | 2011-09-13 | Marvell Israel (Misl) Ltd. | Flow control for multiport PHY |
CN102045253A (en) * | 2010-12-06 | 2011-05-04 | 杭州华三通信技术有限公司 | Method and device for sending video surveillance service messages in link aggregation network |
CN103391582A (en) * | 2012-05-08 | 2013-11-13 | 普天信息技术研究院有限公司 | Combined dispatching method and device based on buffering state |
CN105245826A (en) * | 2015-08-27 | 2016-01-13 | 浙江宇视科技有限公司 | Method and device for controlling transmission of monitoring video stream |
CN106533964A (en) * | 2015-09-09 | 2017-03-22 | 中兴通讯股份有限公司 | Method and device for managing packet loss of link aggregation member ports |
Also Published As
Publication number | Publication date |
---|---|
CN111200567A (en) | 2020-05-26 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP5402484B2 (en) | Communication apparatus and communication control method | |
CN107005485A (en) | A kind of method, corresponding intrument and system for determining route | |
US8599693B2 (en) | Packet transmission device and packet transmission method | |
EP2728812A1 (en) | Method and device for leaky bucket speed-limitation | |
CN105490967A (en) | Bandwidth allocation method and device | |
CN104980359A (en) | Flow control method of fiber channel over Ethernet (FCoE), flow control device of FCoE and flow control system of FCoE | |
CN109088822B (en) | Data flow forwarding method, device, system, computer equipment and storage medium | |
JP2002223223A (en) | Flow control device and flow control method | |
JP4983438B2 (en) | Packet transmission load balancing control method and apparatus | |
CN109167737A (en) | The CAN dispatching method of fused data reduction and dynamic priority queue | |
CN111200567B (en) | Dispatching method and dispatching system applied to member port of switch | |
CN109286582B (en) | Method and device for realizing network bandwidth priority in virtualized switch | |
CN113726690A (en) | Method and system for uploading protocol message, electronic equipment and storage medium | |
CN101232508A (en) | Equipment and method for speeding up poly spanning tree protocol network topological convergence | |
CN109547352B (en) | Dynamic allocation method and device for message buffer queue | |
JP2005522145A (en) | Method and system for allocating resources dedicated to non-voice data traffic processing | |
CN110347518B (en) | Message processing method and device | |
CN113438185B (en) | Bandwidth allocation method, device and equipment | |
CN110855570A (en) | SDN (software defined network) packet loss judgment method, device and system and multi-network control system | |
WO2016173133A1 (en) | Load sharing implementation method, interface machine, service processor and system | |
CN107623638B (en) | Fault processing method and device for load balancing path | |
CN102595497B (en) | The CDMA data service system of automatic alleviation processor overload and method thereof | |
CN112468393B (en) | Method and device for selecting forwarding port based on link EEE and power consumption and electronic equipment | |
JPH08130560A (en) | Inter-network connecting device and its method | |
JP2008167180A (en) | Packet repeating method, and packet repeating installation |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |