CN101447939A - Functional distribution method and load balancer - Google Patents
Functional distribution method and load balancer Download PDFInfo
- Publication number
- CN101447939A CN101447939A CNA2008102400107A CN200810240010A CN101447939A CN 101447939 A CN101447939 A CN 101447939A CN A2008102400107 A CNA2008102400107 A CN A2008102400107A CN 200810240010 A CN200810240010 A CN 200810240010A CN 101447939 A CN101447939 A CN 101447939A
- Authority
- CN
- China
- Prior art keywords
- node
- record
- resource
- traffic assignments
- reports
- 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
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
- Computer And Data Communications (AREA)
Abstract
The invention discloses a functional distribution method and a load balancer which are applied to the functional distribution between multiple systems or multiple nodes. The method of the invention comprises the following steps: receiving self-resource use status reported by each system/each node through the load balancer; determining functional distribution rate of each system/each node according to the resource use status reported by each system/each node; and distributing business to be assigned to each system/each node according to the functional distribution rate. The adoption of the invention accomplishes dynamic business distribution according to current resource use status of each system/each node, so that practical load quantity of each system/each node is basically balanced.
Description
Technical field
The present invention relates to the traffic assignments between multisystem or the multinode, and the load equalizer that carries out traffic assignments.
Background technology
At present, in the every field, as finance, telecommunications etc. all need often to face ultra-large business, all need to adopt many cover systems, and a plurality of nodes carry out load balancing to whole business, to reach the business of guarantee fast, stable purpose.
Usually, load balancing is in order to allow a plurality of systems or node handle a large amount of concurrent service requests simultaneously, to reach high performance processing.Generally speaking, each system of load balancing or node all are average total traffic carrying capacitys of sharing, and do not have difference mutually.This load balancing is a kind of fixing traffic assignments mode, can not change according to actual conditions intelligently.And in the practical application, the loading condition of each system or node can not accomplish that load truly is just the same according to the difference of device fabrication producer, hardware, software, configuration and extent of deterioration.If there is the part system loading too high, still accept the traffic carrying capacity of mean allocation, then may cause subsequent treatment speed to reduce, a large number of services overstocks, and finally causes quitting work.
At present, when carrying out traffic assignments, another kind of solution is to be configured on load equalizer, allows it that business is distributed according to configuration, reaches all systems and distributes whole professional purpose jointly.But, the problem of such solution maximum is exactly when facing the situation of burst, still distribute and to cause some problems according to configuration file, suppose that the unexpected load of certain system uprises, if still distribute to the traffic carrying capacity of its configuration always, then might cause its processing speed slack-off, and influence certain customers for this professional perception.
Summary of the invention
The invention provides a kind of method for distributing business and load equalizer, carry out traffic assignments according to fixed configurations, cause the unbalanced problem of actual load amount between each system or each node in order to solve in the prior art.
Method for distributing business provided by the invention, be applied to multisystem or/traffic assignments between the multinode, comprising:
Load equalizer receives the own resource behaviour in service that each system/each node reports;
According to the described resource behaviour in service that each system/each node reports, determine the traffic assignments ratio of described each system/each node;
According to described traffic assignments ratio, give described each system/each node with traffic assignments to be allocated.
Described load equalizer by with described each system/each node between chain of command be connected, receive the own resource behaviour in service that described each system/each node reports; And
Described load equalizer by with described each system/each node between service surface be connected, give described each system/each node with traffic assignments to be allocated.
Described each system/each node cycle statistics its other resources utilization rate;
When satisfying the report condition of setting, report the resource utilization of statistics to give described load equalizer.
Described when satisfying the report condition of setting, report the resource utilization of statistics to give described load equalizer, specifically comprise:
When one of them correspondence above setting of resource utilization of statistics reports threshold value, report the resource utilization of statistics to give described load equalizer; And/or
When corresponding variable quantity threshold value that one of them variable quantity of the resource utilization of statistics surpass to be set, report the resource utilization of statistics to give described load equalizer.
The described described resource behaviour in service that reports according to each system/each node is determined the traffic assignments ratio of described each system/each node, specifically comprises:
According to the described resource utilization that each system/each node reports, calculate the resource idleness of each system/each node;
With the resource idleness of each system/each node respectively divided by the resource idleness of each system/each node and value, obtain the traffic assignments ratio of described each system/each node.
Before the described traffic assignments ratio of determining each system/each node, also comprise:
The described resource utilization that described load equalizer reports according to this each system/each node is upgraded the record that reports of local each system/each node of storing;
Report in the record described, corresponding utilization rate threshold value and the current not correspondence of the system/node of inefficacy of selecting resource utilization to surpass setting report record, generate the traffic assignments table;
The described traffic assignments ratio of determining each system/each node specifically comprises:
According to the described resource utilization that respectively reports corresponding each system/each node of record in the described traffic assignments table, calculate the resource idleness that respectively reports corresponding each system/each node of record in the described traffic assignments table;
With the resource idleness that respectively reports corresponding each system/each node of record in the described traffic assignments table that calculates, respectively divided by the resource idleness that respectively reports corresponding each system/each node of record in the described traffic assignments table and value, obtain respectively reporting in the described traffic assignments table traffic assignments ratio of each system/each node of record correspondence.
Described generation traffic assignments table specifically comprises:
Every described record that reports is calculated corresponding record update time and the difference of current time respectively, and relatively with the difference threshold of described difference and setting, when described difference surpasses the difference threshold of setting, determine further this reports current whether inefficacy of corresponding system/node of record; If the current inefficacy of corresponding system/node that this reports record then reports this corresponding record of record to be revised as the current time update time, otherwise reports record to be labeled as failure logging (record) this; And
With every described each resource utilization of comprising of record that reports, respectively with the corresponding utilization rate threshold ratio of setting, when one of them surpasses the corresponding utilization rate threshold value of setting when described resource utilization, report record to be labeled as failure logging (record) this;
Select unmarked all the other for failure logging (record) to report record, generate the traffic assignments table.
According to said method provided by the invention, described each system/each node reports and the corresponding resource utilization of described various resource types according to the various resource types that the needs that set in advance report;
The described traffic assignments ratio of determining each system/each node specifically comprises:
According to the resource utilization corresponding with each resource type that respectively reporting of described traffic assignments table stored in the record, each system/each node of determining each record correspondence in the described traffic assignments table respectively is at each resource type corresponding service allotment ratio;
Described according to the traffic assignments ratio, give described each system/each node with traffic assignments to be allocated, specifically comprise:
According to the respective weights of described various resource types, the traffic carrying capacity of business to be allocated is allocated to described various resource type;
At each resource type corresponding service allotment ratio, the corresponding traffic carrying capacity that each resource type is divided is distributed to described each system/each node concurrently according to described each system/each node.
The present invention also provides a kind of load equalizer, comprising:
Resource behaviour in service receiving element is used to receive the own resource behaviour in service that each system/each node reports;
Traffic assignments ratio determining unit is used for the described resource behaviour in service that reports according to each system/each node, determines the traffic assignments ratio of described each system/each node;
The traffic assignments unit is used for according to described traffic assignments ratio, gives described each system/each node with traffic assignments to be allocated.
Described resource behaviour in service receiving element by and described each system/each node between the chain of command connecting interface, receive the own resource behaviour in service that described each system/each node reports;
Described traffic assignments unit by and described each system/each node between the service surface connecting interface, give described each system/each node with traffic assignments to be allocated.
Above-mentioned load equalizer provided by the invention also comprises:
Record storage update unit is used for the record that reports according to described each system/each node of this resource utilization updated stored that reports of each system/each node;
The traffic assignments table generates the unit, is used for the described record that reports in storage, selects resource utilization to report record above the correspondence of corresponding utilization rate threshold value of setting and the current system/node that did not lose efficacy, and generates the traffic assignments table;
Described traffic assignments ratio determining unit, specifically be used for respectively reporting the described resource utilization of corresponding each system/each node of record, calculate the resource idleness that respectively reports corresponding each system/each node of record in the described traffic assignments table according to described traffic assignments table;
With the resource idleness that respectively reports corresponding each system/each node of record in the described traffic assignments table that calculates, respectively divided by the resource idleness that respectively reports corresponding each system/each node of record in the described traffic assignments table and value, obtain respectively reporting in the described traffic assignments table traffic assignments ratio of each system/each node of record correspondence.
Described traffic assignments table generates the unit, specifically be used for every described record that reports that described record storage update unit is stored, calculate corresponding record update time and the difference of current time respectively, and relatively with the difference threshold of described difference and setting, when described difference surpasses the difference threshold of setting, determine further this reports current whether inefficacy of corresponding system/node of record; If the current inefficacy of corresponding system/node that this reports record then reports this corresponding record of record to be revised as the current time update time, otherwise reports record to be labeled as failure logging (record) this; And
With every described each resource utilization of comprising of record that reports, respectively with the corresponding utilization rate threshold ratio of setting, when one of them surpasses the corresponding utilization rate threshold value of setting when described resource utilization, report record to be labeled as failure logging (record) this; And
Select unmarked all the other for failure logging (record) to report record, generate the traffic assignments table.
Described resource behaviour in service receiving element, receive that various resource types that described each system/each node reports according to the needs that set in advance report with the corresponding resource utilization of described various resource types;
Described traffic assignments ratio determining unit, specifically be used for the resource utilization corresponding with each resource type that respectively reports record to store according to described traffic assignments table, each system/each node of determining each record correspondence in the described traffic assignments table respectively is at each resource type corresponding service allotment ratio; And
According to the respective weights of described various resource types, the traffic carrying capacity of business to be allocated is allocated to described various resource type; And
At each resource type corresponding service allotment ratio, the corresponding traffic carrying capacity that each resource type is divided is distributed to described each system/each node concurrently according to described each system/each node.
Method for distributing business provided by the invention is applied to the traffic assignments between multisystem or the multinode, receives the own resource behaviour in service that each system/each node reports by load equalizer; According to the resource behaviour in service that each system/each node reports, determine each system/each node corresponding service allotment ratio; According to the traffic assignments ratio, give each system/each node with traffic assignments to be allocated.Because the own resource behaviour in service that each system/each node reports, characterized the current real resource behaviour in service of each system/each node, load equalizer is when carrying out traffic assignments, current real resource behaviour in service according to each system/each node, determine a corresponding service allotment ratio traffic assignments to be allocated being given each system/each node, carry out traffic assignments according to the traffic assignments ratio of determining, thereby realize according to the current resource behaviour in service of each system/each node carry out professional dynamic assignment (current resource utilization higher system or node, the corresponding service allotment ratio is lower; Otherwise, system that current resource utilization is lower or node, the corresponding service allotment ratio is higher), make that the actual load amount between each system/each node reaches balanced substantially.
Description of drawings
The method for distributing business flow chart that Fig. 1 provides for the embodiment of the invention;
Multisystem/the multinode that Fig. 2 provides for the embodiment of the invention and the annexation schematic diagram of load equalizer;
The flow chart that Fig. 3 provides for the embodiment of the invention by load equalizer realization multisystem traffic assignments;
Fig. 4 distributes schematic diagram for the traffic carrying capacity that the embodiment of the invention provides;
Fig. 5 A, Fig. 5 B organize the corresponding relation schematic diagram of the traffic assignments that realizes that various resource types are divided for what the invention process row provided by processing;
The load equalizer structural representation that Fig. 6 provides for the invention process row.
Embodiment
The invention provides a kind of method for distributing business and load equalizer, be applied to the traffic assignments between the multisystem, the traffic assignments that perhaps is applied between the multinode is (easy for describing, the mode of employing system/node, expression system or node), report its other resources behaviour in service by each system/each node, according to the resource behaviour in service on each system/each node, carry out professional dynamic assignment by load equalizer.
Below in conjunction with accompanying drawing,, be described in detail method for distributing business provided by the invention.
Referring to Fig. 1, the method for distributing business flow chart for the embodiment of the invention provides specifically comprises:
Step S101, load equalizer receive the own resource behaviour in service that each system/each node reports;
Step S102, according to the resource behaviour in service that each system/each node reports, determine the traffic assignments ratio of each system/each node;
Step S103, according to the traffic assignments ratio of determining, give each system/each node with traffic assignments to be allocated.
Referring to Fig. 2, be the annexation schematic diagram of a plurality of systems (each system also can be a node) with load equalizer.Among Fig. 2, black dotted line represents that load equalizer is connected with chain of command between each system, and black solid line represents that load equalizer is connected with user's face between each system.Load equalizer by with each system between chain of command be connected, receive the own resource behaviour in service that each system reports; And by with each system between service surface be connected, give each system with traffic assignments to be allocated.
Networking diagram shown in Figure 2 is divided into the chain of command connection with the connection of the network between load equalizer and each system and is connected with service surface, and the chain of command connection is connected with service surface can logically be cut apart also and can cut apart physically.Can realize that like this control and carrying are separated, its benefit is because the load-balancing technique of intelligence often needs relatively large and frequent up-downgoing control information, like this control information and business information are separated fully, can guarantee not only can not to conflict fully in two aspects or mutually between bandwidth occupancy, aspect expansion, also can only consider wherein demand in a certain respect simultaneously respectively and can not cause negative effect on the other hand.In general, chain of command is more paid attention to the complexity of control information and intelligent, and service surface is more paid attention to accurately and fast issuing processing for business.Therefore diverse two aspects of characteristic if can separately consider, then can improve the efficient of Intelligent treatment.
Report the own resource utilization rate with a plurality of systems to load equalizer below, realize that by load equalizer load balancing is an example, said method provided by the invention is elaborated.
Referring to Fig. 3, the specific implementation step comprises:
Step S301, each system statistics own resource behaviour in service.
Step S302, judge whether to satisfy report condition,, then go to step S301 if do not satisfy report condition; If satisfy report condition, continue step S303.
Step S303, each system report the own resource utilization rate of statistics to load equalizer.
The resource utilization that step S304, load equalizer report according to this each system is upgraded the record that reports of local storage.In this step, load equalizer is also stored the record update time of each record correspondence.
Step S305, load equalizer generate the traffic assignments table according to the record of storing that reports.
Step S306, according to the resource utilization that the system that each record is corresponding in the traffic assignments table reports, determine each system's corresponding service allotment ratio.
Step S307, according to each system's corresponding service allotment ratio, give each system with traffic assignments to be allocated.
Above steps specifically describes as follows:
Among the step S301, each system can be provided with an enough little time interval △ t according to the situation of self and add up by certain mode the current utilization rate x of the various resource types of native system.The resource type of concrete statistics depends on whether it is influential to the whole load-bearing capacity of system.For example: the concrete resource type that supposing the system reports comprises: CPU, internal memory and I/O (I/O) port, then system adds up CPU usage x (1), memory usage x (2), I/O port utilization x (3) every △ t duration.
Among the step S302, trigger the report condition that reports, for example:
When one of them correspondence above setting of resource utilization of statistics reports threshold value, report the resource utilization of statistics to give load equalizer; And when corresponding variable quantity threshold value that one of them variable quantity of the resource utilization of statistics surpass to be set, report the resource utilization of statistics to give load equalizer.
Supposing the system CPU usage correspondence report that threshold value is 60%, the threshold value that reports of Installed System Memory utilization rate correspondence is 70%, the threshold value that reports of system I/O port utilization correspondence is 80%; If the utilization rate x (1) of statistics acquisition CPU is 75% in the current period, memory usage x (2) is 35%, I/O port utilization x (3) is 45%, then, CPU usage x (1) reports threshold value 60% owing to having surpassed, satisfy report condition, CPU usage, memory usage and I/O port utilization that system will report this statistics to obtain;
In addition, system should preserve nearest twice statistical value at least, just can obtain the variation delta x of certain resource utilization in Δ t then.When Δ x surpasses the variable quantity threshold value of setting, CPU usage, memory usage and I/O port utilization that system also will report this statistics to obtain.
Among one embodiment, certain system can be shown in following table one at a statistical form of this locality storage:
Table one:
| IP | x(1) | Δx(1) | x(2) | Δx(2) | …… | x(n) | Δx(n) | Time (Time) |
| 10.0.0.1 | 55％ | 8 | 44％ | -5 | 10％ | 30 | 2008-6-10?20:23 |
In the table, IP represents the IP address of this system's correspondence, the utilization rate of the n kind resource type correspondence of x (1)～x (n) expression system statistics, Δ x (1)～Δ x (n) expression is corresponding to the variable quantity of every kind of nearest twice statistical value of resource type, and the time (Time) represents that concrete statistics is constantly.
In step S305, generate the traffic assignments table and be specially:
On the one hand, all report in the record, report record to calculate corresponding record update time and the difference of current time respectively to every, and relatively with the difference threshold of difference and setting, when the difference that calculates surpasses the difference threshold of setting, determine further this reports current whether inefficacy of corresponding system of record; If the current inefficacy of corresponding system that this reports record then reports this corresponding record of record to be revised as the current time update time, otherwise reports record to be labeled as failure logging (record) this;
On the other hand, report each resource utilization of comprising of record with every, respectively with the corresponding utilization rate threshold ratio of setting, when one of them surpasses the corresponding utilization rate threshold value of setting when resource utilization, report record to be labeled as failure logging (record) this;
At last, select all unmarked all the other for failure logging (record) to report record, generate the traffic assignments table.
Determine to report current whether inefficacy of corresponding system of record, can adopt to corresponding system and send detect-message,, judge that then corresponding system did not lose efficacy if receive normally receiveing the response of corresponding system; Otherwise determining that corresponding system is current lost efficacy.
Among one embodiment, can be to respectively reporting an identification field of recording setting (Flag), the Flag field initial value of each record is defaulted as effectively value (value of establishing is OK) of this record of expression, for the corresponding Flag field value that is labeled as failure logging (record) according to the said method needs, be revised as the value NOT OK of this protocol failure of expression.Choose and revise the corresponding record generation traffic assignments table of back Flag value for OK.
Load equalizer can generate this traffic assignments table before carrying out traffic assignments; Also can regularly generate the traffic assignments table.
Following table two is an object lesson of the traffic assignments table of load equalizer generation:
Table two
| IP | x(1) | x(2) | x(3) | …… | x(n) | Time (Time) | Flag |
| 10.0.0.1 | 25％ | 75％ | 30％ | …… | 20％ | 2008-6-10?20:23 | OK |
| 10.0.0.2 | 30％ | 80％ | 46％ | …… | 43％ | 2008-6-10?20:26 | OK |
| 10.0.0.3 | 15％ | 40％ | 85％ | …… | 10％ | 2008-6-10?20:29 | OK |
In the table two, expression has three systems, and the IP address of system's 1 correspondence is respectively 10.0.0.1, and the IP address of system's 2 correspondences is respectively 10.0.0.2, the IP address of system's 3 correspondences is respectively 10.0.0.3, and each system has reported the resource utilization of n kind resource type correspondence.
Among the step S306,, determine each system's corresponding service allotment ratio, specifically comprise according to the resource utilization that the system that each record is corresponding in the traffic assignments table reports:
According to the resource utilization that each system reports, calculate the resource idleness of each system; With the resource idleness of each system respectively divided by the resource idleness of each system and value, obtain the traffic assignments ratio of each system.
Above table two is an example, supposes that x (1) represents the CPU usage of each system, and x (2) represents each Installed System Memory utilization rate, and all the other x (n) represent other resource utilization of each system.Current business need be distributed to system 1, system 2 and the system 3 in the table two, and then the CPU usage x (1) according to each system determines at resource type CPU corresponding service allotment ratio, and concrete grammar is:
The CPU usage of system 1 is 25%, and then the CPU idleness of system 1 is 75%;
The CPU usage of system 2 is 30%, and then the CPU idleness of system 2 is 70%;
The CPU usage of system 3 is 15%, and then the CPU idleness of system 3 is 85%;
The CPU idleness of system 1, system 2 and system 3 and value are 75%+70%+85%=230%;
System 1 is at resource type CPU, and corresponding traffic assignments ratio is 75%/230%=75/230=33%;
System 2 is at resource type CPU, and corresponding traffic assignments ratio is 70%/230%=70/230=30%;
System 3 is at resource type CPU, and corresponding traffic assignments ratio is 85%/230%=85/230=37%.
In like manner, determine at resource type internal memory corresponding service allotment ratio according to the memory usage x (2) of each system, concrete grammar is:
The memory usage of system 1 is 75%, and then the internal memory idleness of system 1 is 25%;
The memory usage of system 2 is 80%, and then the internal memory idleness of system 2 is 20%;
The memory usage of system 3 is 40%, and then the internal memory idleness of system 3 is 60%;
The internal memory idleness of system 1, system 2 and system 3 and value are 25%+20%+60%=105%;
System 1 is at the resource type internal memory, and corresponding traffic assignments ratio is 25%/105%=25/105=24%;
System 2 is at the resource type internal memory, and corresponding traffic assignments ratio is 20%/105%=20/105=19%;
System 3 is at the resource type internal memory, and corresponding traffic assignments ratio is 60%/105%=60/105=57%.
For other resource type, in like manner can calculate the corresponding service allotment ratio.
In the practical application, the various resource types that each system can report according to the needs that set in advance report corresponding resource utilization.
In step S307, according to each system's corresponding service allotment ratio, give each system with traffic assignments to be allocated, specifically comprise:
If each system has only reported a kind of utilization rate of resource type, then can obtain each system's corresponding service allotment ratio according to said method, according to this traffic assignments ratio, give each system with traffic assignments to be allocated.
If each system has reported the utilization rate of more than a kind of resource type, then need respective weights according to the various resource types that set in advance, the traffic carrying capacity of business to be allocated is allocated to various resource types; According to the corresponding service allotment ratio that each system calculates at each resource type, the corresponding traffic carrying capacity that each resource type is divided is distributed to each system concurrently.
Still above table two is an example, at the respective weights of n kind resource type, traffic carrying capacity to be allocated is allocated to n kind resource type, and the corresponding traffic carrying capacity that each resource type is divided is distributed to each system concurrently.Referring to Fig. 4, for traffic carrying capacity is distributed schematic diagram.Among Fig. 4, the n kind resource type according to each system reports with the weight of traffic carrying capacity to be allocated according to each resource type correspondence, is allocated to n kind resource type (weight coefficient of each resource type of hypothesis is identical among Fig. 4, all is 1/n).Each resource type is divided the traffic carrying capacity that obtains, all need to distribute to each system (s1, s2 among Fig. 4 ... sp represents that total P system participates in the traffic carrying capacity distribution), the traffic assignments principle is:
According to the corresponding traffic assignments ratio of each system that calculates, the traffic carrying capacity that is allocated to this kind resource type is distributed to each system at every kind of resource type.
For example: suppose that traffic carrying capacity to be allocated is M, according to the respective weights of the resource type CPU of system, the traffic carrying capacity that is allocated to resource type CPU is m1, and according to the respective weights of the resource type internal memory of system, the traffic carrying capacity that is allocated to the resource type internal memory is m2; Still above table two is an example, and then according to the corresponding traffic assignments ratio at resource type CPU of each system, the traffic carrying capacity of distributing to each system is respectively:
System 1 is 33% at the corresponding traffic assignments ratio of resource type CPU, and the traffic carrying capacity of distributing to system 1 at resource type CPU is m1*33%;
System 2 is 30% at the corresponding traffic assignments ratio of resource type CPU, and the traffic carrying capacity of distributing to system 2 at resource type CPU is m1*30%;
System 3 is 37% at the corresponding traffic assignments ratio of resource type CPU, and the traffic carrying capacity of distributing to system 3 at resource type CPU is m1*37%.
In like manner, according to the corresponding traffic assignments ratio at the resource type internal memory of each system, the traffic carrying capacity of distributing to each system is respectively:
System 1 is 24% at the corresponding traffic assignments ratio of resource type internal memory, and giving the traffic carrying capacity of system 1 at the resource type Memory Allocation is m2*24%;
System 2 is 19% at the corresponding traffic assignments ratio of resource type internal memory, and giving the traffic carrying capacity of system 2 at the resource type Memory Allocation is m2*19%;
System 3 is 57% at the corresponding traffic assignments ratio of resource type internal memory, and giving the traffic carrying capacity of system 3 at the resource type Memory Allocation is m2*57%.
According to each system at various resource type corresponding service allotment ratios (as corresponding traffic assignments ratio, at the corresponding traffic assignments ratio of resource type internal memory etc.), parallel carrying out when the distribution service amount is given each system at resource type CPU.
Among one embodiment, can start m process simultaneously and carry out traffic assignments concurrently.Referring to Fig. 5 A, represent the corresponding traffic carrying capacity that m process divided at n kind resource type simultaneously, carry out the corresponding relation schematic diagram of traffic assignments at the corresponding traffic assignments ratio of every kind of resource type according to each system.Among Fig. 5 A, the respective weights of supposing each resource type is also unequal, for example, the weight of second kind of resource type is 2 times of first kind of resource type, then carry out traffic carrying capacity when dividing by the weight size, it will be 2 times that first kind of resource type divided the traffic carrying capacity that obtains that second kind of resource type divided the traffic carrying capacity that obtains, therefore, second kind of resource type can adopt two process P2 and P3 to carry out traffic assignments, and first kind of resource type can only adopt a process P1 to carry out traffic assignments.Fig. 5 B also is the corresponding traffic carrying capacity that m process of expression divided at n kind resource type simultaneously, carries out the corresponding relation schematic diagram of traffic assignments at the corresponding traffic assignments ratio of every kind of resource type according to each system.Different with Fig. 5 A is, among Fig. 5 B, the weight of various resource type correspondences equates, therefore every kind of resource type is all enabled two corresponding processes and carried out traffic assignments.
Adopt said method of the present invention, utilization rate according to the various resource types of each system, determine the corresponding traffic assignments ratio of each system at each resource type, give various resource types with delineation of activities to be allocated earlier, the traffic carrying capacity that every kind of resource type is divided is distributed to each system concurrently according to the corresponding service allotment ratio again.Realized occupancy, reasonably, dynamically carried out traffic assignments, to guarantee the load balancing between each system according to the various resources of each system.
The foregoing description is to be example with the traffic assignments between the multisystem, for the traffic assignments between the multinode, can adopt the disclosed method for distributing business of the above embodiment of the present invention equally.
One of ordinary skill in the art will appreciate that all or part of step that realizes in the foregoing description method is to instruct relevant hardware to finish by program, this program can be stored in the computer read/write memory medium, as: ROM/RAM, magnetic disc, CD etc.
Based on same inventive concept, the present invention also provides a kind of load equalizer, and its structural representation comprises as shown in Figure 6:
Resource behaviour in service receiving element 61 is used to receive the own resource behaviour in service that each system/each node reports;
Traffic assignments ratio determining unit 62 is used for the resource behaviour in service that reports according to each system/each node, determines the traffic assignments ratio of each system/each node;
Among one embodiment, resource behaviour in service receiving element 61 by and each system/each node between the chain of command connecting interface, receive the own resource behaviour in service that each system/each node reports; Traffic assignments unit 63 by and each system/each node between the service surface connecting interface, give each system/each node with traffic assignments to be allocated.
Among one embodiment, above-mentioned load equalizer provided by the invention also comprises:
Record storage update unit 64 is used for the record that reports according to each system/each node of this resource utilization updated stored that reports of each system/each node;
The traffic assignments table generates unit 65, is used for the record that reports in 64 storages of record storage update unit, selects resource utilization to report record above the correspondence of corresponding utilization rate threshold value of setting and the current system/node that did not lose efficacy, and generates the traffic assignments table;
Traffic assignments ratio determining unit 62 specifically is used for respectively reporting the resource utilization that writes down corresponding each system/each node according to the traffic assignments table that generates, and calculates the resource idleness that respectively reports corresponding each system/each node of record in the traffic assignments table; With the resource idleness that respectively reports corresponding each system/each node of record in the traffic assignments table that calculates, respectively divided by the resource idleness that respectively reports corresponding each system/each node of record in the traffic assignments table and value, obtain respectively reporting in the traffic assignments table traffic assignments ratio of each system/each node of record correspondence.
Among one embodiment, the traffic assignments table generates unit 65, every that specifically is used for record storage update unit 64 is stored reports record, calculate corresponding record update time and the difference of current time respectively, and relatively with the difference threshold of difference and setting, when difference surpasses the difference threshold of setting, determine further this reports current whether inefficacy of corresponding system/node of record; If the current inefficacy of corresponding system/node that this reports record then reports this corresponding record of record to be revised as the current time update time, otherwise reports record to be labeled as failure logging (record) this; And report each resource utilization of comprising of record with every, respectively with the corresponding utilization rate threshold ratio of setting, when one of them surpasses the corresponding utilization rate threshold value of setting when resource utilization, report record to be labeled as failure logging (record) this; And select unmarked all the other for failure logging (record) to report record, generate the traffic assignments table.
Among one embodiment, resource behaviour in service receiving element 61 receives the resource utilization corresponding with various resource types that various resource types that each system/each node reports according to the needs that set in advance report;
Traffic assignments ratio determining unit 62, specifically be used for the resource utilization corresponding with each resource type that respectively reports record to store according to the traffic assignments table, each system/each node of determining each record correspondence in the traffic assignments table respectively is at each resource type corresponding service allotment ratio; And, give various resource types with delineation of activities to be allocated according to the respective weights of various resource types; Again according to each system/each node at each resource type corresponding service allotment ratio, the corresponding traffic assignments that each resource type is divided is given each system/each node concurrently.
In sum, according to method for distributing business provided by the invention, load equalizer is when carrying out traffic assignments, the real resource behaviour in service that reports according to each system/each node, determine a corresponding service allotment ratio traffic assignments to be allocated being given each system/each node, carry out traffic assignments according to the traffic assignments ratio of determining, thereby realize carrying out professional dynamic assignment according to the current resource behaviour in service of each system/each node.Current resource utilization higher system or node, the corresponding service allotment ratio is lower; Otherwise, system that current resource utilization is lower or node, the corresponding service allotment ratio is higher, thereby reaches the actual load amount equilibrium between each system/each node.
Obviously, those skilled in the art can carry out various changes and modification to the present invention and not break away from the spirit and scope of the present invention.Like this, if of the present invention these are revised and modification belongs within the scope of claim of the present invention and equivalent technologies thereof, then the present invention also is intended to comprise these changes and modification interior.
Claims (13)
1, a kind of method for distributing business, be applied to multisystem or/traffic assignments between the multinode, it is characterized in that, comprising:
Load equalizer receives the own resource behaviour in service that each system/each node reports;
According to the described resource behaviour in service that each system/each node reports, determine the traffic assignments ratio of described each system/each node;
According to described traffic assignments ratio, give described each system/each node with traffic assignments to be allocated.
2, the method for claim 1 is characterized in that, described load equalizer by with described each system/each node between chain of command be connected, receive the own resource behaviour in service that described each system/each node reports; And
Described load equalizer by with described each system/each node between service surface be connected, give described each system/each node with traffic assignments to be allocated.
3, method as claimed in claim 2 is characterized in that, described each system/each node cycle statistics its other resources utilization rate;
When satisfying the report condition of setting, report the resource utilization of statistics to give described load equalizer.
4, method as claimed in claim 3 is characterized in that, and is described when satisfying the report condition of setting, and reports the resource utilization of statistics to give described load equalizer, specifically comprises:
When one of them correspondence above setting of resource utilization of statistics reports threshold value, report the resource utilization of statistics to give described load equalizer; And/or
When corresponding variable quantity threshold value that one of them variable quantity of the resource utilization of statistics surpass to be set, report the resource utilization of statistics to give described load equalizer.
5, method as claimed in claim 4 is characterized in that, the described described resource behaviour in service that reports according to each system/each node is determined the traffic assignments ratio of described each system/each node, specifically comprises:
According to the described resource utilization that each system/each node reports, calculate the resource idleness of each system/each node;
With the resource idleness of each system/each node respectively divided by the resource idleness of each system/each node and value, obtain the traffic assignments ratio of described each system/each node.
6, method as claimed in claim 5 is characterized in that, before the described traffic assignments ratio of determining each system/each node, also comprises:
The described resource utilization that described load equalizer reports according to this each system/each node is upgraded the record that reports of local each system/each node of storing;
Report in the record described, corresponding utilization rate threshold value and the current not correspondence of the system/node of inefficacy of selecting resource utilization to surpass setting report record, generate the traffic assignments table;
The described traffic assignments ratio of determining each system/each node specifically comprises:
According to the described resource utilization that respectively reports corresponding each system/each node of record in the described traffic assignments table, calculate the resource idleness that respectively reports corresponding each system/each node of record in the described traffic assignments table;
With the resource idleness that respectively reports corresponding each system/each node of record in the described traffic assignments table that calculates, respectively divided by the resource idleness that respectively reports corresponding each system/each node of record in the described traffic assignments table and value, obtain respectively reporting in the described traffic assignments table traffic assignments ratio of each system/each node of record correspondence.
7, method as claimed in claim 6 is characterized in that, described generation traffic assignments table specifically comprises:
Every described record that reports is calculated corresponding record update time and the difference of current time respectively, and relatively with the difference threshold of described difference and setting, when described difference surpasses the difference threshold of setting, determine further this reports current whether inefficacy of corresponding system/node of record; If the current inefficacy of corresponding system/node that this reports record then reports this corresponding record of record to be revised as the current time update time, otherwise reports record to be labeled as failure logging (record) this; And
With every described each resource utilization of comprising of record that reports, respectively with the corresponding utilization rate threshold ratio of setting, when one of them surpasses the corresponding utilization rate threshold value of setting when described resource utilization, report record to be labeled as failure logging (record) this;
Select unmarked all the other for failure logging (record) to report record, generate the traffic assignments table.
8, method as claimed in claim 6 is characterized in that, described each system/each node reports and the corresponding resource utilization of described various resource types according to the various resource types that the needs that set in advance report;
The described traffic assignments ratio of determining each system/each node specifically comprises:
According to the resource utilization corresponding with each resource type that respectively reporting of described traffic assignments table stored in the record, each system/each node of determining each record correspondence in the described traffic assignments table respectively is at each resource type corresponding service allotment ratio;
Described according to the traffic assignments ratio, give described each system/each node with traffic assignments to be allocated, specifically comprise:
According to the respective weights of described various resource types, the traffic carrying capacity of business to be allocated is allocated to described various resource type;
At each resource type corresponding service allotment ratio, the corresponding traffic carrying capacity that each resource type is divided is distributed to described each system/each node concurrently according to described each system/each node.
9, a kind of load equalizer is characterized in that, comprising:
Resource behaviour in service receiving element is used to receive the own resource behaviour in service that each system/each node reports;
Traffic assignments ratio determining unit is used for the described resource behaviour in service that reports according to each system/each node, determines the traffic assignments ratio of described each system/each node;
The traffic assignments unit is used for according to described traffic assignments ratio, gives described each system/each node with traffic assignments to be allocated.
10, load equalizer as claimed in claim 9 is characterized in that, described resource behaviour in service receiving element by and described each system/each node between the chain of command connecting interface, receive the own resource behaviour in service that described each system/each node reports;
Described traffic assignments unit by and described each system/each node between the service surface connecting interface, give described each system/each node with traffic assignments to be allocated.
11, load equalizer as claimed in claim 10 is characterized in that, also comprises:
Record storage update unit is used for the record that reports according to described each system/each node of this resource utilization updated stored that reports of each system/each node;
The traffic assignments table generates the unit, is used for the described record that reports in storage, selects resource utilization to report record above the correspondence of corresponding utilization rate threshold value of setting and the current system/node that did not lose efficacy, and generates the traffic assignments table;
Described traffic assignments ratio determining unit, specifically be used for respectively reporting the described resource utilization of corresponding each system/each node of record, calculate the resource idleness that respectively reports corresponding each system/each node of record in the described traffic assignments table according to described traffic assignments table;
With the resource idleness that respectively reports corresponding each system/each node of record in the described traffic assignments table that calculates, respectively divided by the resource idleness that respectively reports corresponding each system/each node of record in the described traffic assignments table and value, obtain respectively reporting in the described traffic assignments table traffic assignments ratio of each system/each node of record correspondence.
12, load equalizer as claimed in claim 11, it is characterized in that, described traffic assignments table generates the unit, specifically be used for every described record that reports that described record storage update unit is stored, calculate corresponding record update time and the difference of current time respectively, and with the difference threshold of described difference and setting relatively, when described difference surpasses the difference threshold of setting, determine further this reports current whether inefficacy of corresponding system/node of record; If the current inefficacy of corresponding system/node that this reports record then reports this corresponding record of record to be revised as the current time update time, otherwise reports record to be labeled as failure logging (record) this; And
With every described each resource utilization of comprising of record that reports, respectively with the corresponding utilization rate threshold ratio of setting, when one of them surpasses the corresponding utilization rate threshold value of setting when described resource utilization, report record to be labeled as failure logging (record) this; And
Select unmarked all the other for failure logging (record) to report record, generate the traffic assignments table.
13, load equalizer as claimed in claim 12, it is characterized in that, described resource behaviour in service receiving element, receive that various resource types that described each system/each node reports according to the needs that set in advance report with the corresponding resource utilization of described various resource types;
Described traffic assignments ratio determining unit, specifically be used for the resource utilization corresponding with each resource type that respectively reports record to store according to described traffic assignments table, each system/each node of determining each record correspondence in the described traffic assignments table respectively is at each resource type corresponding service allotment ratio; And
According to the respective weights of described various resource types, the traffic carrying capacity of business to be allocated is allocated to described various resource type; And
At each resource type corresponding service allotment ratio, the corresponding traffic carrying capacity that each resource type is divided is distributed to described each system/each node concurrently according to described each system/each node.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2008102400107A CN101447939B (en) | 2008-12-16 | 2008-12-16 | Functional distribution method and load balancer |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2008102400107A CN101447939B (en) | 2008-12-16 | 2008-12-16 | Functional distribution method and load balancer |
Related Child Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201110008514 Division CN102075425B (en) | 2008-12-16 | 2008-12-16 | Business allocation method and load balancer |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101447939A true CN101447939A (en) | 2009-06-03 |
| CN101447939B CN101447939B (en) | 2011-09-28 |
Family
ID=40743351
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2008102400107A Active CN101447939B (en) | 2008-12-16 | 2008-12-16 | Functional distribution method and load balancer |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101447939B (en) |
Cited By (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101945353A (en) * | 2009-07-07 | 2011-01-12 | 中国移动通信集团山东有限公司 | Method and system for remote adjustment of system resources |
| CN102111337A (en) * | 2011-03-14 | 2011-06-29 | 浪潮(北京)电子信息产业有限公司 | Method and system for task scheduling |
| CN102324967A (en) * | 2011-05-24 | 2012-01-18 | 中兴通讯股份有限公司 | Method and system for mutual scheduling of equipment resources |
| CN101631360B (en) * | 2009-08-19 | 2012-05-09 | 中兴通讯股份有限公司 | Method, device and system for realizing load balancing |
| CN102469023A (en) * | 2010-11-19 | 2012-05-23 | 中国移动通信集团公司 | Dispatching method, unit and system based on cloud computing |
| CN102573072A (en) * | 2010-12-29 | 2012-07-11 | 中国移动通信集团湖南有限公司 | Self-adaptive regulating method and device of network resource allocation as well as communication system |
| CN102891881A (en) * | 2012-07-09 | 2013-01-23 | 北京中创信测科技股份有限公司 | Method for implementing equivalence and balance of nodes under cloud environment |
| CN102932281A (en) * | 2012-10-31 | 2013-02-13 | 华为技术有限公司 | Method and equipment for dynamic allocation of resources |
| CN102932015A (en) * | 2011-08-12 | 2013-02-13 | 国家广播电影电视总局广播科学研究院 | Small power short wave scheduling method |
| CN101594642B (en) * | 2009-06-18 | 2013-05-08 | 中兴通讯股份有限公司 | Resource allocation method and resource allocation device used for communication network element device |
| CN103617086A (en) * | 2013-11-20 | 2014-03-05 | 东软集团股份有限公司 | Parallel computation method and system |
| CN103703830A (en) * | 2013-05-31 | 2014-04-02 | 华为技术有限公司 | Physical resource adjustment method, device and controller |
| CN103841126A (en) * | 2012-11-20 | 2014-06-04 | 中国移动通信集团辽宁有限公司 | Shared resource deployment method and device |
| CN103973586A (en) * | 2014-05-14 | 2014-08-06 | 江苏省无线电科学研究所有限公司 | Method for achieving automatic meteorological station wireless networking load balance |
| CN104780210A (en) * | 2015-04-13 | 2015-07-15 | 杭州华三通信技术有限公司 | Load balancing method and device |
| CN105389212A (en) * | 2015-10-21 | 2016-03-09 | 浪潮(北京)电子信息产业有限公司 | Job assigning method and apparatus |
| CN105450537A (en) * | 2014-09-02 | 2016-03-30 | 阿尔卡特朗讯 | Method and device for determining node load information, as well as carrying out load balancing and overload control based thereon |
| CN105491138A (en) * | 2015-12-15 | 2016-04-13 | 国网智能电网研究院 | Load rate based graded triggering distributed load scheduling method |
| CN105656973A (en) * | 2014-11-25 | 2016-06-08 | 中国科学院声学研究所 | Distributed method and system for scheduling tasks in node group |
| CN106815061A (en) * | 2015-12-01 | 2017-06-09 | 阿里巴巴集团控股有限公司 | A kind of method for processing business and device |
| WO2018233299A1 (en) * | 2017-06-22 | 2018-12-27 | 平安科技（深圳）有限公司 | Method, apparatus and device for scheduling processor, and medium |
| CN110175074A (en) * | 2019-04-18 | 2019-08-27 | 北京奇艺世纪科技有限公司 | Load-balancing method and server, load unit, service processing apparatus and medium |
| WO2019165836A1 (en) * | 2018-02-28 | 2019-09-06 | 深圳市网心科技有限公司 | Computing device, blockchain-based resource contribution computing method and storage medium |
| CN110311933A (en) * | 2018-03-20 | 2019-10-08 | 中国移动通信集团有限公司 | A kind of method, apparatus, equipment and the storage medium of equilibrium service traffics |
| CN112559130A (en) * | 2020-12-16 | 2021-03-26 | 恒生电子股份有限公司 | Container distribution method, device, electronic equipment and storage medium |
| CN118158222A (en) * | 2024-05-11 | 2024-06-07 | 中移(苏州)软件技术有限公司 | Load balancer deployment method, device, electronic equipment, storage medium and product |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100542175C (en) * | 2006-01-10 | 2009-09-16 | 华为技术有限公司 | A kind of method for balancing load in multiprocessing unit and system of multiprocessing unit |
| US7640023B2 (en) * | 2006-05-03 | 2009-12-29 | Cisco Technology, Inc. | System and method for server farm resource allocation |
-
2008
- 2008-12-16 CN CN2008102400107A patent/CN101447939B/en active Active
Cited By (40)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101594642B (en) * | 2009-06-18 | 2013-05-08 | 中兴通讯股份有限公司 | Resource allocation method and resource allocation device used for communication network element device |
| CN101945353B (en) * | 2009-07-07 | 2012-10-31 | 中国移动通信集团山东有限公司 | Method and system for remote adjustment of system resources |
| CN101945353A (en) * | 2009-07-07 | 2011-01-12 | 中国移动通信集团山东有限公司 | Method and system for remote adjustment of system resources |
| CN101631360B (en) * | 2009-08-19 | 2012-05-09 | 中兴通讯股份有限公司 | Method, device and system for realizing load balancing |
| CN102469023B (en) * | 2010-11-19 | 2015-10-07 | 中国移动通信集团公司 | Based on the dispatching method of cloud computing, unit and system |
| CN102469023A (en) * | 2010-11-19 | 2012-05-23 | 中国移动通信集团公司 | Dispatching method, unit and system based on cloud computing |
| CN102573072A (en) * | 2010-12-29 | 2012-07-11 | 中国移动通信集团湖南有限公司 | Self-adaptive regulating method and device of network resource allocation as well as communication system |
| CN102111337A (en) * | 2011-03-14 | 2011-06-29 | 浪潮(北京)电子信息产业有限公司 | Method and system for task scheduling |
| CN102111337B (en) * | 2011-03-14 | 2013-05-15 | 浪潮(北京)电子信息产业有限公司 | Method and system for task scheduling |
| CN102324967A (en) * | 2011-05-24 | 2012-01-18 | 中兴通讯股份有限公司 | Method and system for mutual scheduling of equipment resources |
| CN102932015B (en) * | 2011-08-12 | 2015-01-28 | 国家广播电影电视总局广播科学研究院 | Small power short wave scheduling method |
| CN102932015A (en) * | 2011-08-12 | 2013-02-13 | 国家广播电影电视总局广播科学研究院 | Small power short wave scheduling method |
| CN102891881A (en) * | 2012-07-09 | 2013-01-23 | 北京中创信测科技股份有限公司 | Method for implementing equivalence and balance of nodes under cloud environment |
| CN102932281A (en) * | 2012-10-31 | 2013-02-13 | 华为技术有限公司 | Method and equipment for dynamic allocation of resources |
| CN102932281B (en) * | 2012-10-31 | 2015-08-26 | 华为技术有限公司 | A kind of dynamic allocation method of resource and equipment |
| CN103841126B (en) * | 2012-11-20 | 2017-11-21 | 中国移动通信集团辽宁有限公司 | The concocting method and device of a kind of shared resource |
| CN103841126A (en) * | 2012-11-20 | 2014-06-04 | 中国移动通信集团辽宁有限公司 | Shared resource deployment method and device |
| CN103703830A (en) * | 2013-05-31 | 2014-04-02 | 华为技术有限公司 | Physical resource adjustment method, device and controller |
| US10142997B2 (en) | 2013-05-31 | 2018-11-27 | Huawei Technologies Co., Ltd. | Method and apparatus for adjusting physical resource, and controller |
| WO2014190557A1 (en) * | 2013-05-31 | 2014-12-04 | 华为技术有限公司 | Physical resource adjustment method, device and controller |
| CN103703830B (en) * | 2013-05-31 | 2017-11-17 | 华为技术有限公司 | A kind of physical resource adjustment, device and controller |
| CN103617086A (en) * | 2013-11-20 | 2014-03-05 | 东软集团股份有限公司 | Parallel computation method and system |
| CN103617086B (en) * | 2013-11-20 | 2017-02-08 | 东软集团股份有限公司 | Parallel computation method and system |
| CN103973586A (en) * | 2014-05-14 | 2014-08-06 | 江苏省无线电科学研究所有限公司 | Method for achieving automatic meteorological station wireless networking load balance |
| CN105450537A (en) * | 2014-09-02 | 2016-03-30 | 阿尔卡特朗讯 | Method and device for determining node load information, as well as carrying out load balancing and overload control based thereon |
| CN105656973A (en) * | 2014-11-25 | 2016-06-08 | 中国科学院声学研究所 | Distributed method and system for scheduling tasks in node group |
| CN105656973B (en) * | 2014-11-25 | 2018-11-13 | 中国科学院声学研究所 | Method for scheduling task and system in a kind of distributed node group |
| CN104780210A (en) * | 2015-04-13 | 2015-07-15 | 杭州华三通信技术有限公司 | Load balancing method and device |
| CN105389212A (en) * | 2015-10-21 | 2016-03-09 | 浪潮(北京)电子信息产业有限公司 | Job assigning method and apparatus |
| CN106815061A (en) * | 2015-12-01 | 2017-06-09 | 阿里巴巴集团控股有限公司 | A kind of method for processing business and device |
| CN105491138A (en) * | 2015-12-15 | 2016-04-13 | 国网智能电网研究院 | Load rate based graded triggering distributed load scheduling method |
| CN105491138B (en) * | 2015-12-15 | 2020-01-24 | 国网智能电网研究院 | Distributed load scheduling method based on load rate graded triggering |
| WO2018233299A1 (en) * | 2017-06-22 | 2018-12-27 | 平安科技（深圳）有限公司 | Method, apparatus and device for scheduling processor, and medium |
| WO2019165836A1 (en) * | 2018-02-28 | 2019-09-06 | 深圳市网心科技有限公司 | Computing device, blockchain-based resource contribution computing method and storage medium |
| CN110311933A (en) * | 2018-03-20 | 2019-10-08 | 中国移动通信集团有限公司 | A kind of method, apparatus, equipment and the storage medium of equilibrium service traffics |
| CN110175074A (en) * | 2019-04-18 | 2019-08-27 | 北京奇艺世纪科技有限公司 | Load-balancing method and server, load unit, service processing apparatus and medium |
| CN112559130A (en) * | 2020-12-16 | 2021-03-26 | 恒生电子股份有限公司 | Container distribution method, device, electronic equipment and storage medium |
| CN112559130B (en) * | 2020-12-16 | 2024-01-19 | 恒生电子股份有限公司 | Container distribution method, device, electronic equipment and storage medium |
| CN118158222A (en) * | 2024-05-11 | 2024-06-07 | 中移(苏州)软件技术有限公司 | Load balancer deployment method, device, electronic equipment, storage medium and product |
| CN118158222B (en) * | 2024-05-11 | 2024-08-09 | 中移(苏州)软件技术有限公司 | Load balancer deployment method, device, electronic equipment, storage medium and product |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101447939B (en) | 2011-09-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101447939B (en) | Functional distribution method and load balancer | |
| CN102075425B (en) | Business allocation method and load balancer | |
| CN110858161B (en) | Resource allocation method, device, system, equipment and medium | |
| CN102611735B (en) | A kind of load-balancing method of application service and system | |
| CN103401947A (en) | Method and device for allocating tasks to multiple servers | |
| CN102546782B (en) | Distribution system and data operation method thereof | |
| CN108268317B (en) | Resource allocation method and device | |
| US9703610B2 (en) | Extensible centralized dynamic resource distribution in a clustered data grid | |
| CN105141541A (en) | Task-based dynamic load balancing scheduling method and device | |
| CN103227838B (en) | A kind of multi-load equilibrium treatment apparatus and method | |
| CN106471773A (en) | The distribution of integrated form global resource and load balance | |
| US8356098B2 (en) | Dynamic management of workloads in clusters | |
| CN105721350A (en) | Intelligent bandwidth allocation method and apparatus | |
| CN105468458A (en) | Resource scheduling method and system of computer cluster | |
| CN102917077A (en) | Resource allocation method in cloud computing system | |
| CN110244901B (en) | Task allocation method and device and distributed storage system | |
| CN108282526B (en) | Dynamic allocation method and system for servers between double clusters | |
| CN105450784B (en) | The device and method of message distribution consumption node into MQ | |
| CN114816241B (en) | Disk allocation method and device of distributed storage cluster and distributed storage cluster | |
| CN104715044A (en) | Distributed system and data manipulation method thereof | |
| CN101853185B (en) | Blade server and service dispatching method thereof | |
| CN103905473A (en) | Cloud computing system, load balancing system, load balancing method and device | |
| CN103260196A (en) | Method, device and system of controlling of transmission bandwidth | |
| CN115840649A (en) | Method and device for allocating partitioned capacity block type virtual resources, storage medium and terminal | |
| CN104834565A (en) | Method and device for dynamically deploying system services |
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 |