CN106210028B - A kind of server prevents method, server and the system of overload - Google Patents
A kind of server prevents method, server and the system of overload Download PDFInfo
- Publication number
- CN106210028B CN106210028B CN201610526094.5A CN201610526094A CN106210028B CN 106210028 B CN106210028 B CN 106210028B CN 201610526094 A CN201610526094 A CN 201610526094A CN 106210028 B CN106210028 B CN 106210028B
- Authority
- CN
- China
- Prior art keywords
- concurrent request
- storage server
- server
- handling capacity
- request amount
- 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 29
- 230000005540 biological transmission Effects 0.000 claims abstract description 22
- 238000004891 communication Methods 0.000 claims description 4
- 230000008569 process Effects 0.000 description 8
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000006185 dispersion Substances 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1097—Protocols in which an application is distributed across nodes in the network for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Computer And Data Communications (AREA)
Abstract
The technical issues of the embodiment of the invention discloses method, server and systems that a kind of server prevents overload, solve current distributed memory system in high load, and the storage server as caused by excessive concurrent request amount is collapsed.The method that server of the embodiment of the present invention prevents overload includes: constantly to send concurrent request to storage server according to initial concurrent request amount window；The handling capacity returned packet quantity and calculate storage server returned according to storage server, and practical concurrent request amount window is determined according to handling capacity；Using practical concurrent request amount window size as concurrent request threshold value, until the concurrent request of transmission reaches concurrent request threshold value, then stop sending concurrent request to storage server.
Description
Technical field
The present invention relates to technical field of data processing more particularly to a kind of server prevent overload method, server and
System.
Background technique
Distributed memory system is that data dispersion is stored in more independent equipment.Traditional network store system
All data, bottleneck and reliability and peace of the storage server as system performance are stored using the storage server of concentration
The focus of full property, is not able to satisfy the needs of Mass storage application.Distributed network storage system uses expansible system knot
Structure shares storage load using more storage servers, positions storage information using location server, it not only increases system
Reliability, availability and access efficiency, be also easy to extend.
Current distributed memory system is two-tiered structure, and first layer is proxy server, and the second layer is message synchronization collection
Group and storage service cluster, the handling capacity when high load, it will since excessive concurrent request amount causes storage server to collapse
Routed technical problem.
Summary of the invention
A kind of server provided in an embodiment of the present invention prevents method, server and the system of overload, solves current
Distributed memory system in high load, the storage server as caused by excessive concurrent request amount collapse the technical issues of.
The method that a kind of server provided in an embodiment of the present invention prevents overload, comprising:
Concurrent request constantly is sent to storage server according to initial concurrent request amount window；
The handling capacity of storage server is calculated according to the packet quantity of returning that the storage server returns, and is gulped down according to described
The amount of spitting determines practical concurrent request amount window；
Using the practical concurrent request amount window size as concurrent request threshold value, until the concurrent request of transmission reaches
To the concurrent request threshold value, then stop sending the concurrent request to the storage server.
Optionally, it is also wrapped before constantly sending concurrent request to storage server according to initial concurrent request amount window size
It includes:
The size of the initial concurrent request amount window is set for infinity.
Optionally, it is specifically wrapped according to the handling capacity that time packet quantity that the storage server returns calculates storage server
It includes:
Storage server is calculated by the packet quantity of returning that preset time period and the concurrent request sent return
Handling capacity.
Optionally, determine that practical concurrent request amount window specifically includes according to the handling capacity:
When the request number of times of the concurrent request reaches preset request number of times threshold values, it is determined that the storage server is
High load；
Presently described practical concurrent request amount window is set by the current calculated handling capacity.
Optionally, presently described practical concurrent request amount window is set by the current calculated handling capacity specifically to wrap
It includes:
During constantly sending concurrent request to the storage server, to the handling capacity of the storage server
It is calculated in real time；
Judge whether the calculated handling capacity is greater than the practical concurrent request amount window in real time, if so, updating
The practical concurrent request amount window is the real-time calculated handling capacity.
A kind of server provided in an embodiment of the present invention, comprising:
Concurrent request transmission unit, for constantly concurrently being asked to storage server transmission according to initial concurrent request amount window
It asks；
Computing unit, the packet quantity of returning for being returned according to the storage server calculate handling up for storage server
Amount, and practical concurrent request amount window is determined according to the handling capacity；
System load capacity unit is determined, for using the practical concurrent request amount window size as concurrent request threshold
Value then stops sending the concurrent request and deposit to described until the concurrent request of transmission reaches the concurrent request threshold value
Store up server.
Optionally, server further include:
Setting unit, for the size of the initial concurrent request amount window to be arranged as infinity.
Optionally, computing unit specifically includes:
Computation subunit, the packet quantity of returning for being returned by preset time period and the concurrent request sent calculate
The handling capacity of storage server out.
Optionally, computing unit is specific further include:
High load determines subelement, for when the request number of times of the concurrent request reaches preset request number of times threshold values,
Then determine that the storage server is high load；
Practical concurrent request amount determines subelement, for setting presently described reality for the current calculated handling capacity
Border concurrent request amount window.
Optionally, practical concurrent request amount determines that subelement specifically includes:
Real-time computing module is used for during constantly sending concurrent request to the storage server, to the storage
The handling capacity of server is calculated in real time；
Judgment module, for judging whether the real-time calculated handling capacity is greater than the practical concurrent request amount window
Mouthful, if so, updating the practical concurrent request amount window is the real-time calculated handling capacity.
The system that a kind of server provided in an embodiment of the present invention prevents overload, comprising:
Server described in any one referred in several storage servers and the present embodiment；
Several storage servers and server foundation have communication connection relationship.
As can be seen from the above technical solutions, the embodiment of the present invention has the advantage that
A kind of server provided in an embodiment of the present invention prevents method, server and the system of overload, wherein server is anti-
The method only overloaded includes: constantly to send concurrent request to storage server according to initial concurrent request amount window；According to storage
The handling capacity returned packet quantity and calculate storage server that server returns, and practical concurrent request amount window is determined according to handling capacity
Mouthful；Using practical concurrent request amount window size as concurrent request threshold value, until the concurrent request of transmission reaches concurrent request threshold
Value then stops sending concurrent request to storage server.In the present embodiment, by according to initial concurrent request amount window constantly to
Storage server sends concurrent request, then calculates handling up for storage server according to the packet quantity of returning that storage server returns
Amount, and practical concurrent request amount window is determined according to handling capacity, finally using practical concurrent request amount window size as concurrently asking
Threshold value is sought, until the concurrent request of transmission reaches concurrent request threshold value, then stops sending concurrent request to storage server, realizes
The load capacity for first assessing storage server solves current further according to the traffic volume of load capacity control concurrent request
Distributed memory system in high load, the storage server as caused by excessive concurrent request amount collapse the technical issues of.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention without any creative labor, may be used also for those of ordinary skill in the art
To obtain other attached drawings according to these attached drawings.
Fig. 1 is the process signal of the one embodiment for the method that a kind of server provided in an embodiment of the present invention prevents overload
Figure；
Fig. 2 is that the process of another embodiment of the method that a kind of server provided in an embodiment of the present invention prevents overload is shown
It is intended to；
Fig. 3 is a kind of structural schematic diagram of one embodiment of server provided in an embodiment of the present invention；
Fig. 4 is a kind of structural schematic diagram of another embodiment of server provided in an embodiment of the present invention；
Fig. 5 is the structural representation of the one embodiment for the system that a kind of server provided in an embodiment of the present invention prevents overload
Figure；
Fig. 6 is distributed memory system configuration diagram.
Specific embodiment
A kind of server provided in an embodiment of the present invention prevents method, server and the system of overload, solves current
Distributed memory system in high load, the storage server as caused by excessive concurrent request amount collapse the technical issues of.
In order to make the invention's purpose, features and advantages of the invention more obvious and easy to understand, below in conjunction with the present invention
Attached drawing in embodiment, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that disclosed below
Embodiment be only a part of the embodiment of the present invention, and not all embodiment.Based on the embodiments of the present invention, this field
Those of ordinary skill's all other embodiment obtained without making creative work, belongs to protection of the present invention
Range.
Referring to Fig. 1, a kind of server provided in an embodiment of the present invention prevents one embodiment of method of overload from including:
101, concurrent request constantly is sent to storage server according to initial concurrent request amount window；
In the present embodiment, when the processing capacity for needing to assess entire distributed memory system system to prevent locking system collapse,
Concurrent request constantly is sent to storage server firstly the need of according to initial concurrent request amount window.
102, the handling capacity returned packet quantity and calculate storage server returned according to storage server, and according to handling capacity
Determine practical concurrent request amount window；
After constantly sending concurrent request to storage server according to initial concurrent request amount window, need according to storage
The handling capacity returned packet quantity and calculate storage server that server returns, and practical concurrent request amount window is determined according to handling capacity
Mouthful.
103, using practical concurrent request amount window size as concurrent request threshold value, until the concurrent request of transmission reaches simultaneously
Hair request threshold value then stops sending concurrent request to storage server.
When the packet quantity of returning returned according to storage server calculates the handling capacity of storage server, and it is true according to handling capacity
After fixed practical concurrent request amount window, need using practical concurrent request amount window size as concurrent request threshold value, Zhi Daofa
The concurrent request sent reaches concurrent request threshold value, then stops sending concurrent request to storage server.
In the present embodiment, by constantly sending concurrent request to storage server according to initial concurrent request amount window, so
The handling capacity of storage server is calculated according to the packet quantity of returning that storage server returns afterwards, and determines reality simultaneously according to handling capacity
Request amount window is sent out, finally using practical concurrent request amount window size as concurrent request threshold value, until the concurrent request of transmission
Reach concurrent request threshold value, then stops sending concurrent request to storage server, realize the load for first assessing storage server
Ability solves current distributed memory system in high load further according to the traffic volume of load capacity control concurrent request,
The technical issues of storage server as caused by excessive concurrent request amount is collapsed.
The above is that the process of the method for overload is prevented to be described in detail to server, will be carried out below to detailed process
Detailed description, referring to Fig. 2, a kind of server provided in an embodiment of the present invention prevents another embodiment of the method for overload
Include:
201, the size that initial concurrent request amount window is arranged is infinity；
In the present embodiment, when the processing capacity for needing to assess entire distributed memory system system to prevent locking system collapse,
It is infinity firstly the need of the size that initial concurrent request amount window is arranged.
202, concurrent request constantly is sent to storage server according to initial concurrent request amount window；
After the size that initial concurrent request amount window is arranged is infinitely great, need according to initial concurrent request amount window
Constantly concurrent request is sent to storage server.
It should be noted that concurrent request amount window indicates the quantity that can transmit a request to external service simultaneously.
203, storage server is calculated by the packet quantity of returning that preset time period and the concurrent request sent return
Handling capacity；
After constantly sending concurrent request to storage server according to initial concurrent request amount window, need by preset
The handling capacity returned packet quantity and calculate storage server that period and the concurrent request sent return.
204, when the request number of times of concurrent request reaches preset request number of times threshold values, it is determined that storage server is high negative
It carries；
When the packet quantity of returning returned by preset time period and the concurrent request sent calculates gulping down for storage server
After the amount of spitting, if the request number of times of concurrent request reaches preset request number of times threshold values, it is determined that storage server is high load.
205, during constantly sending concurrent request to storage server, the handling capacity of storage server is carried out real-time
It calculates；
When the request number of times of concurrent request reaches preset request number of times threshold values, it is determined that storage server be high load it
Afterwards, it needs to set currently practical concurrent request amount window for current calculated handling capacity, it is specifically currently calculated to gulp down
The amount of spitting is set as currently practical concurrent request amount window, can be during constantly sending concurrent request to storage server,
The handling capacity of storage server is calculated in real time.
206, judge whether calculated handling capacity is greater than practical concurrent request amount window in real time, if so, thening follow the steps
207；
During constantly sending concurrent request to storage server, the handling capacity of storage server is counted in real time
After calculation, need to judge whether real-time calculated handling capacity is greater than practical concurrent request amount window, if so, thening follow the steps
207。
207, updating practical concurrent request amount window is real-time calculated handling capacity；
When judge in real time calculated handling capacity be greater than practical concurrent request amount window, then update practical concurrent request amount
Window is real-time calculated handling capacity.
208, using practical concurrent request amount window size as concurrent request threshold value, until the concurrent request of transmission reaches simultaneously
Hair request threshold value then stops sending concurrent request to storage server.
When the packet quantity of returning returned according to storage server calculates the handling capacity of storage server, and it is true according to handling capacity
After fixed practical concurrent request amount window, need using practical concurrent request amount window size as concurrent request threshold value, Zhi Daofa
The concurrent request sent reaches concurrent request threshold value, then stops sending concurrent request to storage server.
The antioverloading process of server is described in detail with a concrete application scene below, as shown in fig. 6, using
Example include:
The practical framework of Fig. 6 is that distributed memory system horsetable (packet type framework table) uses two-tiered structure,
First layer service is horse_proxy (proxy server)；Second layer service be sync_broker (message synchronization cluster) and
Storage (storage service cluster).
In the distributed memory system, horse_proxy process performance is substantially better than sync_broker and storage,
Therefore the performance bottle of system is tightly in sync_broker and storage.Horse_proxy service is by calculating external service
Handling capacity when (sync_broker and storage) is in high load controls concurrent request amount window, to prevent excessive
Concurrent request amount pressure is across external service, to guarantee the normal service ability of whole system.Wherein handling capacity refers to the unit time
The interior number of request that can be handled；Concurrent request refers to the request amount sent in the unit time.
Following step is for horse_proxy and storage is serviced:
(1) concurrent request amount window W indicates the quantity that can transmit a request to external service simultaneously, for assessing external clothes
The ability that business can be handled simultaneously.Concurrent request amount window W is initialized to infinity, i.e., at the beginning with no restriction；horse_
Proxy service is constantly serviced to storage with W window size sends request.
(2) horse_proxy service is serviced by the packet amount R of returning of T in a period of time (such as 10 seconds) to calculate storage
Handling capacity: handling capacity=R/T
(3) when request timed out number reaches certain threshold values V, horse_proxy service thinks storage service in height
Load, and using handling capacity at this time as concurrent request amount window W；
(4) as long as handling capacity is greater than concurrent request amount window W, horse_proxy service is concurrently asked with handling capacity update
The amount of asking window W；
(5) as long as concurrent request amount window W is less than, horse_proxy service can be serviced to send to storage and be asked
It asks, has expired and just stopped sending.
In (3) if in V it is too small, external service may be will mistakenly believe that and be in high load, if too big, may not enough and
Shi Faxian storage service is in the state of high load, so V value will be with being arranged according to actual conditions in tradeoff.
Horse_proxy service service ability better than under storage service scenario, it is ensured that do not press across
Storage is serviced and is kept best service ability, to guarantee whole system nonoverload, to stop working.It needs to illustrate
, fluctuation may be generated in state change, but this fluctuation is at one or two in T seconds, and state change number is few.
In horsetable two-tiered structure system, the case where service using first layer service processing performance better than the second layer
Under, first layer service controls concurrent request amount by assessment second layer service ability, to guarantee the positive informal dress of second service
Business ability, the final normal service ability for guaranteeing whole system, even if in the case of an overload.
In the present embodiment, by constantly sending concurrent request to storage server according to initial concurrent request amount window, so
The handling capacity of storage server is calculated according to the packet quantity of returning that storage server returns afterwards, and determines reality simultaneously according to handling capacity
Request amount window is sent out, finally using practical concurrent request amount window size as concurrent request threshold value, until the concurrent request of transmission
Reach concurrent request threshold value, then stops sending concurrent request to storage server, realize the load for first assessing storage server
Ability solves current distributed memory system in high load further according to the traffic volume of load capacity control concurrent request,
The technical issues of storage server as caused by excessive concurrent request amount is collapsed, and constantly calculate handling capacity and current
The data of concurrent request amount window determine last practical concurrent request amount window, the more accurate service ability of system
Assessment, ensure that the normal service ability of whole system.
Referring to Fig. 3, a kind of one embodiment of server provided in an embodiment of the present invention includes:
Concurrent request transmission unit 301, for constantly being sent simultaneously to storage server according to initial concurrent request amount window
Hair request；
Computing unit 302, the handling capacity returned packet quantity and calculate storage server for being returned according to storage server,
And practical concurrent request amount window is determined according to handling capacity；
It determines system load capacity unit 303, is used for using practical concurrent request amount window size as concurrent request threshold value,
Until the concurrent request of transmission reaches concurrent request threshold value, then stop sending concurrent request to storage server.
In the present embodiment, constantly taken to storage by concurrent request transmission unit 301 according to initial concurrent request amount window
Business device sends concurrent request, and then computing unit 302 calculates storage server according to the packet quantity of returning that storage server returns
Handling capacity, and practical concurrent request amount window is determined according to handling capacity, finally determining system load capacity unit 303 is with reality
Concurrent request amount window size is as concurrent request threshold value, until the concurrent request of transmission reaches concurrent request threshold value, then stops
Concurrent request is sent to storage server, the load capacity for first assessing storage server is realized, is controlled further according to load capacity
The traffic volume of concurrent request solves current distributed memory system in high load, since excessive concurrent request amount is led
The technical issues of storage server collapse of cause.
The above is that each unit of server is described in detail, and sub-unit is described in detail below, asks
Refering to Fig. 4, a kind of another embodiment of server provided in an embodiment of the present invention includes:
Setting unit 401, for the size of initial concurrent request amount window to be arranged as infinity.
Concurrent request transmission unit 402, for constantly being sent simultaneously to storage server according to initial concurrent request amount window
Hair request；
Computing unit 403, the handling capacity returned packet quantity and calculate storage server for being returned according to storage server,
And practical concurrent request amount window is determined according to handling capacity；
Computing unit 403 specifically includes:
Computation subunit 4031, the packet quantity of returning for being returned by preset time period and the concurrent request sent calculate
The handling capacity of storage server out.
High load determines subelement 4032, when reaching preset request number of times threshold values for the request number of times when concurrent request,
Then determine that storage server is high load；
Practical concurrent request amount determines subelement 4033, for by current calculated handling capacity be set as it is currently practical simultaneously
Send out request amount window；
Practical concurrent request amount determines that subelement 4033 specifically includes:
Real-time computing module 4031a, is used for during constantly sending concurrent request to storage server, to storage service
The handling capacity of device is calculated in real time；
Judgment module 4032b, for judging whether real-time calculated handling capacity is greater than practical concurrent request amount window, if
It is to update practical concurrent request amount window then for real-time calculated handling capacity.
It determines system load capacity unit 404, is used for using practical concurrent request amount window size as concurrent request threshold value,
Until the concurrent request of transmission reaches concurrent request threshold value, then stop sending concurrent request to storage server.
In the present embodiment, constantly taken to storage by concurrent request transmission unit 402 according to initial concurrent request amount window
Business device sends concurrent request, and then computing unit 403 calculates storage server according to the packet quantity of returning that storage server returns
Handling capacity, and practical concurrent request amount window is determined according to handling capacity, finally determining system load capacity unit 404 is with reality
Concurrent request amount window size is as concurrent request threshold value, until the concurrent request of transmission reaches concurrent request threshold value, then stops
Concurrent request is sent to storage server, the load capacity for first assessing storage server is realized, is controlled further according to load capacity
The traffic volume of concurrent request solves current distributed memory system in high load, since excessive concurrent request amount is led
The technical issues of storage server collapse of cause, and constantly calculate the data of handling capacity and current concurrent request amount window
Determine last practical concurrent request amount window, the more accurate assessment of the service ability of system ensure that whole system
Normal service ability.
Referring to Fig. 5, a kind of server provided in the embodiment of the present invention prevents one embodiment packet of the system of overload
It includes:
The server 52 referred in several storage servers 51 and Fig. 3 and Fig. 4 embodiment；
Several storage servers 51 and the foundation of server 52 have communication connection relationship.
It should be noted that storage server 51 can be storage service cluster, further comprise message synchronization cluster,
Several clients.
In the two-tiered structure of distributed memory system, it is better than the feelings of second layer service using first layer service processing performance
Under condition, first layer service controls concurrent request amount by assessment second layer service ability, to guarantee the normal of second service
Service ability, the final normal service ability for guaranteeing whole system, even if in the case of an overload.
It is apparent to those skilled in the art that for convenience and simplicity of description, the system of foregoing description,
The specific work process of device and unit, can refer to corresponding processes in the foregoing method embodiment, and details are not described herein.
In several embodiments provided herein, it should be understood that disclosed system, device and method can be with
It realizes by another way.For example, the apparatus embodiments described above are merely exemplary, for example, the unit
It divides, only a kind of logical function partition, there may be another division manner in actual implementation, such as multiple units or components
It can be combined or can be integrated into another system, or some features can be ignored or not executed.Another point, it is shown or
The mutual coupling, direct-coupling or communication connection discussed can be through some interfaces, the indirect coupling of device or unit
It closes or communicates to connect, can be electrical property, mechanical or other forms.
The unit as illustrated by the separation member may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multiple
In network unit.It can select some or all of unit therein according to the actual needs to realize the mesh of this embodiment scheme
's.
It, can also be in addition, the functional units in various embodiments of the present invention may be integrated into one processing unit
It is that each unit physically exists alone, can also be integrated in one unit with two or more units.Above-mentioned integrated list
Member both can take the form of hardware realization, can also realize in the form of software functional units.
If the integrated unit is realized in the form of SFU software functional unit and sells or use as independent product
When, it can store in a computer readable storage medium.Based on this understanding, technical solution of the present invention is substantially
The all or part of the part that contributes to existing technology or the technical solution can be in the form of software products in other words
It embodies, which is stored in a storage medium, including some instructions are used so that a computer
Equipment (can be personal computer, server or the network equipment etc.) executes the complete of each embodiment the method for the present invention
Portion or part steps.And storage medium above-mentioned includes: USB flash disk, mobile hard disk, read-only memory (ROM, Read-Only
Memory), random access memory (RAM, Random Access Memory), magnetic or disk etc. are various can store journey
The medium of sequence code.
The above, the above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations；Although referring to before
Stating embodiment, invention is explained in detail, those skilled in the art should understand that: it still can be to preceding
Technical solution documented by each embodiment is stated to modify or equivalent replacement of some of the technical features；And these
It modifies or replaces, the spirit and scope for technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution.
Claims (5)
1. a kind of method that server prevents overload characterized by comprising
Concurrent request constantly is sent to storage server according to initial concurrent request amount window；
The handling capacity returned packet quantity and calculate storage server returned according to the storage server, and according to the handling capacity
Determine practical concurrent request amount window；
Using the practical concurrent request amount window size as concurrent request threshold value, until the concurrent request of transmission reaches institute
Concurrent request threshold value is stated, then stops sending the concurrent request to the storage server；
It is specifically included according to the handling capacity that time packet quantity that the storage server returns calculates storage server:
Handling up for storage server is calculated by the packet quantity of returning that preset time period and the concurrent request sent return
Amount；
The handling capacity of the storage server is equal to the packet amount R of returning divided by the preset time period T；
Determine that practical concurrent request amount window specifically includes according to the handling capacity:
When the request number of times of the concurrent request reaches preset request number of times threshold values, it is determined that the storage server is high negative
It carries；
Presently described practical concurrent request amount window is set by the current calculated handling capacity；
Presently described practical concurrent request amount window is set by the current calculated handling capacity to specifically include:
During constantly sending concurrent request to the storage server, the handling capacity of the storage server is carried out
It calculates in real time；
Judge whether the calculated handling capacity is greater than the practical concurrent request amount window in real time, if so, described in updating
Practical concurrent request amount window is the real-time calculated handling capacity.
2. the method that server according to claim 1 prevents overload, which is characterized in that according to initial concurrent request amount window
Before mouth size constantly sends concurrent request to storage server further include:
The size of the initial concurrent request amount window is set for infinity.
3. a kind of server characterized by comprising
Concurrent request transmission unit, for constantly sending concurrent request to storage server according to initial concurrent request amount window；
Computing unit, the handling capacity returned packet quantity and calculate storage server for being returned according to the storage server, and
Practical concurrent request amount window is determined according to the handling capacity；
It determines system load capacity unit, is used for using the practical concurrent request amount window size as concurrent request threshold value, directly
The concurrent request to transmission reaches the concurrent request threshold value, then stops sending the concurrent request to the storage service
Device；
Computing unit specifically includes:
Computation subunit, time packet quantity for being returned by preset time period and the concurrent request sent, which calculates, deposits
Store up the handling capacity of server；
The handling capacity of the storage server is equal to the packet amount R of returning divided by the preset time period T；
Computing unit is specific further include:
High load determines subelement, for when the request number of times of the concurrent request reaches preset request number of times threshold values, then really
The fixed storage server is high load；
Practical concurrent request amount determines subelement, for setting presently described reality simultaneously for the current calculated handling capacity
Send out request amount window；
Practical concurrent request amount determines that subelement specifically includes:
Real-time computing module is used for during constantly sending concurrent request to the storage server, to the storage service
The handling capacity of device is calculated in real time；
Judgment module, for judging whether the real-time calculated handling capacity is greater than the practical concurrent request amount window, if
It is to update the practical concurrent request amount window then for the real-time calculated handling capacity.
4. server according to claim 3, which is characterized in that server further include:
Setting unit, for the size of the initial concurrent request amount window to be arranged as infinity.
5. the system that a kind of server prevents overload characterized by comprising
Several storage servers, and the server as described in any one of claim 3 and 4；
Several storage servers and server foundation have communication connection relationship.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610526094.5A CN106210028B (en) | 2016-07-05 | 2016-07-05 | A kind of server prevents method, server and the system of overload |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610526094.5A CN106210028B (en) | 2016-07-05 | 2016-07-05 | A kind of server prevents method, server and the system of overload |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106210028A CN106210028A (en) | 2016-12-07 |
| CN106210028B true CN106210028B (en) | 2019-09-06 |
Family
ID=57465462
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610526094.5A Active CN106210028B (en) | 2016-07-05 | 2016-07-05 | A kind of server prevents method, server and the system of overload |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106210028B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108255599A (en) * | 2016-12-29 | 2018-07-06 | 北京京东尚科信息技术有限公司 | Based on the treating method and apparatus largely asked |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101882105A (en) * | 2010-06-01 | 2010-11-10 | 华南理工大学 | Method for testing response time of Web page under concurrent environment |
| CN102148759A (en) * | 2011-04-01 | 2011-08-10 | 许旭 | Method for saving export bandwidth of backbone network by cache acceleration system |
| CN103236956A (en) * | 2013-04-18 | 2013-08-07 | 神州数码网络(北京)有限公司 | Method and switch for testing throughput of communication equipment |
| CN105207832A (en) * | 2014-06-13 | 2015-12-30 | 腾讯科技（深圳）有限公司 | Server stress testing method and device |
| CN105701207A (en) * | 2016-01-12 | 2016-06-22 | 腾讯科技（深圳）有限公司 | Request quantity forecast method of resource and application recommendation method and device |
-
2016
- 2016-07-05 CN CN201610526094.5A patent/CN106210028B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101882105A (en) * | 2010-06-01 | 2010-11-10 | 华南理工大学 | Method for testing response time of Web page under concurrent environment |
| CN102148759A (en) * | 2011-04-01 | 2011-08-10 | 许旭 | Method for saving export bandwidth of backbone network by cache acceleration system |
| CN103236956A (en) * | 2013-04-18 | 2013-08-07 | 神州数码网络(北京)有限公司 | Method and switch for testing throughput of communication equipment |
| CN105207832A (en) * | 2014-06-13 | 2015-12-30 | 腾讯科技（深圳）有限公司 | Server stress testing method and device |
| CN105701207A (en) * | 2016-01-12 | 2016-06-22 | 腾讯科技（深圳）有限公司 | Request quantity forecast method of resource and application recommendation method and device |
Non-Patent Citations (1)
| Title |
|---|
| web系统性能测试工具的研究;邵燕琳;《中国优秀硕士学位论文全文库 信息科技辑》;20071215;正文第2、12-13、27、39-53页 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106210028A (en) | 2016-12-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109618002B (en) | Micro-service gateway optimization method, device and storage medium | |
| CN103440202B (en) | A kind of communication means based on RDMA, system and communication equipment | |
| CN105897836A (en) | Back source request processing method and device | |
| CN105554102A (en) | Elastic expansion method based on container cluster and application system thereof | |
| CN106020926B (en) | A kind of method and device transmitted for data in virtual switch technology | |
| EP3745678B1 (en) | Storage system, and method and apparatus for allocating storage resources | |
| CN102281190A (en) | Networking method for load balancing apparatus, server and client access method | |
| CN113742135B (en) | Data backup method, device and computer readable storage medium | |
| CN106878197A (en) | A kind of management system and method for the transmission of cloud platform message | |
| US11316916B2 (en) | Packet processing method, related device, and computer storage medium | |
| CN105635083A (en) | Service processing method and service processing system based on server and client architecture | |
| CN105554125B (en) | A kind of method and its system for realizing webpage fit using CDN | |
| CN107529186A (en) | The method and system of channel transmission upstream data, client, server | |
| CN105978938A (en) | Service processing equipment service status determining method and scheduling equipment | |
| CN106304154B (en) | A kind of data transmission method and PDCP entity of PDCP entity | |
| CN106210028B (en) | A kind of server prevents method, server and the system of overload | |
| CN113726847B (en) | Network system, network segmentation method and electronic equipment | |
| CN108418752A (en) | A kind of creation method and device of aggregation group | |
| CN114116207A (en) | Flow control method, device, equipment and system | |
| CN118041937A (en) | Data access method and device of storage device | |
| WO2020010507A1 (en) | Method for adjusting communication link, communication device, and unmanned aerial vehicle | |
| CN108337328A (en) | A kind of data exchange system, data uploading method and data download method | |
| CN105099753B (en) | The method of Network Management System and its processing business | |
| CN105450679A (en) | Method and system for performing data cloud storage | |
| CN108632921A (en) | A kind of core net switching method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20210111 Address after: 510000 3108, 79 Wanbo 2nd Road, Nancun Town, Panyu District, Guangzhou City, Guangdong Province Patentee after: GUANGZHOU CUBESILI INFORMATION TECHNOLOGY Co.,Ltd. Address before: 511449 Wanda Plaza, Nancun Town, Panyu District, Guangzhou City, Guangdong Province Patentee before: GUANGZHOU HUADUO NETWORK TECHNOLOGY Co.,Ltd. |
|
| EE01 | Entry into force of recordation of patent licensing contract | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20161207 Assignee: GUANGZHOU HUADUO NETWORK TECHNOLOGY Co.,Ltd. Assignor: GUANGZHOU CUBESILI INFORMATION TECHNOLOGY Co.,Ltd. Contract record no.: X2021440000053 Denomination of invention: Method, server and system for preventing overload of server Granted publication date: 20190906 License type: Common License Record date: 20210208 |