CN1852156A - Method for updating programe edition of processing system without operation interruption - Google Patents
Method for updating programe edition of processing system without operation interruption Download PDFInfo
- Publication number
- CN1852156A CN1852156A CN 200510117385 CN200510117385A CN1852156A CN 1852156 A CN1852156 A CN 1852156A CN 200510117385 CN200510117385 CN 200510117385 CN 200510117385 A CN200510117385 A CN 200510117385A CN 1852156 A CN1852156 A CN 1852156A
- Authority
- CN
- China
- Prior art keywords
- unit
- processing unit
- processing
- streamline
- work
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Abstract
The method is suitable to a processing system of using more than one processing unit with independent instruction space. In procedure of processing service by the processing system, the method carries out program upgrade for processing units needed to upgrade one by one. When upgrading program for single processing unit, the method appoints alternative processing unit, and loads upgrade program for processing unit to be upgraded to instruction space of the alternative processing unit. Using the alternative processing unit replaces operation of the processing unit to be upgraded. Features are: upgrading scale is not limited by instruction space of processing system.
Description
Technical field
The present invention relates to the program version of treatment system is carried out technology of upgrading, relate in particular to a kind of method of under the situation of non-interrupting service, the program version of treatment system being upgraded.
Background technology
Reliabilty and availability is a basic demand to high-end communication processing system.The reliabilty and availability of general communication processing system all uses interior (in the 1 year) time of service interruption of certain hour to weigh.Communication processing system generally is made up of program and hardware, and hardware components wherein can be very durable after the design maturation; But for program part, because size of code huge (size of code such as general high-end router has millions of row), so some shortcomings (Bug) are inevitably arranged, in addition because the development of technology is with rapid changepl. never-ending changes and improvements, new traffic performance is also in continuous appearance, so just inevitable to the upgrading of program version.
The program of treatment system is divided into control plane and datum plane.Wherein the brain of the similar communication processing system of control plane is used to handle the protocol package between the communication processing system, simultaneously system is managed, disposes, monitors; Datum plane also claims service plane, is the plane that communication processing system is managed business, and the packet during such as online is all walked datum plane and transmitted.
If the program version of upgrading communication processing system for control plane, generally all has backup, can realize program upgrade by masterslave switchover, and can accomplish not influence business; But for the datum plane service plane, because all business are all managed herein, the ROMPaq version need restart, and therefore must cause professional interruption.
At present, development along with internet (Internet), the network user rapidly increases, and various types of application co-exist on the physical network, and the disposal ability of network node, service quality (QoS) ensure and the network bandwidth becomes the major obstacle that Internet continues development gradually.But along with the development of optical transport technology, transmission bandwidth no longer becomes the bottleneck of network, and the disposal ability of network switching node seems particularly important.In order to satisfy the new variation of IP network, Gigabit switching router (GSR) and tril switch router (TSR) are arisen at the historic moment, and these high-end routers adopt big capacity switching fabric and hardware high speed retransmission technique, have improved message transferring speed greatly; Network processing unit (NP, Network Processor) technology as these devices exchange cores has also obtained swift and violent development and application.
Network processing unit is a kind of programming device, is applied to the various particular tasks of the communications field, for example: the converging of bag processing, protocal analysis, route querying, sound/data, fire compartment wall, QOS or the like.The present application of network processing unit mainly concentrates on the middle and high end equipment of data communication network, transmits and compares with former program forwarding, FPGA realization forwarding, ASIC realization, and it has following characteristics or advantage:
1, the performance height.A lot of algorithms of network processing unit are all realized with hardware, its inside is generally all integrated several in addition tens transmit micro engine (ME) and hardware co-processor, hardware accelerator, under the prerequisite that realizes complicated congestion management, queue scheduling, traffic classification and QOS function, can also reach very high and search, forwarding performance, realize so-called " the hard forwarding ".
2, can carry out function expansion flexibly.Because network processing unit can be programmed,, can realize by software programming very easily that the function of system can be added deletion easily by program module in case have new technology or demand to occur.So for special user's request, can carry out customized development, promptly can delete the product that exploitation can meet the needs of different users by module at short notice.And under the former situation, needing to revise pin function with the hardware logic realization, debugging is upgraded again, and other hardware designs also will be changed under most situation, for system reliability is brought very big hidden danger.Under the situation with the ASIC realization, can't add, can only redesign, more new chip new function.
3, the reliability height.Because most of function of network processing unit all uses one or two chips to realize, has all passed through strict test and various anti-interference and destructive testing before chip is changed the line of production, thereby the reliability of the treatment system that adopts NP is improved greatly.So NP is particularly suitable for developing the carrier class data communication products.
4, management, exploitation are conveniently.NP provides interface or the built-in CPU management with upper strata central processing unit (CPU) standard, can realize high speed communication with other CPU.NP generally provides a large amount of hardware counters, can realize various MIB statistical functions easily, for webmaster provides support.NP generally provides compiling system and program sample, and the NP of main flow provides the procedure simulation development platform at present, can carry out off-line exploitation and checking, even can clock cycle (Cycle) level be arrived in the efficient emulation of program with emulation platform; When on-line debugging, there is single step to follow the tracks of, be provided with means such as breakpoint, can makes the equipment development merchant develop the program that is fit to product needed within a short period of time.
5, can realize flexible configurations.NP provides configuration feature flexibly as a device, multi-form combination that can be by NP or and the combination of other CPU, realize the flexible configuration of system, satisfy the demand of distinct device, make things convenient for system design, accelerated the development progress of equipment.
In view of above-mentioned advantage, network processing unit is becoming the mainstream technology that datum plane is transmitted.Claim at present to adopt network processing unit to carry out router that datum plane handles and be " the 5th generation router ".
Yet on datum plane,, having become and having realized that high-end router and other the treatment system of utilizing network processing unit do not interrupt the key of upgrading for how realizing on the network processing unit ROMPaq version under the situation of non-interrupting service.
At present, in adopting the treatment system of NP, the method that is implemented in the program version of upgrading processing system under the situation of non-interrupting service mainly contains hot patch (Hot Patch) technology.The major programme of this technology is meant when little defective (Bug) appears in program, can realize repairing by covering or increase redirect statement to binary file.Described patch is the program code segments of being upgraded.For example, in following binary system execute file, the program between n～n+m has Bug.
.................................
n xxxxxxxx
n+1 xxxxxxxx
......
n+m xxxxxxxx
.....................
The mode of beating hot patch has two kinds of methods usually:
First kind, in the instruction space (being internal memory) of network processing unit, open up one and be specifically designed to the zone of depositing patch, be assumed to from address a1 to address a2 herein, in a1～a1+x, deposit the binary code of correct program then, deposit the code of go to n+m+1 at a1+x+1, the instruction of revising the n place at last is go to a1, to jump to a1 in program running during to n like this and move correct processing code, therefore can under the situation of non-interrupting service, carry out patch with the ROMPaq version.
If the shared space of second kind of patch code is smaller or equal to the space of failure code, supposing needs the instruction of x bar, and x＜=m can directly cover the binary file that begins from n.If x also needs in the end one to deposit go to a1 less than m, so just can non-interrupting service and revise Bug, reach the upgrading purpose.
But there is following shortcoming in existing this hot patch technology:
1, because the instruction space of network processing unit is very limited,, can only make amendment to limited instruction so its modification is limited in scope.For datum plane, when the patch code number was counted greater than original instruction code, the instruction space of network processing unit was very limited, so the vacant instruction space that is used for hot patch can not be too big, so some big a little modifications can't be implemented to load.
2, the hot patch modification can relate to the modification data structure, if but revise and to relate to too many data structure, then can relate to some intermediatenesses, cause handling failure.
3, for increase new function, when having beaten in the microcode that patch is local to be needed to revise owing to beaten too much patch, wasted a large amount of instruction spaces, therefore can't use hot patch or realize very difficulty.
Summary of the invention
In view of this, the method that provides a kind of non-interrupting service that the program version of treatment system is upgraded is provided main purpose of the present invention, thereby when realizing that non-interrupting service is upgraded to the program version of treatment system, the scale of upgrading is not subject to processing the restriction of system instruction space.
In order to realize the foregoing invention purpose, main technical schemes of the present invention is:
The method that a kind of non-interrupting service is upgraded to the program version of treatment system, be applicable to and utilize more than one processing unit to carry out the treatment system of Business Processing with independent instruction space, carry out in the process of Business Processing in treatment system, the processing unit to the need upgrading carries out program upgrade one by one; When single processing unit is carried out program upgrade: specify the alternate process unit, the ROMPaq of the processing unit of need upgrading is loaded in the instruction space of alternate process unit, the processing unit work that replaces this to upgrade with the alternate process unit.
Described appointment alternate process unit is: reserve idle processing unit in treatment system, the idle processing unit of reserving is appointed as the alternate process unit.
Described appointment alternate process unit is: will can quit work in the process systems programme streamline temporarily and do not influence the processing unit that current business handles and be appointed as the alternate process unit.
Described appointment alternate process unit is: the arbitrary processing unit that carries out in the process systems programme streamline in the above processing unit of parallel processing is appointed as the alternate process unit.
The detailed process that described processing unit of one by one need being upgraded carries out program upgrade is:
First processing unit that need upgrade is set to work as pretreatment unit in the streamline of A, process systems programme;
B, the ROMPaq that will work as pretreatment unit are loaded into the instruction space of alternate process unit;
C, the alternate process unit is added streamline, and will work as pretreatment unit and from streamline, extract;
Whether processing finishes when the task of pretreatment unit for D, judgement, is then to carry out next step; Otherwise repeat this step;
E, stop the task when pretreatment unit, will work as pretreatment unit and be appointed as the alternate process unit, the processing unit that the next one in the streamline need be upgraded is set to work as pretreatment unit;
F, repeat above-mentioned steps B to step e, all processing units that need upgrade are all upgraded and are finished in streamline.
The detailed process that described processing unit of one by one need being upgraded carries out program upgrade is:
First processing unit that need upgrade is set to work as pretreatment unit in the streamline of a, process systems programme;
B, the ROMPaq that will work as pretreatment unit are loaded into the instruction space of alternate process unit;
C, the alternate process unit is added streamline, and will work as pretreatment unit and from streamline, extract;
Whether processing finishes when the task of pretreatment unit for d, judgement, is then to carry out next step; Otherwise repeat this step;
E, stop the task when pretreatment unit, the ROMPaq that will work as pretreatment unit is loaded into the instruction space when pretreatment unit;
F, will work as pretreatment unit and join streamline, the alternate process unit will be extractd from streamline;
G, judge the alternate process unit task whether processing finish, be then to carry out next step; Otherwise repeat this step;
H, stop the task of alternate process unit, the processing unit that the next one in the streamline need be upgraded is set to work as pretreatment unit;
I, repeat above-mentioned steps b to step h, all processing units that need upgrade are all upgraded and are finished in streamline.
Described processing unit is: micro engine or CPU.
Described treatment system is a network processing unit.
Described method does not influence the processing unit of Business Processing to quit work temporarily when upgrading, and does not need the replacement work of alternate process unit, and the instruction space that directly ROMPaq is loaded into this processing unit is upgraded.
With respect to existing hot patch technology, because method of the present invention utilizes the alternate process unit one by one the processing unit that needs upgrading in the treatment system to be carried out program upgrade respectively, therefore Business Processing that can interrupt processing system in escalation process; Because among the present invention is to replace former processing unit to carry out work with the ROMPaq post-processing unit, needn't occupying volume outer instruction space patch installing, therefore the relative prior art of scale of utilizing method of the present invention to upgrade is bigger, be no more than the instruction space of treatment system as long as the integral body of upgrading back program takes up room, can do bigger modification original program; Can save simultaneously the instruction space of treatment system in a large number; In addition, because upgrading mode of the present invention is for to replace former processing cell operation with the processing unit after the ROMPaq, the program of processing unit is whole to be replaced, and does not relate to the modification of data structure, the problem that causes owing to the data structure modification therefore can not occur.
Description of drawings
Fig. 1 is the system block diagram of 2800 network processing units;
Fig. 2 is the distribution condition schematic diagram of ME in the network processing unit;
Fig. 3 is the flow chart of first embodiment of the invention;
The schematic diagram of Fig. 4 for a ME is upgraded;
Fig. 5 is the flow chart of second embodiment of the invention;
Fig. 6 does not reserve ME and specifies the schematic diagram that substitutes ME for third embodiment of the invention is described.
Embodiment
Below by specific embodiments and the drawings the present invention is described in further details.
Core technology scheme of the present invention is: the method that a kind of non-interrupting service is upgraded to the program version of treatment system, be applicable to and utilize more than one processing unit to carry out the treatment system of Business Processing with independent instruction space, carry out in the process of Business Processing in treatment system, the processing unit to the need upgrading carries out program upgrade one by one; When single processing unit is carried out program upgrade: specify the alternate process unit, the ROMPaq of the processing unit of need upgrading is loaded in the instruction space of alternate process unit, the processing unit work that replaces this to upgrade with the alternate process unit.
The treatment system that the present invention was suitable for is to utilize more than one processing unit with independent instruction space to carry out the treatment system of Business Processing, the network processing unit that adopts in for example high-end communication equipment, perhaps other the treatment system of utilizing the general central processing unit (CPU) that has the independent instruction space more than one to handle.Described processing unit is micro engine (ME) or CPU (CPU).
IXP2800 network processing unit (hereinafter to be referred as 2800 network processing units) with Intel Company in following examples is an example, and this method is described.
Fig. 1 is the system block diagram of 2800 network processing units.Referring to Fig. 1, comprised 16 micro engines (ME) and an embedded type CPU in 2800 network processing units, this CPU is named as Xscale.Each micro engine is a lean instruction set computer (RISC) CPU, and each micro engine comprises 8 threads, each thread all have oneself physically independently one the cover memory and program counter (PC) pointer and relevant storehouse.Also comprise the coprocessor unit that some and the present invention have nothing to do in addition and receive parts such as special hardware such as transmissions grade.ME is used to transmit data, Xscale be used for NP management and and the CPU on tension management plane communicate etc.The instruction space of each ME is independently, i.e. the instruction of each ME can be different.Programming language on the ME is a kind of special assembler language, is " microcode ".
Fig. 2 is the distribution condition schematic diagram of ME in the network processing unit.Referring to Fig. 2, in 16 ME of this network processing unit, 2～4 reception, transmissions that are used for packet, remaining ME is used for Business Processing.
Fig. 3 is the flow chart of first embodiment of the invention.Among this first embodiment, load in order to realize non-interrupting service, reserving a ME in treatment system is idle ME, assists other ME ROMPaq when being specifically designed to loading procedure.When the ROMPaq version,, realize ME instruction purpose of upgrading one by one by replacing the ME that is working with this idle ME.
Referring to Fig. 3, suppose that ME0 is the ME that reserves, this flow process comprises:
First ME1 that need upgrade is set to current ME in step 301, the program flow waterline, and ME0 is appointed as initial alternative ME, is used for substituting the ME that is working, and stops to substitute the work of ME;
Whether all ME that need upgrade have passed through above-mentioned upgrading processing in step 306, the judgment processing system, if, process ends then; Otherwise return step 302.
In this first embodiment, when next ME upgrades, can use this ME that frees out ME assistance upgrading as an alternative.
ME is designed to the mode of streamline in flow scheme design, carry out between the ME of each pipeline step and come transmission information by round-robin queue, and ME reads task from round-robin queue, and the back that disposes sends to next ME by next round-robin queue.The schematic diagram of Fig. 4 for a ME is upgraded.Referring to Fig. 4, suppose ME-a is upgraded herein, mainly comprise 4I, 4II, three parts of 4III.At first referring to 4I, the ROMPaq of ME-a is loaded into substitutes among the ME 41; Secondly referring to 4II, will substitute the streamline that ME 41 adds programs and substitute ME-a, and ME-a is extractd from streamline, promptly cut off its input data; At last after the task processing end referring to 4III, ME-a, then stop the work of ME-a, being about to its input and output all cuts off, this moment, ME-a was the ME of a free time, therefore can be used as an alternative ME replaces follow-up ME, ME-b for example, the profit ME-b that can upgrade that uses the same method, all upgrading up to all ME finishes.
Among first embodiment, alternative ME changes one by one, is the ME that the back free time of at every turn upgrading goes out to be appointed as alternative ME assist next ME to upgrade.In a second embodiment, the ME that can adopt an alternative ME to assist all to upgrade all the time upgrades.Fig. 5 is the flow chart of second embodiment of the invention.Referring to Fig. 5, suppose that ME0 is the ME that reserves, this flow process comprises:
First ME1 that need upgrade is set to current ME in step 501, the program flow waterline, specifies described ME0 for substituting ME, stops the work of ME0;
Whether all ME that need upgrade have passed through above-mentioned upgrading processing in step 509, the judgment processing system, if, process ends then; Otherwise return step 502.
Above-mentioned two embodiment's all is to utilize the idle ME that reserves in network processing unit to upgrade as initial alternative ME, the present invention also can adopt the method for not reserving ME to upgrade, this method is applicable to have in the described program flow waterline and can quits work temporarily and do not influence the ME that the network processing unit current business is handled, can with this ME as an alternative ME assist to finish program upgrade.Especially in the Treatment Design of network processing unit, certain link on the program flow waterline has more than one ME parallel processing, any ME among the ME of parallel processing can be appointed as alternative ME so and assist to finish program upgrade.
Fig. 6 is that the described ME that do not reserve of the 3rd embodiment specifies the schematic diagram that substitutes ME.Referring to Fig. 6, for the situation that an above ME parallel processing is arranged on the streamline link, for example referring to 6I, ME-a and ME-a1 parallel processing on a link, can therefrom choose the carrier that a ME is appointed as upgrading wantonly, promptly substitute ME, the ME that the need in the network processing unit are upgraded upgrades one by one.For example specifying ME-a1 herein is initial alternative ME, and its assignment procedure is referring to 6II and 6III:6II, extract ME-a1 from streamline; After the task processing of ME-a1 finishes, stop the work of this ME-a1, this ME-a1 is appointed as initial alternative ME, utilize and to upgrade by alternative ME, its concrete escalation process can be identical with flow process shown in Figure 3, also can be identical with flow process shown in Figure 5, just ME-a is handled as wherein ME0.And in this 3rd embodiment, owing to there is not idle ME, if ME-a1 also need upgrade, then the alternative ME of last appointment assists ME-a1 to carry out program upgrade, have two kinds of situations this moment: if utilize flow process shown in Figure 3 to upgrade, then load the ROMPaq of ME-a1 among the alternative ME of last appointment, and replace ME-a1 work; If utilize flow process shown in Figure 5 to upgrade, the alternative ME or the ME-a1 of then last appointment directly load himself ROMPaq, add streamline and carry out work.
Because in some Business Stream waterline of the processing of network processing unit, may comprise quits work does not temporarily influence the ME of Business Processing, when the program of these ME is upgraded, can not need substituting ME assists, but directly this ME is extractd from streamline, after the task processing of this ME finishes, stop the work of this ME, and join streamline again after the instruction space that ROMPaq is loaded into this ME upgraded; Influencing the ME of Business Processing for quit work temporarily then utilizes above-mentioned first, second and third the described method of embodiment to upgrade.
The present invention is when upgrading to ME one by one, also can adjust the position of ME in the program flow waterline, therefore the present invention not only is applicable to small-scale program upgrade optimization under the non-interrupting service situation, also is applicable to the extensive modification of program under the non-interrupting service situation.
The above; only for the preferable embodiment of the present invention, but protection scope of the present invention is not limited thereto, and anyly is familiar with the people of this technology in the disclosed technical scope of the present invention; the variation that can expect easily or replacement all should be encompassed within protection scope of the present invention.
Claims (10)
1, a kind of non-interrupting service method that the program version of treatment system is upgraded, be applicable to and utilize more than one processing unit to carry out the treatment system of Business Processing with independent instruction space, it is characterized in that, carry out in the process of Business Processing in treatment system, the processing unit to the need upgrading carries out program upgrade one by one; When single processing unit is carried out program upgrade: specify the alternate process unit, the ROMPaq of the processing unit of need upgrading is loaded in the instruction space of alternate process unit, the processing unit work that replaces this to upgrade with the alternate process unit.
2, method according to claim 1 is characterized in that, described appointment alternate process unit is: reserve idle processing unit in treatment system, the idle processing unit of reserving is appointed as the alternate process unit.
3, method according to claim 1 is characterized in that, described appointment alternate process unit is: will can quit work in the process systems programme streamline temporarily and do not influence the processing unit that current business handles and be appointed as the alternate process unit.
4, method according to claim 1 is characterized in that, described appointment alternate process unit is: the arbitrary processing unit that carries out in the process systems programme streamline in the above processing unit of parallel processing is appointed as the alternate process unit.
According to claim 1,2,3 or 4 described methods, it is characterized in that 5, the detailed process that described processing unit of one by one need being upgraded carries out program upgrade is:
First processing unit that need upgrade is set to work as pretreatment unit in the streamline of A, process systems programme;
B, the ROMPaq that will work as pretreatment unit are loaded into the instruction space of alternate process unit;
C, the alternate process unit is added streamline, and will work as pretreatment unit and from streamline, extract;
Whether processing finishes when the task of pretreatment unit for D, judgement, is then to carry out next step; Otherwise repeat this step;
E, stop the task when pretreatment unit, will work as pretreatment unit and be appointed as the alternate process unit, the processing unit that the next one in the streamline need be upgraded is set to work as pretreatment unit;
F, repeat above-mentioned steps B to step e, all processing units that need upgrade are all upgraded and are finished in streamline.
6, method according to claim 5 is characterized in that, described processing unit is: micro engine or CPU.
According to claim 1,2,3 or 4 described methods, it is characterized in that 7, the detailed process that described processing unit of one by one need being upgraded carries out program upgrade is:
First processing unit that need upgrade is set to work as pretreatment unit in the streamline of a, process systems programme;
B, the ROMPaq that will work as pretreatment unit are loaded into the instruction space of alternate process unit;
C, the alternate process unit is added streamline, and will work as pretreatment unit and from streamline, extract;
Whether processing finishes when the task of pretreatment unit for d, judgement, is then to carry out next step; Otherwise repeat this step;
E, stop the task when pretreatment unit, the ROMPaq that will work as pretreatment unit is loaded into the instruction space when pretreatment unit;
F, will work as pretreatment unit and join streamline, the alternate process unit will be extractd from streamline;
G, judge the alternate process unit task whether processing finish, be then to carry out next step; Otherwise repeat this step;
H, stop the task of alternate process unit, the processing unit that the next one in the streamline need be upgraded is set to work as pretreatment unit;
I, repeat above-mentioned steps b to step h, all processing units that need upgrade are all upgraded and are finished in streamline.
8, method according to claim 7 is characterized in that, described processing unit is: micro engine or CPU.
According to claim 1,2,3 or 4 described methods, it is characterized in that 9, described treatment system is a network processing unit.
10, method according to claim 1, it is characterized in that, described method does not influence the processing unit of Business Processing to quit work temporarily when upgrading, and does not need the replacement work of alternate process unit, and the instruction space that directly ROMPaq is loaded into this processing unit is upgraded.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2005101173850A CN100414888C (en) | 2005-11-03 | 2005-11-03 | Method for updating programe edition of processing system without operation interruption |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2005101173850A CN100414888C (en) | 2005-11-03 | 2005-11-03 | Method for updating programe edition of processing system without operation interruption |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1852156A true CN1852156A (en) | 2006-10-25 |
| CN100414888C CN100414888C (en) | 2008-08-27 |
Family
ID=37133626
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2005101173850A Active CN100414888C (en) | 2005-11-03 | 2005-11-03 | Method for updating programe edition of processing system without operation interruption |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100414888C (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101729428A (en) * | 2010-01-07 | 2010-06-09 | 中兴通讯股份有限公司 | Microcode switching method and microcode switching device |
| CN101222374B (en) * | 2008-01-30 | 2010-09-22 | 华为技术有限公司 | System, device and method for implementing service upgrade |
| CN102025607B (en) * | 2009-09-19 | 2013-04-17 | 华为技术有限公司 | Data processing method, network processor and network equipment |
| CN104486394A (en) * | 2014-12-10 | 2015-04-01 | 杭州华三通信技术有限公司 | Non-interrupting service software upgrading method and device |
| WO2015103876A1 (en) * | 2014-01-13 | 2015-07-16 | 中兴通讯股份有限公司 | Method and apparatus for processing version upgrading |
| CN106792790A (en) * | 2016-03-31 | 2017-05-31 | 新华三技术有限公司 | A kind of AP upgrade methods and device |
| WO2018001044A1 (en) * | 2016-06-27 | 2018-01-04 | 中兴通讯股份有限公司 | Method and apparatus for upgrading single-stage router to cluster router |
| CN110011822A (en) * | 2018-12-19 | 2019-07-12 | 北京乐我无限科技有限责任公司 | Upgrade method, device, management server and the system of Edge Server |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH07319683A (en) * | 1994-05-30 | 1995-12-08 | Nippon Telegr & Teleph Corp <Ntt> | In-operation program updating system |
| US6085333A (en) * | 1997-12-19 | 2000-07-04 | Lsi Logic Corporation | Method and apparatus for synchronization of code in redundant controllers in a swappable environment |
| JP3052919B2 (en) * | 1997-12-25 | 2000-06-19 | 日本電気株式会社 | I / O controller |
-
2005
- 2005-11-03 CN CNB2005101173850A patent/CN100414888C/en active Active
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101222374B (en) * | 2008-01-30 | 2010-09-22 | 华为技术有限公司 | System, device and method for implementing service upgrade |
| CN102025607B (en) * | 2009-09-19 | 2013-04-17 | 华为技术有限公司 | Data processing method, network processor and network equipment |
| CN101729428A (en) * | 2010-01-07 | 2010-06-09 | 中兴通讯股份有限公司 | Microcode switching method and microcode switching device |
| WO2015103876A1 (en) * | 2014-01-13 | 2015-07-16 | 中兴通讯股份有限公司 | Method and apparatus for processing version upgrading |
| CN104486394A (en) * | 2014-12-10 | 2015-04-01 | 杭州华三通信技术有限公司 | Non-interrupting service software upgrading method and device |
| CN104486394B (en) * | 2014-12-10 | 2018-01-12 | 新华三技术有限公司 | In-service Software Upgrade method and device |
| CN106792790A (en) * | 2016-03-31 | 2017-05-31 | 新华三技术有限公司 | A kind of AP upgrade methods and device |
| CN106792790B (en) * | 2016-03-31 | 2021-04-16 | 新华三技术有限公司 | AP upgrading method and device |
| WO2018001044A1 (en) * | 2016-06-27 | 2018-01-04 | 中兴通讯股份有限公司 | Method and apparatus for upgrading single-stage router to cluster router |
| CN107547222A (en) * | 2016-06-27 | 2018-01-05 | 中兴通讯股份有限公司 | Upgrade method and device of the single-stage router to cluster routers |
| CN110011822A (en) * | 2018-12-19 | 2019-07-12 | 北京乐我无限科技有限责任公司 | Upgrade method, device, management server and the system of Edge Server |
Also Published As
| Publication number | Publication date |
|---|---|
| CN100414888C (en) | 2008-08-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1852156A (en) | Method for updating programe edition of processing system without operation interruption | |
| CN109445944B (en) | DPDK-based network data acquisition and processing system and method thereof | |
| Grasso et al. | LibWater: heterogeneous distributed computing made easy | |
| Maqsood et al. | Dynamic task mapping for network-on-chip based systems | |
| CN104156264B (en) | A kind of base band signal process tasks in parallel real-time scheduling method based on many GPU | |
| CN1516419A (en) | Syste mand method for processing customer computer request on host computer network | |
| CN1601476A (en) | System and method for CPI scheduling in SMT processor | |
| CN1906580A (en) | Method and system for a grid-enabled virtual machine with movable objects | |
| CN101196832B (en) | Resource proxy machine facing illumination variant equipment | |
| CN1645330A (en) | Method and system for grid-enabled virtual machines with distributed management of applications | |
| Göhringer et al. | CAP-OS: Operating system for runtime scheduling, task mapping and resource management on reconfigurable multiprocessor architectures | |
| US11347735B2 (en) | Scalable query processing | |
| CN102420709A (en) | Method and equipment for managing scheduling task based on task frame | |
| CN112835714A (en) | Container arrangement method, system and medium for CPU heterogeneous cluster in cloud edge environment | |
| CN1326567A (en) | Job-parallel processor | |
| US20230028008A1 (en) | Checkpoints in batch file processing | |
| CN111367632B (en) | Container cloud scheduling method based on periodic characteristics | |
| CN1306407C (en) | Method for autonomic self-learning in selecting resources for dynamic provisioning | |
| Suresh et al. | Automating cluster management with weave | |
| Nardelli | QoS-aware deployment and adaptation of data stream processing applications in geo-distributed environments | |
| Coppola et al. | Towards a common deployment model for grid systems | |
| Gong et al. | Collaborative inference via dynamic composition of tiny ai accelerators on mcus | |
| CN1595351A (en) | A processor multithreading method and apparatus based on MIPS instruction set | |
| Khalil et al. | Multi-agent model for job scheduling in cloud computing | |
| D'amico | Scheduling and resource management solutions for the scalable and efficient design of today's and tomorrow's HPC machines |
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 | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20201204 Address after: 225700 No.1 science and Technology Park, Zhangguo Town, Xinghua City, Taizhou City, Jiangsu Province Patentee after: Jiangsu Shuangjin New Material Co.,Ltd. Address before: 510640 Unit 2414-2416, Main Building, No. 371 Wushan Road, Tianhe District, Guangzhou City, Guangdong Province Patentee before: GUANGDONG GAOHANG INTELLECTUAL PROPERTY OPERATION Co.,Ltd. Effective date of registration: 20201204 Address after: 510640 Unit 2414-2416, Main Building, No. 371 Wushan Road, Tianhe District, Guangzhou City, Guangdong Province Patentee after: GUANGDONG GAOHANG INTELLECTUAL PROPERTY OPERATION Co.,Ltd. Address before: 518129 Bantian HUAWEI headquarters office building, Longgang District, Guangdong, Shenzhen Patentee before: HUAWEI TECHNOLOGIES Co.,Ltd. |
|
| TR01 | Transfer of patent right |