CN101222522A - Gridding task scheduling method considering gridding task importance and time urgency - Google Patents
Gridding task scheduling method considering gridding task importance and time urgency Download PDFInfo
- Publication number
- CN101222522A CN101222522A CNA2008100467070A CN200810046707A CN101222522A CN 101222522 A CN101222522 A CN 101222522A CN A2008100467070 A CNA2008100467070 A CN A2008100467070A CN 200810046707 A CN200810046707 A CN 200810046707A CN 101222522 A CN101222522 A CN 101222522A
- Authority
- CN
- China
- Prior art keywords
- task
- resource
- grid
- gridding
- agent
- 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 invention relates to a grid task scheduling method taking the importance and the time urgency of the grid task into consideration; the method adopted in the invention is that: an agent based network resource management model is established, wherein the model includes a grid user, a grid request, a grid task, grid resources and a grid router; the particular method of the grid task(Agent) is that: (1) based on characteristics and budget conditions of the task, a bid for the resource is taken part in, and the right-to-use of the resource is obtained at a certain ratio; (2) in the competitive bidding strategy of the grid task, how to bid is specifically defined in order to optimize the utility function of the grid task; (3) the main aim of the grid task is to complete the task as soon as possible under the capital budgeting limit. The invention takes the importance and the time urgency of each sub task in the task scheduling process into consideration, makes the quantitative definition and provides a model simulating the value law in the market economy; the model can guide the resource selection in the grid task bid and approximately ensure that the task is finally finished in the prospective date.
Description
Technical field
The invention belongs to the computer network grid task scheduling method, the grid task dispatching method of particularly a kind of considering gridding task importance and time urgency.
Background technology
Grid computing is the technological revolution for the third time after Internet, Web technology, also is accompanied by the Internet technology and grows up together.Along with development of science and technology, the appearance of the computational problem of extensive property in science, engineering and the commercial field, make single environment (P2P), single technology (cluster calculating) is powerless, it must seek a kind of cheap, the high-performance computing environment that disposal ability is superpower, grid come to this and arise at the historic moment.The core concept of grid computing is to seek a kind of virtual machine of supercomputing capability, it utilizes Internet technology at present all the fashion with the distributed heterogeneous resource on the geographical position, as server, work station, local area network (LAN), cluster, file system, processor, memory or the like overall sharing, this sharing is not that Internet of today just realizes uploading and downloading of information, it utilizes various agencies, realize the visit of transparent resource, make Internet constitute one super, the virtual processor of high-performance calculation ability.
Structurally, grid computing is actually the computational resource that utilizes the Internet will be dispersed on the different regions and organizes, and forms virtual " supercomputer ", and the computer of each participation is exactly one " node ", thousands of groups of nodes becomes a grid altogether.Grid computing has two advantages: the one, and superpower computing capability; Another is the idle computing capability that can make full use of in the network.Thereby make computational resource on the Internet, storage resources, overall sharing such as data resource, information resources, knowledge resource, Expert Resources also is fully utilized.
Resource is an isomery in the grid, mainly shows different on structure, configuration and the capacity of resource, and it comprises processor resource and the memory resource of the communal space and the various resources of other situations of the time of sharing; Resource belongs to multitube reason field in the grid, and all there is oneself management strategy in each field, this just make resource in the grid can not image set group (Cluster) in resource equally carry out centralized management, and necessary implementation distributed management strategy; The resource of grid is dynamic change; because in so extensive environment, there is resource to add wherein at any time, also there is resource to withdraw from wherein at any time; the factor that resource quantity available minimizing or the like dynamic change is also arranged is so grid must have the ability that can monitor change in resources in the grid in real time.The network of scale extend over the entire globe, resource isomery and a dynamic change like this certainly will be complicated unusually to the management and the scheduling of resource.
Therefore, on the one hand because grid environment is cheap, computing capability is superpower, make grid become the suitable environment of extensive property problem in solution science, engineering and the commercial field; On the other hand because characteristics such as the isomerism of the diversity of the distributivity on the resource geographical position in the grid environment, management strategy, resource distribution and dynamic make grid computing very challenging.The best approach of these two kinds of contradictions of balance can only be learnt from other's strong points to offset one's weaknesses, and overcomes the deficiency.Therefore, resource management in the grid and scheduling become the key problem of grid computing.
Summary of the invention
The purpose of this invention is to provide a kind of rational management gridding resource, optimize the considering gridding task importance of gridding task processing and the grid task dispatching method of time urgency.
To achieve these goals, the method applied in the present invention is:
Set up a kind of grid resource model, comprise in its model based on the agency:
Grid user: under the grid environment, the people who submits to gridding task to carry out in the grid environment;
Grid request Agent: the corresponding grid request Agent of each grid user, its responsibility is as follows:
1. receive the resource request of grid user;
2. seek satisfactory resource according to resource request;
3. the request with the user is decomposed into a plurality of subtasks, and each subtask is corresponding with a gridding task Agent;
4. the characteristic of each subtask of shining upon according to the resource request of grid request Agent and the price of resource market, the expense budget and the task that work up each task are finished the time limit;
5. according to dependence between the task and time sequencing, submit a tender by the gridding task Agent of each task correspondence and to obtain resource;
6. the result of calculation with each gridding task Agent gathers, and at last the result is returned grid user.
Gridding task Agent: each task all has gridding task Agent corresponding with it in computing grid, and gridding task Agent buys resource by competitive bidding to one or more gridding resource Agent, finishes calculation task, and its function comprises:
1. according to the characteristic of task and the budget situation of task, participate in the bid of certain resource and the right to use of acquisition certain proportion resource;
2. the competitive bidding strategy of gridding task Agent has defined it clearly how competitive bidding is to have optimized its utility function;
3. the main target of gridding task Agent is to finish the work as quickly as possible under certain capital budgeting restriction.
Gridding resource Agent: the main task of gridding resource Agent is:
1. apply for resource description is published to the gridding resource router;
2. sell resource with certain price to the grid request agency, wherein each gridding task Agent will obtain the resource of certain share of certain hour section, and the price of resource is by auction and competitive bidding decision;
3. provide the calling interface that uses this resource to grid request Agent.
The gridding resource router: the gridding resource router is finished following function:
1. resource registering/cancellation: resource router is the access device of grid computing resource, after computational resource is registered on resource router, has been equivalent to distribute in grid a unique identity to indicate, and can be shared by the whole mesh system;
2. resource routing iinformation collection/renewal: the resource routing iinformation is the information of relevant resource position, as the foundation of resource request being carried out route and forwarding, because the dynamic change of resource between the resource router, needs periodically to carry out the renewal of routing iinformation between router and the resource;
3. route/the forwarding of resource request: after resource router was received a resource request, it need select a road warp and it is transmitted to corresponding resource router for this request according to the resource routing iinformation.
The present invention fully takes into account the distance of gridding resource, and takes into full account the transmission time and the transmission cost of gridding task in the gridding task scheduling process.And importance and time urgency characteristics according to the subtask, a kind of scheduling of resource model is proposed: the law of value of the commodity under this modeling market economy, gridding task Agent is according to the importance factor of subtask and the sex factor gridding resource of submitting a tender targetedly it is pressed for time, average cost and speed budget that the disposal cost of the resource that obtains and execution speed are in task fluctuate, thereby can take into account the characteristic (importance and time urgency) of subtask, can guarantee roughly that again final total expense and time are still in the expense of expection with within the time.The present invention compares with the conventional mesh scheduling of resource, and the performance of its advantage is as follows: 1, the present invention proposes a kind of grid resource model (AGRM) based on the agency, and concise and to the point introduction the function of each module; 2, taken into full account the distance of resource in the grid system, and considered transmission time and the transmission cost of gridding task in assigning, for the grid system based on the market economy model of reality, this be quite reasonable also be necessary, and in present grid resource, seldom consider; 3, the importance and the time urgency of each subtask have been considered in the task scheduling process, and have made quantitative definition that this is that present gridding task scheduling is seldom considered; And this consideration has realistic meaning very much, because be not equal between each subtask, but free urgent and the importance difference; 4, proposed a kind of model of simulating the law of value under the market economy, this model can instruct gridding task to submit a tender and select resource, and the task that finally can roughly guarantee is finished in expection.
Description of drawings
Fig. 1 is the grid resource illustraton of model that the present invention is based on the agency.
Fig. 2 is expense of the present invention and time graph.
Fig. 3 is the present invention's method flow diagram of submitting a tender.
Embodiment
The present invention is described in further detail below in conjunction with accompanying drawing.
The present invention propose dispatching method be to be based upon under the market economy model, the task in the grid environment, the attribute of resource are at first described, take into full account the importance of task and time urgency and resource and the factors such as distance between the task; Simulate the law of value of commodity under the market economy then, propose a kind of task Agent bid and select the model of gridding resource, this model can guarantee about the average cost and speed that the gridding resource price selected and speed floats on task, thereby the assurance task can be finished in expection.Its concrete grammar is:
1, grid user is submitted to a group task T=(T of task agent
1T
2..., T
m), all there is certain length M I each subtask
i(i=1,2...m), whole task has budget (Budget) and two QoS constraints of deadline (DeadLine), and (B D), that is to say that this group subtask must be in expense B, and D finishes within the time, otherwise invalid.Wherein, the length of task represents that with MI (Million Instruct, 1,000,000 instructions) budget B represents with CU (Cost Unit, expense unit); Deadline D represents with S (Second, second).And because the importance and the time urgency of each subtask are different, the subtask is defined as follows:
Subtask T
i(MI
i, Q
i* B, Z
i* D)
MI wherein
iBe the length of subtask, Q
iExpression subtask importance factor, Z
iBe urgent factor task time, require below satisfying:
The antithetical phrase task significance is defined as follows:
Q
iB/MI
i＞B/M……………………………………………(1)
And it is the expense of the unit of subtask instruction is higher more than the average cost of general assignment, shows that the subtask is important more, otherwise inessential more;
Time urgency to the subtask is defined as follows:
MI
i/Z
iD＞MI/D……………………………………………………(2)
And subtask expectation to finish speed faster more than the average speed of general assignment, show that the subtask is urgent more, otherwise more not urgent;
And it is generally acknowledged, than higher subtask, should allow it execute as far as possible for important property; Can select the resource expense than higher for time ratio than urgent task, speed ratio resource execution faster simultaneously.
2, one group of resource set in the grid: (R
1, R
2..., R
n), each resource has (distance, processing speed are used price) (D
i, MIPS
i, CUPMI
i) attribute, represent respectively to act on behalf of distance between the gridding resource from gridding task, the execution speed of resource and execution cost, its unit uses respectively: parasang DU, MI/S, CU/MI represents.And the average resource router number in the hypothetical trellis on the unit distance be m (individual/DU), the forwarding time of the unit of each router (1,000,000) instruction is n (S), the forwarding expense of unit (1,000,000) instruction is k (CU).
3, grid task dispatching method
Gridding task Agent submits a tender and selects the resource method:
Gridding task Agent is according to the subtask characteristics (importance, time urgency) of the own distribution selection gridding resource of submitting a tender, and with its bargaining.Suppose the task T of i task Agent correspondence
i(MI
i, Q
iB, Z
iD) select j resource R
j(D
i, MIPS
i, CUPMI
i)
So, execution cost:
MI
i*D
j*k/m+MI
i*CUPMI
j
Time of implementation:
MI
i*D
j*n/m+MI
i/MIPS
j
If the subtask is spent height (Z it is pressed for time
i* D/MI
iVery little), allow this task Agent fast resource of those execution speeds of submitting a tender so, even its execution cost is very high, and this task has exceeded its expense budget; Otherwise the low resource of those execution speeds of just submitting a tender, the time budget of balance overall task; If subtask importance height (Q
i* B/MI is very big), allow this task Agent those execution costs high (it is generally acknowledged that the resource that execution cost is high may be more stable, better quality) of submitting a tender so, even execution speed is more slowly, and this task has exceeded its time budget; Otherwise the execution cost of just submitting a tender is hanged down a little resources, the expense budget of balance overall task.No matter above-mentioned the sort of situation, the cost curve of whole task and time graph should satisfy curve shown in Figure 2: that is to say that the resource expense of all task Agent and resource speed should fluctuate up and down in average cost and speed budget, thereby guarantee that last whole cost and time are within expection, and for the high task of time urgency, its execution cost can exceed the expense budget of oneself, otherwise should be lower than the budget of oneself; For the high task of importance, its time of implementation can be longer than the time budget of oneself, otherwise should be shorter than the time budget of oneself, and overall task should satisfy:
Task Agent submits a tender and selects the gridding resource method:
1. to each subtask Agent, according to its importance of task analysis and the time urgency characteristics of oneself assigning, specifically with reference to formula 1 and formula 2;
2. task Agent submits a tender and selects gridding resource, for the high task of importance, can submit a tender and select the high resource of execution cost, even its execution speed is slow, allow its expense and time on average speed/expense budget line, otherwise, under average speed/expense budget line.For the high task of time urgency, can submit a tender and select the fast resource of execution speed, even its execution cost exceeds the expense budget of oneself, otherwise, those execution speeds of submitting a tender resource low, that expense is also low;
3. 2. 1. repeating submits a tender until all task Agent selects to finish.
Task Agent bid result passes judgment on:
Whether the resource that above-mentioned all task Agent bids are selected is reasonable, see all resource expenses and VELOCITY DISTRIBUTION situation exactly: be distributed in average speed/expense budget both sides, the symmetry symmetry is just good more for curve, show that this resource of submitting a tender selection can satisfy feature (time urgency and the importance) requirement of task itself, overall task can be finished in the constraint of expection; Otherwise, just unreasonable more.
The content that is not described in detail in this specification belongs to this area professional and technical personnel's known prior art.
Claims (4)
1. the grid task dispatching method of considering gridding task importance and time urgency, the method that is adopted is:
Set up a kind of grid resource model, comprise in its model based on the agency:
Grid user: under the grid environment, the people who submits to gridding task to carry out in the grid environment;
Grid request Agent: the corresponding grid request Agent of each grid user;
Gridding task Agent: each task all has gridding task Agent corresponding with it in computing grid, and gridding task Agent buys resource by competitive bidding to one or more gridding resource Agent, finishes calculation task;
Gridding resource Agent;
The gridding resource router.
2. the grid task dispatching method of considering gridding task importance as claimed in claim 1 and time urgency is characterized in that: the concrete grammar of grid request Agent is:
1. receive the resource request of grid user;
2. seek satisfactory resource according to resource request;
3. the request with the user is decomposed into a plurality of subtasks, and each subtask is corresponding with a gridding task Agent;
4. the characteristic of each subtask of shining upon according to the resource request of grid request Agent and the price of resource market, the expense budget and the task that work up each task are finished the time limit;
5. according to dependence between the task and time sequencing, submit a tender by the gridding task Agent of each task correspondence and to obtain resource;
6. the result of calculation with each gridding task Agent gathers, and at last the result is returned grid user.
3. the grid task dispatching method of considering gridding task importance as claimed in claim 1 and time urgency is characterized in that: the concrete grammar of gridding task Agent is:
1. according to the characteristic of task and the budget situation of task, participate in the bid of certain resource and the right to use of acquisition certain proportion resource;
2. the competitive bidding strategy of gridding task Agent has defined it clearly how competitive bidding is to have optimized its utility function;
3. the main target of gridding task Agent is to finish the work as quickly as possible under certain capital budgeting restriction.
4. as the grid task dispatching method of claim 1 or 3 described considering gridding task importances and time urgency, it is characterized in that: the concrete steps of gridding task Agent competitive bidding are:
First step: the task of the distribution of each task Agent foundation following formula analysis oneself, time urgency that sets the tasks and importance characteristics,
Q
iB/MI
i＞B/M……………………………………………………(1)
MI
i/Z
iD＞MI/D……………………………………………………(2)
Wherein: MI
iBe the length of subtask, Q
iExpression subtask importance factor, Z
iBe urgent factor task time, B is budget, and D is the time;
Second step: task Agent submits a tender and selects gridding resource, for the high task of importance, submits a tender and selects the high resource of execution cost, even its execution speed is slow, allow its expense and time on average speed/expense budget line, otherwise, under average speed/expense budget line; For the high task of time urgency, submit a tender and select the fast resource of execution speed, even its execution cost exceeds the expense budget of oneself, otherwise, those execution speeds of submitting a tender resource low, that expense is also low;
Third step: repeat said process and select resource to finish until all task Agent bids.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2008100467070A CN101222522B (en) | 2008-01-17 | 2008-01-17 | Gridding task scheduling method considering gridding task importance and time urgency |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2008100467070A CN101222522B (en) | 2008-01-17 | 2008-01-17 | Gridding task scheduling method considering gridding task importance and time urgency |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101222522A true CN101222522A (en) | 2008-07-16 |
| CN101222522B CN101222522B (en) | 2011-08-10 |
Family
ID=39632087
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2008100467070A Expired - Fee Related CN101222522B (en) | 2008-01-17 | 2008-01-17 | Gridding task scheduling method considering gridding task importance and time urgency |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101222522B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102081554A (en) * | 2011-01-30 | 2011-06-01 | 浪潮(北京)电子信息产业有限公司 | Cloud computing operating system as well as kernel control system and method thereof |
| CN104244020A (en) * | 2014-09-22 | 2014-12-24 | 内蒙古工业大学 | Request routing method for could environment video live service |
| CN104765640A (en) * | 2014-01-02 | 2015-07-08 | 中国科学院声学研究所 | Intelligent service scheduling method |
| CN105338074A (en) * | 2015-10-21 | 2016-02-17 | 武汉工程大学 | Contract net task distribution method, obtaining method, intelligent proxy and MAS |
| CN107329822A (en) * | 2017-01-15 | 2017-11-07 | 齐德昱 | Towards the multi-core dispatching method based on super Task Network of multi-source multiple nucleus system |
| CN111191352A (en) * | 2019-12-18 | 2020-05-22 | 江苏科技大学 | System elastic recovery algorithm considering time and task importance |
| CN112591568A (en) * | 2020-12-15 | 2021-04-02 | 海南大学 | Device sharing method for fusion of data and information facing intention calculation and reasoning |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7467303B2 (en) * | 2004-03-25 | 2008-12-16 | International Business Machines Corporation | Grid mutual authorization through proxy certificate generation |
| CN101034362A (en) * | 2007-02-05 | 2007-09-12 | 南京邮电大学 | Method for implementing network job scheduling using mobile proxy |
-
2008
- 2008-01-17 CN CN2008100467070A patent/CN101222522B/en not_active Expired - Fee Related
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102081554A (en) * | 2011-01-30 | 2011-06-01 | 浪潮(北京)电子信息产业有限公司 | Cloud computing operating system as well as kernel control system and method thereof |
| CN104765640A (en) * | 2014-01-02 | 2015-07-08 | 中国科学院声学研究所 | Intelligent service scheduling method |
| CN104765640B (en) * | 2014-01-02 | 2018-02-16 | 中国科学院声学研究所 | A kind of intelligent Service dispatching method |
| CN104244020A (en) * | 2014-09-22 | 2014-12-24 | 内蒙古工业大学 | Request routing method for could environment video live service |
| CN104244020B (en) * | 2014-09-22 | 2017-12-12 | 内蒙古工业大学 | A kind of request method for routing of cloud environment net cast service |
| CN105338074A (en) * | 2015-10-21 | 2016-02-17 | 武汉工程大学 | Contract net task distribution method, obtaining method, intelligent proxy and MAS |
| CN107329822A (en) * | 2017-01-15 | 2017-11-07 | 齐德昱 | Towards the multi-core dispatching method based on super Task Network of multi-source multiple nucleus system |
| CN107329822B (en) * | 2017-01-15 | 2022-01-28 | 齐德昱 | Multi-core scheduling method based on hyper task network and oriented to multi-source multi-core system |
| CN111191352A (en) * | 2019-12-18 | 2020-05-22 | 江苏科技大学 | System elastic recovery algorithm considering time and task importance |
| CN112591568A (en) * | 2020-12-15 | 2021-04-02 | 海南大学 | Device sharing method for fusion of data and information facing intention calculation and reasoning |
| CN112591568B (en) * | 2020-12-15 | 2021-11-02 | 海南大学 | Device sharing method for fusion of data and information facing intention calculation and reasoning |
| WO2022127091A1 (en) * | 2020-12-15 | 2022-06-23 | 海南大学 | Intention calculation and reasoning-oriented data and information fusion device sharing method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101222522B (en) | 2011-08-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101222522B (en) | Gridding task scheduling method considering gridding task importance and time urgency | |
| CN100495984C (en) | Grid task dispatching method based on time and expense restriction | |
| Gao et al. | Auction-based VM allocation for deadline-sensitive tasks in distributed edge cloud | |
| CN108537440A (en) | A kind of building scheme project management system based on BIM | |
| CN103401939A (en) | Load balancing method adopting mixing scheduling strategy | |
| CN102281290A (en) | Emulation system and method for a PaaS (Platform-as-a-service) cloud platform | |
| CN114356587B (en) | Calculation power task cross-region scheduling method, system and equipment | |
| Wang et al. | A market-oriented incentive mechanism for emergency demand response in colocation data centers | |
| Zhou et al. | Concurrent workflow budget-and deadline-constrained scheduling in heterogeneous distributed environments | |
| Hao et al. | Adaptive energy-aware scheduling method in a meteorological cloud | |
| CN113283785A (en) | Cooperative scheduling optimization method for multi-task manufacturing resources | |
| Qian et al. | A workflow-aided Internet of things paradigm with intelligent edge computing | |
| CN101127700A (en) | Grid resource scheduling method based on network bandwidth and user charge | |
| Danak et al. | Efficient bidding in dynamic grid markets | |
| CN111045827A (en) | Time-validity task scheduling method based on resource sharing in cloud and fog environment | |
| Liu et al. | 5G network education system based on multi-trip scheduling optimization model and artificial intelligence | |
| CN104869154A (en) | Distributed resource scheduling method for balancing resource credibility and user satisfaction | |
| CN101098255A (en) | Option pricing model scheduling algorithm based implementation in interactive gridding system | |
| Volckaert et al. | Network aware scheduling in grids | |
| Wang et al. | Online scheduling of distributed machine learning jobs for incentivizing sharing in multi-tenant systems | |
| Gu et al. | A preemptive truthful VMs allocation online mechanism in private cloud | |
| Zhao et al. | Hypergraph-based task-bundle scheduling towards efficiency and fairness in heterogeneous distributed systems | |
| Sun et al. | T2FA: A heuristic algorithm for deadline-constrained workflow scheduling in cloud with multicore resource | |
| Rani et al. | Stochastic simulator for priority based task in grid environment | |
| Grosu | AGORA: an architecture for strategyproof computing in grids |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20110810 Termination date: 20130117 |