CN101799772B - Kernel dispatching method, kernel backup method and multi-core processor - Google Patents
Kernel dispatching method, kernel backup method and multi-core processor Download PDFInfo
- Publication number
- CN101799772B CN101799772B CN201010115862.0A CN201010115862A CN101799772B CN 101799772 B CN101799772 B CN 101799772B CN 201010115862 A CN201010115862 A CN 201010115862A CN 101799772 B CN101799772 B CN 101799772B
- Authority
- CN
- China
- Prior art keywords
- processor group
- core
- processor
- occupation rate
- resources occupation
- 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
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Power Sources (AREA)
Abstract
The invention discloses a kernel dispatching method and a kernel backup method of a multi-core processor, and the multi-core processor. The method comprises the following steps: acquiring a resource occupancy rate of each processor group in the multi-core processor; selecting a first processor group of which the resource occupancy rate is less than or equal to a first threshold in processor groups; controlling a resource occupied by a part of a transmission face for kernel releases; after the transmission face kernel completes releasing the resource, deleting the transmission face kernel which releases the resource from the first processor group; and setting the transmission face kernel which releases the resource in a standby state. The embodiment of the invention can be applied to the multi-core processor, can reduce the output power consumption of the multi-core processor and can increase the resource utilization rate of the multi-core processor.
Description
Technical field
The present invention relates to field of computer technology, be specifically related to a kind of kernel dispatching method, kernel backup method and polycaryon processor.
Background technology
In communication system, due to increasing and high-performance forwarding demand of need data to be processed, conventionally adopt polycaryon processor to carry out corresponding data retransmission or data processing task, by the concurrent work of multiple kernels, improve execution efficiency.With respect to single core processor, polycaryon processor is by combining multiple kernels (at least two), and multiple kernels can move under lower frequency, and the performance of polycaryon processor significantly exceedes single core processor and power consumption is lower.
In polycaryon processor of the prior art, generally include a chain of command core and at least one forwarding face core, chain of command core has been used for the startup of communication system, the interface to user, the configuration of communication service and the management to forwarding face core, multiple forwarding face core is carried out the concurrent processing to customer service, such as service message forwarding and service message processing etc.
Wherein, polycaryon processor comprises multiple processor groups, and a processor group is made up of multiple forwarding face core.Multiple processor groups are carried out respectively different treatment steps in the processing procedure of service message, have worked in coordination with the processing to service message.
Inventor finds in the process of research prior art, multiple forwarding faces in polycaryon processor in existing communication system are checked customer service and are carried out concurrent processing, can complete the processing of large flow business, but in the time that communication system load is lighter, the resource that partly forwards face core is idle, but polycaryon processor power consumption is constant, communication system energy consumption is higher.
Summary of the invention
The invention provides a kind of in the time that communication system load is lighter, the kernel dispatching method of the idle forwarding face core of release portion, kernel backup method and polycaryon processor.
On the one hand, the kernel dispatching method that the embodiment of the present invention provides, comprising:
Obtain the resources occupation rate of each processor group in polycaryon processor;
Select resources occupation rate in described processor group to be less than or equal to the first processor group of first threshold;
The part forwarding face core of controlling in described first processor group discharges shared resource;
After described forwarding face core releasing resource completes, from described first processor group, delete the forwarding face core of releasing resource;
The forwarding face caryogamy of described releasing resource is set to holding state.
The embodiment of the present invention also provides a kind of kernel dispatching method of polycaryon processor, comprising:
Obtain the resources occupation rate of each processor group in polycaryon processor;
Select the one or three processor group and the second the highest four-processor group of resources occupation rate that in described processor group, resources occupation rate is minimum;
Be less than or equal to described first threshold in the resources occupation rate of described the one or three processor group, when the resources occupation rate of described the second four-processor group is more than or equal to Second Threshold, obtain the first result that resources occupation rate that described first threshold deducts described the one or three processor group obtains and the resources occupation rate of described the second four-processor group and deduct the second result that Second Threshold obtains, described Second Threshold is greater than described first threshold;
In the time that described the first result is more than or equal to the second result, the part forwarding face core of controlling in described the one or three processor group discharges shared resource;
After described forwarding face core releasing resource completes, from described the one or three processor group, delete the forwarding face core of releasing resource;
The forwarding face caryogamy of described releasing resource is set to holding state.
The embodiment of the present invention also provides a kind of kernel backup method, comprising:
Chain of command core obtains the duty of forwarding face core;
When the first forwarding face core in described forwarding face core breaks down, from default backup core, select the first backup core of holding state;
Chain of command core activates described the first backup core and carries out business processing;
Chain of command core restarts described first and forwards face core;
Chain of command core forwards face caryogamy by first after described restarting and is set to holding state.
On the other hand, the embodiment of the present invention also provides a kind of polycaryon processor, comprises at least one processor group and chain of command core, and described processor group comprises forwarding face core, and described chain of command core comprises:
The first acquiring unit, for obtaining the resources occupation rate of the each processor group of polycaryon processor;
The first selected cell, for selecting described processor group resources occupation rate to be less than or equal to the first processor group of first threshold;
Resource release control unit, discharges shared resource for the forwarding face core of controlling described first processor group;
Kernel delete cells after completing at described the first forwarding face core releasing resource, is deleted the forwarding face core of releasing resource from described first processor group;
Kernel setup unit, for being set to holding state by the forwarding face caryogamy of described releasing resource.
The embodiment of the present invention also provides a kind of polycaryon processor, comprising: at least one processor group and chain of command core, and described processor group comprises forwarding face core, described chain of command core comprises:
Second acquisition unit, for obtaining the resources occupation rate of the each processor group of polycaryon processor;
The 3rd selected cell, for selecting first processor group and the second the highest processor group of resources occupation rate that described processor group resources occupation rate is minimum;
The 3rd acquiring unit, for being less than or equal to described first threshold in the resources occupation rate of described first processor group, when the resources occupation rate of described the second processor group is more than or equal to Second Threshold, obtain the first result that resources occupation rate that described first threshold deducts described first processor group obtains and the resources occupation rate of described the second processor group and deduct the second result that Second Threshold obtains, described Second Threshold is greater than described first threshold;
The first trigger element, in the time that described the first result is more than or equal to the second result, triggers the forwarding face core in first processor group described in described resource release control unit controls and discharges shared resource;
Resource release control unit, discharges shared resource for the forwarding face core of controlling described first processor group;
Kernel delete cells after completing at described forwarding face core releasing resource, is deleted the forwarding face core of releasing resource from described first processor group;
Kernel setup unit, for being set to holding state by the forwarding face caryogamy of described releasing resource.
The embodiment of the present invention also provides a kind of polycaryon processor, comprises forwarding face core and chain of command core, and described chain of command core comprises:
State acquiring unit, for obtaining the duty of each forwarding face core;
Backup core selected cell while breaking down, is selected the first backup core of holding state for forwarding face core at first of described forwarding face core from default backup core;
Backup core activates unit, carries out business processing for activating described the first backup core;
Backup core start unit, forwards face core for restarting described first;
Backup core dispensing unit, is set to holding state for first after restarting forwarded to face caryogamy.
In embodiments of the present invention, in chain of command core selection polycaryon processor, resources occupation rate is less than or equal to the first processor group of first threshold, the part in first processor group is forwarded to face caryogamy and be set to holding state.Compared with prior art, when the embodiment of the present invention can the resources occupation rate of processor group be lower in polycaryon processor, reduce the output power consumption of polycaryon processor, improve the resource utilization of polycaryon processor, and then reduce the energy consumption of communication system.
Brief description of the drawings
In order to be illustrated more clearly in the technical scheme in the embodiment of the present invention, below the accompanying drawing of required use during the embodiment of the present invention is described is briefly described, apparently, accompanying drawing in the following describes is only some embodiments of the present invention, for those of ordinary skill in the art, do not paying under the prerequisite of creative work, can also obtain according to these accompanying drawings other accompanying drawing.
Fig. 1 is the process flow diagram of the kernel dispatching method that provides of the embodiment of the present invention one;
Fig. 2 is the process flow diagram of the kernel dispatching method that provides of the embodiment of the present invention two;
Fig. 3 is the process flow diagram of the kernel dispatching method that provides of the embodiment of the present invention three;
Fig. 4 is the method flow diagram that obtains the resources occupation rate of processor group in the embodiment of the present invention;
Fig. 5 is the structural representation of chain of command core in the polycaryon processor that provides of the embodiment of the present invention four;
Fig. 6 is the structural representation of chain of command core in the polycaryon processor that provides of the embodiment of the present invention five;
Fig. 7 is the process flow diagram of the kernel backup method that provides of the embodiment of the present invention seven;
Fig. 8 is the structural representation of chain of command core in the polycaryon processor that provides of the embodiment of the present invention eight.
Embodiment
The embodiment of the present invention provides a kind of kernel dispatching method, kernel backup method and polycaryon processor.The technical scheme of embodiment for a better understanding of the present invention, below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is clearly and completely described, obviously, described embodiment is only the present invention's part embodiment, instead of whole embodiment.Based on the embodiment in the present invention, those of ordinary skill in the art, not making the every other embodiment obtaining under creative work prerequisite, belong to the scope of protection of the invention.
Referring to Fig. 1, Fig. 1 is the process flow diagram of the kernel dispatching method that provides of the embodiment of the present invention one.
In embodiments of the present invention, polycaryon processor comprises a chain of command core and multiple forwarding face core, wherein the multiple forwarding face cores in polycaryon processor form multiple processor groups, each processor group comprises two or more forwarding face core, and the kernel dispatching method that the embodiment of the present invention provides can comprise:
101, obtain the resources occupation rate of each processor group in polycaryon processor.
Concrete, chain of command endorses to obtain the resources occupation rate of each processor group in polycaryon processor.
102, in selection processor group, resources occupation rate is less than or equal to the first processor group of first threshold.
Concrete, chain of command is endorsed to select resources occupation rate to be less than or equal to the processor group of first threshold from multiple processor groups, for expressing conveniently, in embodiments of the present invention such processor group is called to first processor group, first processor group can be one, can be also multiple.Taking the first processed group as a processor group is as example, if the resources occupation rate of first processor group is E1, E1 is less than or equal to first threshold.Wherein, first threshold is the resources occupation rate of the processor group threshold value when lower.
103, the part forwarding face core of controlling in first processor group discharges shared resource.
Concrete, the part that chain of command endorses to control first processor group forwards face core, controls some forwarding face core or multiple forwarding face core discharges shared resource, and reserve part forwards face core and continues to carry out corresponding business.Releasing resource is the first forwarding face core in embodiments of the present invention, for example control the first forwarding face core and discharge the resources such as shared internal memory, lock, serial ports, network interface, the first business that forwards the processing of face core is assigned to other forwarding face core in first processor group and carries out.Wherein, specifically the quantity of the forwarding face core of releasing resource can be determined according to the resources occupation rate of processor group, and chain of command is endorsed to release order to control the first forwarding face core to the first forwarding face core transmission resource and discharged shared resource.
104, after the face of forwarding core releasing resource completes, from first processor group, delete the forwarding face core of releasing resource.
Concrete, chain of command core, after the first forwarding face core releasing resource completes, is deleted the forwarding face core of releasing resource from first processor group, is the first forwarding face core in embodiments of the present invention, no longer to the first forwarding face core distribution services.Wherein, deleting the first forwarding face endorses with by configuring the first forwarding face core for going enabled state to realize.
105, the forwarding face caryogamy of releasing resource is set to holding state.
Concrete, chain of command core, by the forwarding face core of releasing resource, is that the first forwarding face caryogamy is set to holding state in embodiments of the present invention, and first forwards face core will be in low power consumpting state under holding state.For example chain of command core sends dependent instruction to the first forwarding face core and the clock of the first forwarding face core is turned off or the first forwarding face caryogamy is set to low-power consumption mode.
In embodiments of the present invention, in chain of command core selection polycaryon processor, resources occupation rate is less than or equal to the first processor group of first threshold, the part in first processor group is forwarded to face caryogamy and be set to holding state.Compared with prior art, when the embodiment of the present invention can the resources occupation rate of processor group be lower in polycaryon processor, reduce the output power consumption of polycaryon processor, improve the resource utilization of polycaryon processor, and then reduce the energy consumption of communication system.
The kernel dispatching method that the embodiment of the present invention one the provides execution that can circulate, after execution step 105, the resources occupation rate of first processor group will change, chain of command core circulation execution step 101, the resources occupation rate of processor group is maintained the level higher than first threshold by final realization, and then the resource utilization of raising polycaryon processor, the energy consumption of reduction communication system.
Referring to Fig. 2, Fig. 2 is the process flow diagram of the kernel dispatching method that provides of the embodiment of the present invention two.
In embodiments of the present invention, the implementation of step 201-205 is identical with step 101-105 in above-described embodiment one, is no longer repeated in this description at this.
The embodiment of the present invention is being set to holding state (step 205) afterwards by the forwarding face caryogamy of releasing resource, can also comprise:
206, in selection processor group, resources occupation rate is more than or equal to the second processor group of Second Threshold, and Second Threshold is greater than first threshold.
Concrete, chain of command is endorsed to select resources occupation rate to be more than or equal to the processor group of Second Threshold from processor group, for expressing conveniently, in embodiments of the present invention such processor group is called to the second processor group, the second processor group can be one, can be also multiple.Taking the second processed group as one as example, the resources occupation rate of supposing the second processor group is E2, and wherein, Second Threshold is the threshold value of the resources occupation rate of processor group when higher, and Second Threshold is greater than first threshold.
In embodiments of the present invention, when resources occupation rate E2 is more than or equal to Second Threshold, illustrate that the resources occupation rate of the second processor group is higher, it is heavier to load.
207, forwarding face core idle in polycaryon processor is loaded in the second processor group, the forwarding face core of start-up loading is carried out business processing.
Concrete, chain of command core is loaded into forwarding face core idle in polycaryon processor in the second processor group that resources occupation rate is the highest, for example, to the idle resource such as forwarding face core storage allocation, lock.Wherein, chain of command core load the quantity of idle forwarding face core can be for one or more.Chain of command core starts the second forwarding face core and carries out the business processing in the second processor group.For example the second forwarding face core is carried out to hardware initialization, then receive corresponding service message, and service message is processed.
In embodiments of the present invention, polycaryon processor can be set to idle core by some forwarding face core wherein after powering on, so that chain of command core selects idle core to be loaded in the processor group that resources occupation rate is higher.
In embodiments of the present invention, chain of command core selects resources occupation rate to be more than or equal to the second processor group of Second Threshold, and in the second processor group, loading idle forwarding face core, the forwarding face core of start-up loading is carried out business processing, thereby alleviates the processing load of the second processor group.Compared with prior art, the embodiment of the present invention not only can be in the time that the resources occupation rate of polycaryon processor be lower, reduce the output power consumption of polycaryon processor, improve the resource utilization of polycaryon processor, can also make the resources occupation rate of processor group maintain suitable level simultaneously, reduce the probability that processor group breaks down, improve the job stability of polycaryon processor, and then strengthen the stability of communication system.
In addition, the kernel dispatching method that the embodiment of the present invention two provides can be after execution step 207, the resources occupation rate of the second processor group will change, chain of command core circulation execution step 201, final realization maintains the resources occupation rate of processor group in polycaryon processor higher than first threshold, lower than the level of Second Threshold, the embodiment of the present invention not only can improve the resource utilization of polycaryon processor, reduce the energy consumption of communication system, can also improve the job stability of polycaryon processor, strengthen the stability of communication system.
Referring to Fig. 3, Fig. 3 is the process flow diagram of the kernel dispatching method that provides of the embodiment of the present invention three.
The kernel dispatching method that the embodiment of the present invention three provides comprises:
301, obtain the resources occupation rate of each processor group in polycaryon processor.
Concrete, chain of command endorses to obtain the resources occupation rate of each processor group in polycaryon processor.
302, minimum the 3rd processor group and the highest four-processor group of resources occupation rate of resources occupation rate in selection processor group, the 3rd resources occupation rate of processor group and the resources occupation rate of four-processor group are respectively E3 and E4.。
Concrete, chain of command endorses from processor group, to select processor group and the highest processor group of resources occupation rate that resources occupation rate is minimum, in embodiments of the present invention, what resources occupation rate was minimum is the 3rd processor group, the resources occupation rate of the 3rd processor group is E3, the processor group that resources occupation rate is the highest is four-processor group, and the resources occupation rate of four-processor group is E4.
303, be less than or equal to first threshold at E3, when E4 is more than or equal to Second Threshold, obtain first threshold and deduct the poor D1 of E3 and E4 and deduct the poor D2 of Second Threshold, Second Threshold is greater than first threshold.
Concrete, chain of command is endorsed with in the time being less than or equal to first threshold and E4 at E3 and being more than or equal to Second Threshold, the resources occupation rate of the 3rd processor group is low, when the resources occupation rate of four-processor group is higher, obtain first threshold and deduct the poor D1 of E3 and E4 and deduct the poor D2 of Second Threshold.Wherein, Second Threshold is greater than first threshold.In embodiments of the present invention, the first result and the second result are respectively D1 and D2.
In embodiments of the present invention, chain of command core can also all be greater than first threshold at E3 and E4, while being less than Second Threshold, returns to execution step 301.
304,, in the time that D1 is more than or equal to D2, the part forwarding face core of controlling in the 3rd processor group discharges shared resource.
Concrete, chain of command is endorsed with in the time that D1 is more than or equal to D2, and the part of controlling the 3rd processor group forwards face core, and some forwarding face core or multiple forwarding face core discharge shared resource, and reserve part forwards face core and continues to carry out corresponding business.Releasing resource is the first forwarding face core in embodiments of the present invention, for example control the first forwarding face core and discharge the resources such as shared internal memory, lock, serial ports, network interface, the first business that forwards the processing of face core is assigned to other forwarding face core in the 3rd processor group and carries out.Wherein, specifically the quantity of the forwarding face core of releasing resource can be determined according to the resources occupation rate of processor group, and chain of command is endorsed to release order to control the first forwarding face core to the first forwarding face core transmission resource and discharged shared resource.
305, after the face of forwarding core releasing resource completes, from the 3rd processor group, delete the forwarding face core of releasing resource.
Concrete, chain of command core, after the first forwarding face core releasing resource completes, is deleted the forwarding face core of releasing resource from the 3rd processor group, is the first forwarding face core in embodiments of the present invention, no longer to the first forwarding face core distribution services.Wherein, deleting the first forwarding face endorses with by configuring the first forwarding face core for going enabled state to realize.
306, the forwarding face caryogamy of releasing resource is set to holding state.
Concrete, chain of command core, by the forwarding face core of releasing resource, is that the first forwarding face caryogamy is set to holding state in embodiments of the present invention, and first forwards face core will be in low power consumpting state under holding state.For example chain of command core sends dependent instruction to the first forwarding face core and the clock of the first forwarding face core is turned off or the first forwarding face caryogamy is set to low-power consumption mode.
The embodiment of the present invention can further include:
307, in the time that D1 is less than D2, forwarding face core idle in polycaryon processor is loaded in four-processor group, the forwarding face core of start-up loading is carried out business processing.
Concrete, chain of command core is loaded into forwarding face core idle in polycaryon processor in the four-processor group that resources occupation rate is the highest, for example, to the idle resource such as forwarding face core storage allocation, lock.Wherein, chain of command core load the quantity of idle forwarding face core can be for one or more.Chain of command core starts the second forwarding face core and carries out the business processing in four-processor group.For example the second forwarding face core is carried out to hardware initialization, then receive corresponding service message, and service message is processed.
In embodiments of the present invention, polycaryon processor can be set to idle core by some forwarding face core wherein after powering on, so that chain of command core selects idle core to be loaded in the processor group that resources occupation rate is higher.
In embodiments of the present invention, chain of command is endorsed with the resources occupation rate of the 3rd processor group relatively and is departed from the value of first threshold and the resources occupation rate of four-processor group departs from the size of the value of Second Threshold, and then selects the larger processor group of departure degree to process accordingly.Compared with prior art, the embodiment of the present invention not only can be in the time that the resources occupation rate of polycaryon processor be lower, reduce the output power consumption of polycaryon processor, improve the resource utilization of polycaryon processor, can also make the resources occupation rate of processor group maintain suitable level simultaneously, reduce the probability that processor group breaks down, improve the job stability of polycaryon processor, and then strengthen the stability of communication system.
The embodiment of the present invention is after execution step 306 or 307, the resources occupation rate of the 3rd processor group or four-processor group will change, chain of command core circulation execution step 301, final realization maintains the resources occupation rate of processor group higher than first threshold, lower than the level of Second Threshold, improve the resource utilization of polycaryon processor, reduce the energy consumption of communication system.
Referring to Fig. 4, Fig. 4 is the method flow diagram that obtains the resources occupation rate of processor group in the embodiment of the present invention.
In the kernel dispatching method providing at the embodiment of the present invention one, embodiment bis-and embodiment tri-, the resources occupation rate that chain of command core obtains processor group in polycaryon processor can comprise:
401, obtain the message amount that in polycaryon processor, processor group is processed in preset time.
Concrete, chain of command core obtains the message data that in polycaryon processor, processor group is processed in preset time.Preset time can be for example 1 second:
402, obtain the resources occupation rate of processor group according to message amount.
Concrete, the message amount that chain of command core is processed when full load operation in preset time according to the message amount of processor group processing and processor group is obtained the resources occupation rate of this processor group, and then obtains the resources occupation rate of each processor group in polycaryon processor.
In embodiments of the present invention, for a certain definite communication system, its operation flow and processing complexity are also determined substantially, message processing capability based on single forwarding face core under this condition is also determined substantially, such as single forwarding face core is for worldwide interoperability for microwave access (Wimax, WorldwideInteroperability for Microwave Access) processing speed of system is 100000 message (100Kpps per second, 100K packets percent second), therefore, chain of command is endorsed the resources occupation rate to obtain this processor group according to the message amount of processor group processing, and then to adjusting the resources occupation rate of processor group, the embodiment of the present invention can conveniently be obtained the resources occupation rate of processor group.
The kernel dispatching the method above embodiment of the present invention being provided is described in detail, forwarding face core and polycaryon processor that the embodiment of the present invention is corresponding with above-mentioned kernel dispatching method.
Referring to Fig. 5, Fig. 5 is the structural representation of chain of command core in the polycaryon processor that provides of the embodiment of the present invention four.
The polycaryon processor that the embodiment of the present invention four provides comprises chain of command core and processor group, and processor group comprises forwarding face core, and chain of command core specifically comprises:
The first acquiring unit 501, for obtaining the resources occupation rate of the each processor group of polycaryon processor;
The first selected cell 502, in selection processor group, resources occupation rate is less than or equal to the first processor group of first threshold;
Resource release control unit 503, discharges shared resource for the part forwarding face core of controlling first processor group;
Kernel delete cells 504 after completing at the face of forwarding core releasing resource, is deleted the forwarding face core of releasing resource from first processor group;
Kernel setup unit 505, for being set to holding state by the forwarding face caryogamy of releasing resource.
Chain of command in the polycaryon processor that the embodiment of the present invention four provides is endorsed so that be used in kernel dispatching method first embodiment of aforementioned correspondence, and concrete scheduling process is referring to preceding method embodiment.In embodiments of the present invention, chain of command core selects resources occupation rate in polycaryon processor to be less than or equal to first threshold first processor group, the part in first processor group is forwarded to face caryogamy and be set to holding state.Compared with prior art, when the embodiment of the present invention can the resources occupation rate of processor group be lower in polycaryon processor, reduce the output power consumption of polycaryon processor, improve the resource utilization of polycaryon processor, and then reduce the energy consumption of communication system.
Wherein, the chain of command core in the polycaryon processor that the embodiment of the present invention four provides can also comprise:
The second selected cell 506, for the forwarding face caryogamy of releasing resource being set to after holding state in kernel setup unit 505, in selection processor group, resources occupation rate is more than or equal to the second processor group of Second Threshold, and Second Threshold is greater than first threshold;
Kernel loads unit 507, for being loaded into forwarding face core idle polycaryon processor in the second processor group;
Kernel start unit 508, the forwarding face core execution business processing loading for starting the second processor group.
The polycaryon processor that the embodiment of the present invention four provides can select resources occupation rate to be more than or equal to the second processor group of Second Threshold, and the second forwarding face core that loading is left unused in the second processor group, start second and forward face core execution business processing, thereby alleviate the processing load of the second processor group.The embodiment of the present invention can make the resources occupation rate of processor group maintain suitable level, thereby reduces the probability that processor group breaks down, and improves the job stability of polycaryon processor, and then strengthens the stability of communication system.
Referring to Fig. 6, Fig. 6 is the structural representation of chain of command core in the polycaryon processor that provides of the embodiment of the present invention five.
The polycaryon processor that the embodiment of the present invention five provides comprises chain of command core and processor group, and processor group comprises forwarding face core, and chain of command core specifically comprises:
Second acquisition unit 601, for obtaining the resources occupation rate of the each processor group of polycaryon processor;
The 3rd selected cell 602, minimum the 3rd processor group and the highest four-processor group of resources occupation rate of resources occupation rate in selection processor group;
The 3rd acquiring unit 603, for being less than or equal to first threshold in the resources occupation rate of the 3rd processor group, when the resources occupation rate of four-processor group is more than or equal to Second Threshold, obtain the first result that resources occupation rate that first threshold deducts the 3rd processor group obtains and the resources occupation rate of four-processor group and deduct the second result that Second Threshold obtains, Second Threshold is greater than first threshold;
The first trigger element 604, in the time that the first result is more than or equal to the second result, triggers the forwarding face core of controlling in the 3rd processor group resource release control unit 605 and discharges shared resource;
Resource release control unit 605, discharges shared resource for the part forwarding face core of controlling the 3rd processor group;
Kernel delete cells 606 after completing at the face of forwarding core releasing resource, is deleted the forwarding face core of releasing resource from the 3rd processor group;
Kernel setup unit 607, for being set to holding state by the forwarding face caryogamy of releasing resource.
Chain of command core in the polycaryon processor that the embodiment of the present invention five provides may further include:
The 3rd trigger element 608, in the time that the first result is less than the second result, triggers kernel loads unit 609 and loads forwarding face core;
Kernel loads unit 609, for being loaded into forwarding face core idle polycaryon processor in four-processor group;
Kernel start unit 610, for start four-processor group loading forwarding face core carry out business processing.
In embodiments of the present invention, chain of command is endorsed with the resources occupation rate of the 3rd processor group relatively and is departed from the value of first threshold and the resources occupation rate of four-processor group departs from the size of the value of Second Threshold, and then selects the larger processor group of departure degree to process accordingly.Compared with prior art, the embodiment of the present invention not only can be in the time that the resources occupation rate of polycaryon processor be lower, reduce the output power consumption of polycaryon processor, improve the resource utilization of polycaryon processor, can also reduce the probability that processor group breaks down simultaneously, improve the job stability of polycaryon processor, and then strengthen the stability of communication system.
The kernel dispatching method and the polycaryon processor that above the embodiment of the present invention are provided are described in detail, and the embodiment of the present invention also provides a kind of kernel backup method and polycaryon processor, is described in detail respectively below.
Referring to Fig. 7, Fig. 7 is the process flow diagram of the kernel backup method that provides of the embodiment of the present invention seven.
The kernel backup method that the embodiment of the present invention five provides comprises:
701, chain of command core obtains the duty of forwarding face core.
Concrete, the chain of command in polycaryon processor is endorsed the duty that forwards face core with Real-time Obtaining.Wherein, the duty of forwarding face core comprises normal condition, malfunction etc.
702,, when the first forwarding face core in the face of forwarding core breaks down, from default backup core, select the first backup core of holding state.
Concrete, when the first forwarding face core in the face of forwarding core breaks down, when the first duty that forwards face core is malfunction, chain of command core is selected the first backup core of holding state from default backup core.
Wherein, polycaryon processor can be set to back up core by some forwarding face core wherein after powering on, and the duty of backup core is holding state, so that in the time that the first forwarding face core breaks down, chain of command endorses to select the first backup core from backup core.
703, chain of command core activates the first backup core and carries out business processing.
Concrete, chain of command core activates the duty of the first backup core for duty, controls the first backup core and carries out business processing, for example, can allow the first backup core take over the business of the first forwarding face core.
704, chain of command core restarts the first forwarding face core.
Concrete, the first fault that forwards face core is generally software fault, after chain of command core restarts the first forwarding face core, can get rid of the fault of the first forwarding face core.
705, chain of command core forwards face caryogamy by first after restarting and is set to holding state.
Concrete, chain of command core, after the fault of the first forwarding face core is got rid of, forwards face caryogamy by first after restarting and is set to holding state, for example, send and hang up instruction to the first forwarding face core.First forwards face core will, in interrupt latency state, wait for waking up of chain of command core.
In embodiments of the present invention, chain of command core first when forwarding face core and breaking down, is selected the first backup core, and the first backup core is taken over to the business of the first forwarding face core in polycaryon processor, the first forwarding face core is set to spare core, thereby realizes for subsequent use of kernel in polycaryon processor.With backup nuclear phase ratio is not set in prior art, the embodiment of the present invention is by backup core is set, and the resource of backup core is dispatched, and can strengthen the job stability of polycaryon processor, and strengthen the job stability of communication system.
In embodiments of the present invention, normally the forwarding face core of work can constantly be counted in operational process, counts and can constantly increase, and chain of command is endorsed to obtain these and counted, and breaks down if count long-time constant explanation forwarding face core.Therefore, in the embodiment of the present invention, chain of command is endorsed with by the heartbeat that detects forwarding face core, forwards the number of times of face core middle change in count in preset time, obtains the duty of forwarding face core according to the heartbeat of the face of forwarding core.Wherein, the heartbeat that detects forwarding face core can adopt various ways, for example, mutually send message, or the face of forwarding core increase counting, and chain of command core reads counting etc.
Referring to Fig. 8, Fig. 8 is the structural representation of chain of command core in the polycaryon processor that provides of the embodiment of the present invention eight.
The polycaryon processor that the embodiment of the present invention six provides comprises forwarding face core and chain of command core, and chain of command core specifically comprises:
Backup core selected cell 802 while breaking down, is selected the first backup core of holding state for forwarding face core at first of the face of forwarding core from default backup core;
Backup core activates unit 803, carries out business processing for activating the first backup core;
Backup core start unit 804, for restarting the first forwarding face core;
Backup core dispensing unit 805, is set to holding state for first after restarting forwarded to face caryogamy.
Chain of command in the polycaryon processor that the embodiment of the present invention six provides is endorsed so that be used in the kernel backup method embodiment of aforementioned correspondence, chain of command core is by dispatching the resource of backup core, can strengthen the job stability of polycaryon processor, and strengthen the job stability of communication system.
It should be noted that, the content such as information interaction, implementation between the each unit in said apparatus, due to the inventive method embodiment based on same design, particular content can, referring to the narration in the inventive method embodiment, repeat no more herein.
One of ordinary skill in the art will appreciate that all or part of flow process realizing in above-described embodiment method, can carry out the hardware that instruction is relevant by computer program to complete, program can be stored in computer read/write memory medium, this program, in the time carrying out, can comprise as the flow process of the embodiment of above-mentioned each side method.Wherein, storage medium can be magnetic disc, CD, read-only store-memory body (Read-Only Memory, ROM) or random store-memory body (Random Access Memory, RAM) etc.
The polycaryon processor kernel dispatching method, kernel backup method and the polycaryon processor that above the embodiment of the present invention are provided are described in detail, for one of ordinary skill in the art, according to the thought of the embodiment of the present invention, all will change in specific embodiments and applications, this description should not be construed as limitation of the present invention.
Claims (4)
1. a kernel dispatching method for polycaryon processor, is characterized in that, comprising:
Obtain the message amount that in polycaryon processor, each processor group is processed in preset time;
Obtain the resources occupation rate of each processor group according to the message amount of processing in described preset time;
Select minimum the 3rd processor group and the highest four-processor group of resources occupation rate of resources occupation rate in described processor group;
Resources occupation rate in described the 3rd processor group is less than or equal to first threshold, when the resources occupation rate of described four-processor group is more than or equal to Second Threshold, obtain the first result that resources occupation rate that described first threshold deducts described the 3rd processor group obtains and the resources occupation rate of described four-processor group and deduct the second result that Second Threshold obtains, described Second Threshold is greater than described first threshold;
In the time that described the first result is more than or equal to the second result, the part forwarding face core of controlling in described the 3rd processor group discharges shared resource;
After described forwarding face core releasing resource completes, from described the 3rd processor group, delete the forwarding face core of releasing resource;
The forwarding face caryogamy of described releasing resource is set to holding state.
2. method according to claim 1, is characterized in that, also comprises:
In the time that described the first result is less than the second result, forwarding face core idle in described polycaryon processor is loaded in described four-processor group;
The forwarding face core that starts the loading in described four-processor group is carried out business processing.
3. a polycaryon processor, comprises at least two processor groups, and described processor group comprises forwarding face core, it is characterized in that, also comprises chain of command core, and described chain of command core comprises:
Second acquisition unit, for obtaining the resources occupation rate of the each processor group of polycaryon processor;
Described second acquisition unit is specifically for obtaining the message amount that in polycaryon processor, each processor group is processed in preset time; Obtain the resources occupation rate of each processor group according to the message amount of processing in described preset time;
The 3rd selected cell, for selecting the 3rd processor group and the highest four-processor group of resources occupation rate that described processor group resources occupation rate is minimum;
The 3rd acquiring unit, for being less than or equal to first threshold in the resources occupation rate of described the 3rd processor group, when the resources occupation rate of described four-processor group is more than or equal to Second Threshold, obtain the first result that resources occupation rate that described first threshold deducts described the 3rd processor group obtains and the resources occupation rate of described four-processor group and deduct the second result that Second Threshold obtains, described Second Threshold is greater than described first threshold;
The first trigger element, in the time that described the first result is more than or equal to the second result, triggers described in described resource release control unit controls the part in the 3rd processor group and forwards face core and discharge shared resource;
Resource release control unit, discharges shared resource for the part forwarding face core of controlling described the 3rd processor group;
Kernel delete cells after completing at described forwarding face core releasing resource, is deleted the forwarding face core of releasing resource from described the 3rd processor group;
Kernel setup unit, for being set to holding state by the forwarding face caryogamy of described releasing resource.
4. polycaryon processor according to claim 3, is characterized in that, described chain of command core also comprises:
The 3rd trigger element, in the time that described the first result is less than the second result, triggers kernel loads unit and loads forwarding face core;
Kernel loads unit, for being loaded into forwarding face core idle described polycaryon processor in described four-processor group;
Kernel start unit, for start described four-processor group loading forwarding face core carry out business processing.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201010115862.0A CN101799772B (en) | 2010-02-26 | 2010-02-26 | Kernel dispatching method, kernel backup method and multi-core processor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201010115862.0A CN101799772B (en) | 2010-02-26 | 2010-02-26 | Kernel dispatching method, kernel backup method and multi-core processor |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101799772A CN101799772A (en) | 2010-08-11 |
| CN101799772B true CN101799772B (en) | 2014-06-11 |
Family
ID=42595456
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201010115862.0A Active CN101799772B (en) | 2010-02-26 | 2010-02-26 | Kernel dispatching method, kernel backup method and multi-core processor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101799772B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103077081B (en) * | 2012-12-31 | 2017-02-22 | 华为技术有限公司 | Method and device of adjusting resources |
| CN104077266B (en) * | 2013-03-27 | 2017-07-07 | 华为技术有限公司 | Many kernel operating system implementation methods and realize apparatus and system |
| CN107132903B (en) * | 2016-02-29 | 2020-11-13 | 中兴通讯股份有限公司 | Energy-saving management implementation method, device and network equipment |
| CN106020844A (en) * | 2016-05-31 | 2016-10-12 | 广东欧珀移动通信有限公司 | Method and device for determining kernel running configurations in processor |
| CN108664285A (en) * | 2018-05-10 | 2018-10-16 | Oppo广东移动通信有限公司 | Application program preloads method, apparatus, storage medium and mobile terminal |
| CN109522113B (en) * | 2018-09-28 | 2020-12-18 | 迈普通信技术股份有限公司 | Memory management method and device |
| CN109840135B (en) * | 2019-01-30 | 2022-02-18 | 郑州云海信息技术有限公司 | Load balancing method and device and electronic equipment |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1604050A (en) * | 2003-09-30 | 2005-04-06 | 国际商业机器公司 | System and method for autonomic memory leak detection and remediation |
| CN101354664A (en) * | 2008-08-19 | 2009-01-28 | 中兴通讯股份有限公司 | Method and apparatus for interrupting load equilibrium of multi-core processor |
| CN101436098A (en) * | 2008-12-24 | 2009-05-20 | 华为技术有限公司 | Method and apparatus for reducing power consumption of multiple-core symmetrical multiprocessing system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101146089B (en) * | 2007-08-22 | 2011-12-28 | 杭州华三通信技术有限公司 | Method for configuring core resources in multi-core system, multi-core system and management core |
-
2010
- 2010-02-26 CN CN201010115862.0A patent/CN101799772B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1604050A (en) * | 2003-09-30 | 2005-04-06 | 国际商业机器公司 | System and method for autonomic memory leak detection and remediation |
| CN101354664A (en) * | 2008-08-19 | 2009-01-28 | 中兴通讯股份有限公司 | Method and apparatus for interrupting load equilibrium of multi-core processor |
| CN101436098A (en) * | 2008-12-24 | 2009-05-20 | 华为技术有限公司 | Method and apparatus for reducing power consumption of multiple-core symmetrical multiprocessing system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101799772A (en) | 2010-08-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101799772B (en) | Kernel dispatching method, kernel backup method and multi-core processor | |
| US10552222B2 (en) | Task scheduling method and apparatus on heterogeneous multi-core reconfigurable computing platform | |
| EP3073374B1 (en) | Thread creation method, service request processing method and related device | |
| TWI569202B (en) | Apparatus and method for adjusting processor power usage based on network load | |
| CN104238712B (en) | Distributed power conveying | |
| CN101470518B (en) | Method and device for service independent of operating system | |
| CN103124957B (en) | Method and apparatus for the Dynamic Resource Allocation for Multimedia of processing unit | |
| CN102043675B (en) | Thread pool management method based on task quantity of task processing request | |
| CN102955549B (en) | The method for managing power supply of a kind of multi-core CPU, system and CPU | |
| US20130167152A1 (en) | Multi-core-based computing apparatus having hierarchical scheduler and hierarchical scheduling method | |
| CN104050137B (en) | The method and device that kernel is run in a kind of OS | |
| CN107579926A (en) | The QoS methods to set up of Ceph cloud storage systems based on token bucket algorithm | |
| CN101403982A (en) | Task distribution method, system and equipment for multi-core processor | |
| CN103037109B (en) | Multicore equipment energy consumption management method and device | |
| CN102662740B (en) | Asymmetric multi-core system and realization method thereof | |
| CN103246550A (en) | Multitask dispatching method and system based on capacity | |
| CN108345502A (en) | Resource regulating method, device, terminal device based on DPDK and readable storage medium storing program for executing | |
| CN107656813A (en) | The method, apparatus and terminal of a kind of load dispatch | |
| CN103634379A (en) | Management method for distributed storage space and distributed storage system | |
| CN108205475A (en) | EMS memory management process, terminal device, computer installation and readable storage medium storing program for executing | |
| CN110795323A (en) | Load statistical method, device, storage medium and electronic equipment | |
| CN110825212B (en) | Energy-saving scheduling method and device and computer storage medium | |
| CN110308991B (en) | Data center energy-saving optimization method and system based on random tasks | |
| CN104182280B (en) | Low-energy RM real-time task scheduling method for hybrid main memory embedded system | |
| CN102609306B (en) | Method for processing video processing tasks by aid of multi-core processing chip and system using method |
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 |