CN102360314A - System and method for managing resources of data center - Google Patents
System and method for managing resources of data center Download PDFInfo
- Publication number
- CN102360314A CN102360314A CN2011103348959A CN201110334895A CN102360314A CN 102360314 A CN102360314 A CN 102360314A CN 2011103348959 A CN2011103348959 A CN 2011103348959A CN 201110334895 A CN201110334895 A CN 201110334895A CN 102360314 A CN102360314 A CN 102360314A
- Authority
- CN
- China
- Prior art keywords
- resource
- request
- expression formula
- task
- node
- 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.)
- Pending
Links
Images
Abstract
The invention provides a system and a method for managing resources of a data center. The system comprises a resource manager and an execution task node, wherein the resource manager is used for receiving a job submission request of a user, receiving and storing a resource report request of the execution task node, and allocating jobs to the execution task node according to the job submission request and the resource report request; the job submission request comprises a job attribute expression; the resource report request comprises an internet protocol (IP) address of the execution task node and resource description; and the execution task node is used for submitting the resource report request and executing the jobs. By adoption of the system and the method, various requirements for resources of different types from different application loads can be met; the resources can be quickly and efficiently managed; the utilization rate of the resources can be improved; and operational cost is saved.
Description
Technical field
The present invention relates to the computer systems technology field, particularly a kind of data center resource management system and method.
Background technology
Data center (DataCenter) typically refers to the place that focuses on, stores, transmits, exchanges, manages of the information that in a physical space, realizes, is made up of one or more group of planes that is positioned at same machine room.A group of planes is to couple together the computer system that constitutes by numerous servers or workstation (workstation) through express network, and the server or the workstation that constitute a group of planes are called as node.
In recent years; Characteristics such as the internet, applications development is rapid, and it is many to demonstrate application type, and application scale is big; More specifically; Any active ues amount, visit capacity and business datum amount all present explosive the growth, make the data center's scale that supports internet, applications sharply enlarge, and extend to several ten thousand or even hundreds of thousands platform scale by hundreds of platform, several thousand.And; Because the diversity of application load characteristic, the diversity and the node resource dynamic in use of data center's node resource; Such as operation in groove (slot); In virtual machine, move, the diversity of the application program running environment of directly on operating system, moving has brought very big difficulty for the resource management of data center's group of planes.Management resource satisfies the different demands of various load to different resource how flexible and efficiently, dilatation fast automatically when business is in crest, and automatically quick capacity reducing when being in trough, its key is to realize the flexible resource management.
The resource management framework of existing data center is as shown in Figure 1, wherein, the 1st, data center's job manager, the 2nd, be divided into the node of several trough (slot).The node of each operation task is divided into several trough (slot) according to certain rule, in case delimit just no longer change, task of each groove (slot) operation, in case task is complete, this groove (slot) is just to new task of job manager application.This job manager is used to receive the operation that the user submits to, specifies a unique identifier for each operation, operation is sorted according to priority level, and according to this order assignment operation.This method is simple, be prone to realize, resource allocation methods is single; But because each operation is different to the demand of resource; And task actual employed resource in the process of carrying out is dynamic change, so can not satisfy the different demands of different application to various resources.
Summary of the invention
For solving the problems of the technologies described above, the object of the invention is to provide a kind of data center flexible resource management system and method.
According to one aspect of the invention, a kind of data center resource management system is provided, comprise explorer and execute the task node, wherein:
Said explorer; Be used to receive user's operation submission request; Receive and store the resource reporting request of the node of executing the task; And submit to request and said resource reporting request to distribute operation to the node of executing the task according to said operation, and wherein said operation is submitted to and is asked to comprise the job property expression formula, and said resource reporting request comprises execute the task IP addresses of nodes and resource description;
The said node of executing the task is used to submit to the resource reporting request and carries out operation.
According to a further aspect of the invention, a kind of data center method for managing resource is provided also, has comprised:
1) request is submitted in the operation that receives the user to, and wherein said operation submits to request to comprise the job property expression formula;
2) receive and the execute the task resource reporting request of node of storage, wherein said resource reporting request comprises execute the task IP addresses of nodes and resource description;
3) submit to request and said resource reporting request to distribute operation according to said operation to the node of executing the task.
In the said method, also comprise in the said step 3): select to execute the task node according to default resource demand expression formula.
In the said method, said operation submits to request also to comprise allocated resource demand expression formula; And
Also comprise in the said step 3): select to execute the task node according to said allocated resource demand expression formula.
In the said method, also comprise in the said step 3): select to execute the task node according to the default sort expression formula.
In the said method, said operation submits to request also to comprise appointment ordering expression formula, and
Also comprise in the said step 3): express expression formula according to said appointment ordering and select to execute the task node.
Said system of the present invention and method can satisfy the various demands of different application load to dissimilar resources; Can be fast and carry out resource management efficiently.And, can also improve resource utilization, practice thrift operation cost.
Description of drawings
Fig. 1 is data center's resource management Organization Chart of prior art;
Fig. 2 is data center's resource management Organization Chart in accordance with a preferred embodiment of the present invention;
Fig. 3 is a method for managing resource process flow diagram in accordance with a preferred embodiment of the present invention.
Embodiment
In order to make the object of the invention, technical scheme and advantage clearer, below in conjunction with accompanying drawing, to data center's resource management system and method further explain according to an embodiment of the invention.Should be appreciated that specific embodiment described herein only in order to explanation the present invention, and be not used in qualification the present invention.
Fig. 2 shows data center's resource management Organization Chart in accordance with a preferred embodiment of the present invention, wherein, the 1st, data center's job manager, the 2nd, explorer, the 3rd, the node of operation task.Compared to Figure 1, this framework is many explorer is used for the periodically current available resource of each node of record data center, and according to operation the demand of resource is confirmed that appropriate nodes comes operation task.Comprise a monitoring resource unit on each node, be used for periodically reporting the current available various resources of this node to explorer, like cpu, memory, disk etc., preferably, this monitoring resource unit is a monitoring resource process of operation.A node can move several tasks simultaneously.One of ordinary skill in the art will appreciate that; Comprise job manager in the preferred embodiment, it is used to receive the operation submission request that the user submits to, specifies a unique identifier for each operation and discerns to be used for explorer; Operation is sorted according to priority level; And submit to request to be transmitted to explorer the operation of correspondence according to this order, but this job manager is not necessary, can be directly receives operation that the user submits to by explorer yet and submits to and ask.
Fig. 3 is a method for managing resource process flow diagram in accordance with a preferred embodiment of the present invention, and it has expressed the resource reporting of node and the main flow process of submitting to request to carry out resources allocation according to operation intuitively, and it specifically comprises the steps:
A) request is submitted in the operation of the resource reporting request of the explorer wait node of data center and client to; Wherein comprise node IP address and resource description in this resource reporting request, as be the resource reporting request, execution in step b); As submitting request to, change step e) for operation.
B) according to IP addresses of nodes, on explorer, search the resource description of this node, if found, change step d), otherwise execution in step c).
C), content stores in this resource reporting request to explorer, is changeed step a) according to IP addresses of nodes.
D) upgrade resource description corresponding in the explorer, change step a).
E) submit to the request judges whether to specify demand according to operation,, otherwise change step g) if continue step f) to resource.
F) according to the resource description of being preserved in the explorer, confirm that all meet the node of the resource requirement of user's appointment, change step h).
G), confirm that all meet the node by the default resource demand of keeper's appointment according to resource description of being preserved in the explorer and default resource demand expression formula.
H) judge whether the node number meet demand is less than or equal to n, wherein n is the number of the task (task) that comprises in the operation, and wherein n＞=1 is then to continue step I), otherwise change step j).
I) return the node that all meet demand, change step a).
Whether j) submit to request to judge according to operation wherein comprises appointment ordering expression formula, is then to continue step k), otherwise change step l).
K) according to specifying the ordering expression formula that the node that meets demand is sorted, change step m).
L) according to the default sort expression formula of being preserved in the explorer node that meets demand is sorted.
M) return preceding n node after the ordering, change step a).
According to a preferred embodiment of the present invention, describe implementation and the associative operation that request is submitted in above-mentioned resource reporting request and operation to below in detail.
The resource reporting request has comprised IP addresses of nodes and resource description, and it can be expressed as character string.Resource description relates to the attribute of node, has below listed some nodal communitys commonly used.
● IP: expression IP addresses of nodes.
● Memory: represent the size of current free memory, be worth for greater than 0 and be less than or equal to the integer of machine maximum memory that unit is M.
● Cpuusage: the utilization rate of expression current C PU, i.e. number percent is worth for more than or equal to 0 and less than 1 decimal.
● Disk: the size of the free space of expression disk, unit is M.
● DiskRate: the read or write speed of expression disk, be worth for greater than 0 integer, unit is MB/s.
● DiskIn: the byte number of input in the expression current period disk unit interval.
● DiskOut: the byte number of output in the expression current period disk unit interval.
● EthRate: the read or write speed of expression network interface card, be worth for greater than 0 integer, unit is Mb/s.
● EthIn: the byte number that network interface card reads in the expression current period unit interval.
● EthOut: the byte number that network interface card sends in the expression current period unit interval.
● Cores: the number of the CPU nuclear of expression node.
● JobId: the unique identification of the operation that moves on the expression node.
The monitoring resource process of moving on the node is periodically obtained the various available resources of this machine, and through such as the mode of remote function calls (RPC) resource description being reported to explorer, particularly, with " key
1=value
1...; Key
n=value
n", " Cpuusage=0.3 for example; Memory=1000; ... ", with the form of character string a parameter, report explorer, key as the RPC call function
1... key
nThe title of representing various resources is called Resource Properties again, value
1... value
nThe value of representing each corresponding Resource Properties.
The resource reporting request of all nodes of explorer storage.This mode one of ordinary skill in the art will appreciate that the resource reporting request can be stored with different modes such as Hash array or map structures, as long as can realize with node and its resource corresponding stored.Be that example is described the storage of resource reporting request on explorer with the Hash array below.With the resource description request of each node a element as the Hash array; This element can be organized as the form of structure; IP addresses of nodes is as a variable in this structure, and the arbitrary node attribute in the resource description is also as the variable in this structure.Preferably, the subscript of element in the Hash array also identified IP addresses of nodes.Mode stored resource for the map structure reports request, can be with IP addresses of nodes as a key word (Key), and as value (Value), preferably, this resource description is organized as the form of structure with the resource description of this node.
Operation submits to request can comprise some attribute expression formulas of operation self; Preferably; Said operation is submitted to request also to comprise allocated resource demand expression formula and/or is specified the ordering expression formula; The job property expression formula is used to describe some characteristics of operation self, and this allocated resource demand expression formula is used for by the demand of user's assignment to resource, and this appointment ordering expression formula is used for specifying the node sequencing algorithm.
The job property expression formula can represent with the version of Key=Value, has listed some Property Names commonly used below:
Cmd: the title of executable program, for example a.out.
Args: by the executable program parameters needed of Cmd description.
TaskCount: the number of carrying out of task, the number of the executable program that promptly will carry out.
TransferFiles: need copy the file on the node to, this node is the task actuator.
TaskType: the type of operation, such as common operation, cron operation (operation of regularly carrying out) etc.
Allocated resource demand expression formula can be Requirements=Expression1, and wherein Requirements also can ((character string that {) and underscore (_) constitute be replaced for (), brace by capitalization (A-Z) arbitrarily, lowercase (a-z), numeral (0-9), parentheses.Expression1 representes an expression formula, for example meets the logical expression of logical expression standard in the C language.According to a preferred embodiment of the present invention, this allocated resource demand expression formula for example: Requirements=Memory＞10&&Cpuusage＜0.5.
In order to increase user's dirigibility; In this allocated resource demand expression formula; The user can specify a machine or a collection of both candidate nodes to come running job through the IP address; Allocated resource demand expression formula can be write as Requirements=IP=={10.0.0.1, and 10.0.0.2}, user can also specify two operations to have to operate on the machine; To reduce by two communication overheads between the operation, then allocated resource demand expression formula can be write as the form of Requirements=(JobId==100).
One of ordinary skill in the art will appreciate that; Operation is submitted in the request also can not provide resource requirement; Do not providing under the resource requirement situation; The resource requirement that can adopt the keeper to give tacit consent to, perhaps arbitrarily specified node is executed the task, the resource requirement of this keeper's acquiescence for example: Requirements=Memory＞50&&Cpuusage＜0.8&&Disk＞500.
Specify the ordering expression formula can be Rank=Expression2, wherein Rank also can be replaced by the character string that capitalization (A-Z), lowercase (a-z), numeral (0-9) and underscore () arbitrarily constitute.Expression2 representes an expression formula, for example meets the arithmetic expression of logical expression standard in the C language.Step k for said method), can corresponding nodes be sorted according to the size of Rank value.
One of ordinary skill in the art will appreciate that; Except through specifying the ordering expression formula to specify an ordering strategy to select the only machine, also ordering strategy can not specified, for the latter; Can adopt the default sort expression formula of keeper's appointment; In accordance with a preferred embodiment of the present invention, specify Rank=Cpuusage, and sort according to the Rank value.The preferred embodiment can concentrate on task on the plurality of nodes.When explorer according to the state-detection of node Cpu when business is in trough, task only can be moved on some nodes, remaining node then can be in dormant state; When being in crest, being in dormant node can be waken up gradually, all is in active state or all tasks can both find resource up to all nodes.
The user can wait other modes to submit request to job manager submit job and operation with the api function interface or through order through RPC.Operation submits to request to be transmitted to explorer by job manager.If with the mode submit job of RPC function call, the same with the resource reporting request, with allocated resource demand expression formula and/or specify ordering expression formula and job property with " A1=B1; ... An=Bn " character string of form is as the parameter of RPC function interface; If with the mode submit job of order, need so " A1=B1; ... An=Bn " write on the text, with-the file=filename, as the parameter of operation submiting command.
Explorer is whenever received an operation submission request; Just go for and seek the node that meets demand according to wherein resource requirement; Calculation Method is: for each node; The name of all properties in the expression formula on Requirements the right is replaced to the pairing value of this attribute, judge that then whether this expression formula is true, meets demand if then show the node that finds.For example: " Requirements=Memory＞100&&Cpuusage＜0.5 ", being put into corresponding position in the expression formula of the right to the value of the value of Memory in each node and Cpuusage, the value of judgment expression is very or vacation then.After having confirmed to meet the node of resource requirement, submit to the appointment ordering expression formula in the request to go ordering according to operation, the mode that the algorithm of ordering and above-mentioned is looked for node is similar.Then, select to come top one or more node and carry out operation, preferably, select a plurality of nodes, this is because an operation comprises a plurality of tasks usually.
One of ordinary skill in the art will appreciate that the job property in the request is submitted in the nodal community and the operation that can increase according to concrete demand or reduce in the resource description to
Lot of advantages of the present invention has guaranteed that concrete advantage is following to the elasticity at data center and efficient resource management:
One of advantage of the present invention is to solve the different demands of different application load to dissimilar resources well; No matter how complicacy can be described out with logical expression to the demand of resource in operation; And support application program operates in the different environment, for example directly on operating system, slot is last or virtual machine is first-class.Concrete example as: if application program need operate among the slot, then report the number of idle slot, the user can be that unit proposes resource request with slot.
Two of advantage of the present invention is management resources fast and efficiently.When receiving the resource reporting request of node; If what adopt is the structure of Hash array, then can find corresponding resource description through the IP address rapidly, upgrade; Because node adopts periodic resource reporting; So explorer can perceive the dynamic change of resource delicately, when receiving resource request, just can find the node that meets demand as long as calculate the logical value of Requirements; Then the node that meets demand is carried out quicksort according to the Rank value, so the ability fast selecting meets the node of demand.
Three of advantage of the present invention is ability dilatation capacity reducings automatically, improves resource utilization, reduces power consumption, saves operation cost.For example, the user can select task is concentrated on the sort algorithm on some the machines, and perhaps the keeper can be with such sort algorithm as default algorithm.Because business exists crest and trough; Task is only moved on some nodes when being in trough; Remaining node then can be in dormant state; When being in crest, being in dormant node can be waken up gradually, all is in active state or all tasks can both find resource up to all nodes.
Four of advantage of the present invention is very strong dirigibility to the user to be provided.The user can fill in any resource requirement, can give operation a minimum resource quota under the default situations; The user also can fill in the stock number that needs, and can fill out all types of resources and also can only fill out part, can also refer to fix on certain the specific node and executes the task, and can specify this task to want to monopolize a node, in a word, can satisfy the various demands of user.
Five of advantage of the present invention is very strong extendability to the manager of programming to be provided.The description of resource can be different according to actual conditions; For example: if the network performance of data center is not fine; Need frequent communicating between two tasks again, so preferably these two tasks are put into a machine, so preferably; Comprise the Resource Properties of a description in the resource reporting request, and the requirement that the user writes this Resource Properties gets final product in Requirements in submit job about the All Jobs that moves on the node.
Should be noted that and understand, under the situation that does not break away from the desired the spirit and scope of the present invention of accompanying Claim, can make various modifications and improvement the present invention of above-mentioned detailed description.Therefore, the scope of the technical scheme of requirement protection does not receive the restriction of given any specific exemplary teachings.
Claims (15)
1. data center's resource management system comprises explorer and executes the task node, wherein:
Said explorer; Be used to receive user's operation submission request; Receive and store the resource reporting request of the node of executing the task; And submit to request and said resource reporting request to distribute operation to the node of executing the task according to said operation, and wherein said operation is submitted to and is asked to comprise the job property expression formula, and said resource reporting request comprises execute the task IP addresses of nodes and resource description;
The said node of executing the task is used to submit to the resource reporting request and carries out operation.
2. system according to claim 1; It is characterized in that, also comprise job manager, be used to receive user's operation submission request; And, submit to request to be transmitted to said explorer said operation according to ranking results according to the said operation of the prioritization of said operation.
3. system according to claim 1 and 2 is characterized in that, said explorer is stored said resource reporting request with the mode of Hash array or map structure.
4. according to each described system in the claim 1 to 3, it is characterized in that, also store default resource demand expression formula in the said explorer, said explorer also is used for selecting to execute the task node according to said default resource demand expression formula.
5. according to each described system in the claim 1 to 3, it is characterized in that, also store the default sort expression formula in the said explorer, said explorer also is used for selecting to execute the task node according to said default sort expression formula.
6. system according to claim 5 is characterized in that, comprises the utilization rate of CPU in the said resource reporting request, and said default sort expression formula is the utilization rate of Rank=Cpuusage.
7. according to each described system in the claim 1 to 3, it is characterized in that said operation submits to request to comprise that also allocated resource demand expression formula, said explorer also are used for selecting to execute the task node according to said allocated resource demand expression formula.
8. according to each described system in the claim 1 to 3, it is characterized in that said operation submits to request also to comprise appointment ordering expression formula, said explorer also is used for selecting to execute the task node according to said appointment ordering expression formula.
9. according to each described system in the claim 1 to 3, it is characterized in that, comprise the sign of moving on the executing the task node of task in the said resource reporting request.
10. according to each described system in the claim 1 to 3, it is characterized in that the said node of executing the task comprises the monitoring resource unit, is used to submit to the resource reporting request.
11. data center's method for managing resource comprises:
1) request is submitted in the operation that receives the user to, and wherein said operation submits to request to comprise the job property expression formula;
2) receive and the execute the task resource reporting request of node of storage, wherein said resource reporting request comprises execute the task IP addresses of nodes and resource description;
3) submit to request and said resource reporting request to distribute operation according to said operation to the node of executing the task.
12. method according to claim 11 is characterized in that, also comprises in the said step 3): select to execute the task node according to default resource demand expression formula.
13. method according to claim 11 is characterized in that,
Said operation submits to request also to comprise allocated resource demand expression formula; And
Also comprise in the said step 3): select to execute the task node according to said allocated resource demand expression formula.
14. according to claim 11,12 or 13 described methods, it is characterized in that, also comprise in the said step 3): select to execute the task node according to the default sort expression formula.
15. according to claim 11,12 or 13 described methods, it is characterized in that,
Said operation submits to request also to comprise appointment ordering expression formula, and
Also comprise in the said step 3): express expression formula according to said appointment ordering and select to execute the task node.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011103348959A CN102360314A (en) | 2011-10-28 | 2011-10-28 | System and method for managing resources of data center |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011103348959A CN102360314A (en) | 2011-10-28 | 2011-10-28 | System and method for managing resources of data center |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102360314A true CN102360314A (en) | 2012-02-22 |
Family
ID=45585645
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2011103348959A Pending CN102360314A (en) | 2011-10-28 | 2011-10-28 | System and method for managing resources of data center |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102360314A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103049326A (en) * | 2013-01-16 | 2013-04-17 | 浪潮(北京)电子信息产业有限公司 | Method and system for managing job program of job management and scheduling system |
| CN103383653A (en) * | 2012-05-02 | 2013-11-06 | 中国科学院计算技术研究所 | Method and system for managing and dispatching cloud resource |
| CN103810046A (en) * | 2012-11-15 | 2014-05-21 | 百度在线网络技术（北京）有限公司 | Method and system for managing resources in single machine |
| CN104038540A (en) * | 2014-06-03 | 2014-09-10 | 山东乾云启创信息科技有限公司 | Method and system for automatically selecting application proxy server |
| CN105897837A (en) * | 2015-12-07 | 2016-08-24 | 乐视云计算有限公司 | Content distribution task submitting method and system |
| CN105975334A (en) * | 2016-04-25 | 2016-09-28 | 深圳市永兴元科技有限公司 | Distributed scheduling method and system of task |
| CN106790092A (en) * | 2016-12-23 | 2017-05-31 | 北京奇艺世纪科技有限公司 | Remote procedure call services end control system and method |
| CN106844040A (en) * | 2016-12-20 | 2017-06-13 | 北京并行科技股份有限公司 | A kind of job submission method, system and server |
| CN108073453A (en) * | 2016-11-11 | 2018-05-25 | 阿里巴巴集团控股有限公司 | The dispatching method and device of cpu resource in distributed type assemblies |
| WO2019184739A1 (en) * | 2018-03-29 | 2019-10-03 | 阿里巴巴集团控股有限公司 | Data query method, apparatus and device |
| CN111240857A (en) * | 2020-01-03 | 2020-06-05 | 上海瀚之友信息技术服务有限公司 | Remote service calling system and method |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1917464A (en) * | 2006-09-01 | 2007-02-21 | 清华大学 | Distribution type task assignment and computation method based on lower bound to be raised step by step |
| CN101034362A (en) * | 2007-02-05 | 2007-09-12 | 南京邮电大学 | Method for implementing network job scheduling using mobile proxy |
| CN102073546A (en) * | 2010-12-13 | 2011-05-25 | 北京航空航天大学 | Task-dynamic dispatching method under distributed computation mode in cloud computing environment |
-
2011
- 2011-10-28 CN CN2011103348959A patent/CN102360314A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1917464A (en) * | 2006-09-01 | 2007-02-21 | 清华大学 | Distribution type task assignment and computation method based on lower bound to be raised step by step |
| CN101034362A (en) * | 2007-02-05 | 2007-09-12 | 南京邮电大学 | Method for implementing network job scheduling using mobile proxy |
| CN102073546A (en) * | 2010-12-13 | 2011-05-25 | 北京航空航天大学 | Task-dynamic dispatching method under distributed computation mode in cloud computing environment |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103383653A (en) * | 2012-05-02 | 2013-11-06 | 中国科学院计算技术研究所 | Method and system for managing and dispatching cloud resource |
| CN103810046A (en) * | 2012-11-15 | 2014-05-21 | 百度在线网络技术（北京）有限公司 | Method and system for managing resources in single machine |
| CN103049326B (en) * | 2013-01-16 | 2015-04-15 | 浪潮(北京)电子信息产业有限公司 | Method and system for managing job program of job management and scheduling system |
| CN103049326A (en) * | 2013-01-16 | 2013-04-17 | 浪潮(北京)电子信息产业有限公司 | Method and system for managing job program of job management and scheduling system |
| CN104038540A (en) * | 2014-06-03 | 2014-09-10 | 山东乾云启创信息科技有限公司 | Method and system for automatically selecting application proxy server |
| CN104038540B (en) * | 2014-06-03 | 2017-03-22 | 山东乾云启创信息科技股份有限公司 | Method and system for automatically selecting application proxy server |
| CN105897837A (en) * | 2015-12-07 | 2016-08-24 | 乐视云计算有限公司 | Content distribution task submitting method and system |
| CN105975334A (en) * | 2016-04-25 | 2016-09-28 | 深圳市永兴元科技有限公司 | Distributed scheduling method and system of task |
| CN108073453A (en) * | 2016-11-11 | 2018-05-25 | 阿里巴巴集团控股有限公司 | The dispatching method and device of cpu resource in distributed type assemblies |
| CN108073453B (en) * | 2016-11-11 | 2022-05-10 | 阿里巴巴集团控股有限公司 | Method and device for scheduling CPU (Central processing Unit) resources in distributed cluster |
| CN106844040B (en) * | 2016-12-20 | 2020-08-28 | 北京并行科技股份有限公司 | Job submitting method, system and server |
| CN106844040A (en) * | 2016-12-20 | 2017-06-13 | 北京并行科技股份有限公司 | A kind of job submission method, system and server |
| CN106790092A (en) * | 2016-12-23 | 2017-05-31 | 北京奇艺世纪科技有限公司 | Remote procedure call services end control system and method |
| CN106790092B (en) * | 2016-12-23 | 2020-06-19 | 北京奇艺世纪科技有限公司 | Remote procedure call server control system and method |
| CN110321214A (en) * | 2018-03-29 | 2019-10-11 | 阿里巴巴集团控股有限公司 | A kind of data query method, device and equipment |
| WO2019184739A1 (en) * | 2018-03-29 | 2019-10-03 | 阿里巴巴集团控股有限公司 | Data query method, apparatus and device |
| US11556541B2 (en) | 2018-03-29 | 2023-01-17 | Alibaba Group Holding Limited | Data query method, apparatus and device |
| CN111240857A (en) * | 2020-01-03 | 2020-06-05 | 上海瀚之友信息技术服务有限公司 | Remote service calling system and method |
| CN111240857B (en) * | 2020-01-03 | 2023-02-21 | 上海瀚之友信息技术服务有限公司 | Remote service calling system and method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102360314A (en) | System and method for managing resources of data center | |
| JP6961844B2 (en) | Storage volume creation method and device, server, and storage medium | |
| CA2978889C (en) | Opportunistic resource migration to optimize resource placement | |
| CN109565515B (en) | System, apparatus, and process for dynamic tenant fabric adjustment in a distributed resource management system | |
| US9460185B2 (en) | Storage device selection for database partition replicas | |
| US9489443B1 (en) | Scheduling of splits and moves of database partitions | |
| US20160275123A1 (en) | Pipeline execution of multiple map-reduce jobs | |
| US10158709B1 (en) | Identifying data store requests for asynchronous processing | |
| JPWO2012066640A1 (en) | Computer system, migration method and management server | |
| JP4748950B2 (en) | Storage area management method and system | |
| US9841989B2 (en) | Parallel data processing system, computer, and parallel data processing method | |
| US11080207B2 (en) | Caching framework for big-data engines in the cloud | |
| WO2011071104A1 (en) | Distributed file system, data selection method of same and program | |
| WO2021057108A1 (en) | Data reading method, data writing method, and server | |
| JP3944176B2 (en) | Search request transmitting apparatus and program | |
| WO2021139431A1 (en) | Data synchronization method and apparatus for microservice, electronic device and storage medium | |
| CN107528871B (en) | Data analysis in storage systems | |
| JP2011100263A (en) | Virtual computer system, virtual computer management method and management program | |
| US8832176B1 (en) | Method and system for processing a large collection of documents | |
| JP6418419B2 (en) | Method and apparatus for hard disk to execute application code | |
| US10944814B1 (en) | Independent resource scheduling for distributed data processing programs | |
| US9870152B2 (en) | Management system and management method for managing data units constituting schemas of a database | |
| US10824640B1 (en) | Framework for scheduling concurrent replication cycles | |
| JP2013088920A (en) | Computer system and data management method | |
| JP2012181673A (en) | Resource allocation device, resource allocation method, and program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20120222 |