CN104461722A - Job scheduling method used for cloud computing system - Google Patents
Job scheduling method used for cloud computing system Download PDFInfo
- Publication number
- CN104461722A CN104461722A CN201410774633.8A CN201410774633A CN104461722A CN 104461722 A CN104461722 A CN 104461722A CN 201410774633 A CN201410774633 A CN 201410774633A CN 104461722 A CN104461722 A CN 104461722A
- Authority
- CN
- China
- Prior art keywords
- job
- queue
- subjob
- scheduling method
- cloud computing
- 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
Landscapes
- Multi Processors (AREA)
Abstract
The invention discloses a job scheduling method used for a cloud computing system. A system adopted in the job scheduling method comprises a global job queue, sub job queues, a job scheduler and a virtual machine corresponding to the job queue. The job scheduling method includes the global job queue receives jobs reaching the system, and the job scheduler schedules the jobs in the global job queue into proper sub job queues for queuing to wait for being executed by the virtual machine; the job scheduling method is based on a first-in first-out working mode, wherein scheduling rules include that when the jobs reach the global job queue, only when remaining buffer space capacity of the sub job queues is larger than 0 or job maximum waiting distribution time delay is smaller than minimum remaining execution time of the job which is on the way of being executed by the virtual machine, the jobs wait for being scheduled to the sub job queues, otherwise, the jobs are abandoned. By the job scheduling method, system resource utilization rate can be increased, time of job requesters is saved, and service quality of the cloud computing system is improved.
Description
Technical field
The present invention relates to field of cloud calculation, be specifically related to a kind of job scheduling method for cloud computing system.
Background technology
Cloud computing, as a kind of emerging computation schema, has the process storage capacity of magnanimity, and has the features such as efficient, virtual and cheap, therefore obtains showing great attention to of business circles and academia.Cloud computing system can be deployed on cheap common server, by network for user provides different services, and the various operations that execution efficiently and leading subscriber are submitted to.And the whether excellent execution efficiency affecting cloud computing of job scheduling algorithm, outstanding job scheduling algorithm can reduce the scheduling time of operation, the resource utilization of raising system, the key problem of therefore cloud computing how to reach efficient job scheduling and Resourse Distribute.
Common job scheduling algorithm has three kinds: the job scheduling algorithm of the job scheduling algorithm of first in first out, the job scheduling algorithm of the fair share based on the large job run of guarantee of Facebook exploitation, the computing power based on capacity of Yahoo exploitation.These cloud computing job scheduling schemes also exist deficiency: do not consider each job queue remaining cache space, operation patient maximum wait distribute residue execution time of time delay and the current execution operation of virtual machine, operation can not be assigned with within its institute's patient stand-by period, still to lose time to wait for.Like this, this operation not only occupies system resource, reduces the resource utilization of system, also wastes the time of job request person, makes Consumer's Experience bad.
Summary of the invention
The present invention, just based on above-mentioned technical matters, proposes a kind of job scheduling method for cloud computing system, and improve resource utilization and the service quality of cloud computing system, the technical scheme of employing is as follows:
A kind of job scheduling method for cloud computing system, the system adopted in method comprises overall job queue, subjob queue, job scheduler and the virtual machine corresponding with subjob queue, described overall job queue receives the operation of arrival system, job scheduling in overall job queue is queued up by described job scheduler to suitable subjob queue, wait virtual machine performs, described job scheduling method is based on first in first out working method, wherein, scheduling rule is: when operation arrives overall job queue, group job queue remaining cache spatial content is only had to be greater than 0, or operation maximum wait distribution time delay is less than the least residue execution time that virtual machine is performing operation, operation just can be waited for and be scheduled for subjob queue, otherwise, this operation will be dropped.
Traditional first in first out dispatching method, do not consider each subjob queue remaining cache spatial content, operation patient maximum wait distribute residue execution time of time delay and the current execution operation of virtual machine, operation can not be assigned with in its patient maximum latency of institute, still to lose time etc. to be allocated, occupying system resources, reduce resource utilization ratio, the time of waste job request person.And the job scheduling rule that the present invention increases makes, be 0 in each subjob queue remaining cache space and operation maximum wait distributes time delay when being less than the least residue execution time performing operation in each virtual machine, this operation can not to continue etc. to be dispensed, but be directly dropped, discharge the system resource that this operation takies, improve resource utilization ratio, save the time of job request person simultaneously.
As preferably, described maximum wait is distributed time delay and is not more than the difference that operation next operation adjacent with this operation arrives the time of overall job queue.
Like this, before the adjacent next operation of this operation arrives, this operation must be processed, or is scheduled, or is dropped, and can avoid the long-time occupying system resources of a certain operation like this, causes operation thereafter not have resource to utilize.
As preferably, the present invention comprises further, when operation arrives overall job queue, reads this operation end-to-end time delay requirement, and when end-to-end time delay requires to be satisfied, this operation just can be waited for and be scheduled for subjob queue, otherwise this operation will be dropped.
In the present invention, end-to-end time delay is identical with the end-to-end time delay concept in computer network and implication, and namely operation is transferred to the time required for subjob queue from overall job queue.When the delay requirement of operation can not be satisfied, namely the delay requirement of operation is less than operation when being transferred to the time required for subjob queue from overall job queue, this operation also can not to continue etc. to be dispensed, so just can free system resources further, improve resource utilization and service quality, improve Consumer's Experience.
As preferably, the present invention comprises further, the reason that the different situations be dropped according to operation are dropped to user's backtracking, be specially: group job queue remaining cache spatial content is 0, and operation maximum wait distributes time delay when being less than the least residue execution time performing operation in virtual machine, returns " queue full " information to user; When the end-to-end time delay of operation requires to be satisfied, return " time-out " information to user.
By the reason be dropped to job request person's backtracking, make user can grasp job run situation in real time, and ensuing action can be determined according to feedback message, comprise and resubmit operation or abandon performing operation, improve service quality, improve Consumer's Experience.
As preferably, the present invention comprises further, calculates the previous operations execution time sum of each subjob queue, the job scheduling in overall job queue is arrived, and queues up in the subjob queue that previous operations execution time sum is the shortest.
The operation that some subjob queue is queued up is less, but some operation wherein but needs the execution time grown very much, if ignore this point, directly by job-shop in the less subjob queue of operation, the problems such as the operation stand-by period is unreasonable, and job scheduling efficiency is low can be caused.And the present invention is by calculating the previous operations execution time sum of subjob queue, namely in each subjob queue virtual machine just in the residue execution time of running job and the execution time sum of queued jobs.Then by meeting the job scheduling of the rule that is scheduled in the shortest subjob queue of the previous operations execution time sum of subjob queue, decreasing the stand-by period of operation, improve the efficiency of job scheduling and the service quality of cloud computing system.
Particularly, a kind of job scheduling method for cloud computing system of the present invention comprises the following steps:
S1. operation arrives overall job queue;
S2. read the end-to-end time delay requirement of this operation, distribute job number, the patient maximum wait time delay of this operation is set, estimates the execution time of this operation;
S3. judged whether that subjob queue remaining cache spatial content is greater than 0, if so, then performed S4, if not, then perform S8;
S4. the previous operations execution time sum of each subjob queue is calculated;
S5. judge whether this operation meets end-to-end time delay requirement, if so, then performs S6, if not, then perform S7;
S6. this operation enters the shortest subjob queue of previous operations execution time sum, terminates this scheduling;
S7. abandon this operation and " time-out " information that sends to job request person, terminate this scheduling;
S8. judge that maximum wait is distributed time delay and whether is greater than the least residue execution time that virtual machine is performing operation: if so, then perform step S9, if not, then perform step S10;
S9., after the least residue execution time is waited in this operation, step S4 is performed;
S10. abandon this operation and " queue full " information that sends to job request person, terminate this scheduling.
Wherein, estimate in step S2 that the method for Job execution time is as follows:
If the size of operation is F, the cloud computing system resource taken is C, then the execution time of this operation is estimated as t, then t meets, and wherein cloud computing system resource C is the virtual center processor of 256,000,000 internal memories.
As preferably, the job number in described step S2 was uniquely constant before this operation completes.
The present invention is that a job number is set up in operation, and this job number is responsible for packaging operation and recorded information, ensures that job number is uniquely constant between operational period, to follow the tracks of Job execution state and process.
Particularly, in described subjob queue remaining cache spatial content, each virtual machine, current execution operation remains the previous operations execution time sum of execution time and each subjob queue by job scheduler monitoring calculation.
As preferably, described in the present invention, the length of overall job queue meets: described job scheduler completes overall job queue required time and is not more than the difference that operation next operation adjacent with this operation arrives the time of overall job queue.
Such restriction is done to the length of overall job queue, the operation caused because the medium operation number to be scheduled of overall job queue exceedes maximum operation number that overall job queue can hold can be avoided to run off, reduce the blocking rate of operation, improve service quality.
As preferably, the queue of subjob described in the present invention is one or more, and its queue length meets: virtual machine completes subjob queue required time and is not more than the difference that operation next operation adjacent with this operation arrives the time of subjob queue.
The length of antithetical phrase job queue does such restriction, the medium operation number to be scheduled of factor job queue can be avoided to exceed maximum operation number that subjob queue can hold and the operation that causes is run off, and reduces the blocking rate of operation, improves service quality.
Beneficial effect of the present invention: compared with prior art, the present invention consider when carrying out job scheduling operation patient maximum wait distribute the end-to-end time delay of time delay, subjob queue remaining cache space and job transfer, improve resource utilization ratio, save the time of job request person; Be dropped reason to job request person's backtracking, make user can grasp job run situation in real time; Job scheduling is queued up in the shortest subjob queue of previous operations execution time sum, decreases the stand-by period of operation, improve the service quality of system.Do rationally to limit to the length of overall job queue and subjob queue, reduce the blocking rate of operation, improve service quality.
Accompanying drawing explanation
Fig. 1 is scheduling rule schematic diagram of the present invention;
Fig. 2 is processing flow chart of the present invention.
Embodiment
Below in conjunction with the drawings and specific embodiments, the invention will be further described.
Embodiment:
As shown in Figure 1, a kind of job scheduling method for cloud computing system, the system adopted in method comprises overall job queue, subjob queue, job scheduler and the virtual machine corresponding with subjob queue, described overall job queue receives the operation of arrival system, job scheduling in overall job queue is queued up by described job scheduler to suitable subjob queue, wait virtual machine performs, described job scheduling method is based on first in first out working method, wherein, scheduling rule is: when operation arrives overall job queue, group job queue remaining cache spatial content is only had to be greater than 0, or operation maximum wait distribution time delay is less than the least residue execution time that virtual machine is performing operation, operation just can be waited for and be scheduled for subjob queue, otherwise, this operation will be dropped.
As preferably, in the present embodiment, described maximum wait is distributed time delay and is not more than the difference that operation next operation adjacent with this operation arrives the time of overall job queue.
The difference making maximum wait distribute time delay equaling operation time of arrival next operation adjacent with this operation time of arrival, namely before the adjacent next operation of this operation arrives, this operation must be processed, be scheduled, be dropped, can avoid the long-time occupying system resources of a certain operation like this, cause operation thereafter not have resource to utilize.
As preferably, the present embodiment comprises further, when operation arrives overall job queue, read this operation end-to-end time delay requirement, when end-to-end time delay requires to be satisfied, this operation just can be waited for and be scheduled for subjob queue, otherwise this operation will be dropped.
As preferably, the present embodiment comprises further, the reason that the different situations be dropped according to operation are dropped to user's backtracking, be specially: group job queue remaining cache spatial content is 0, and operation maximum wait distributes time delay when being less than the least residue execution time performing operation in virtual machine, returns " queue full " information to user; When the end-to-end time delay of operation requires to be satisfied, return " time-out " information to user.
As preferably, the present embodiment comprises further, calculates the previous operations execution time sum of each subjob queue, the job scheduling in overall job queue is arrived, and queues up in the subjob queue that previous operations execution time sum is the shortest.
Job scheduling is queued up in the shortest subjob queue of previous operations execution time sum, decreases the stand-by period of operation, improve dispatching efficiency, system resource is got the more reasonable use.
As shown in Figure 2, a kind of job scheduling method for cloud computing system of the present embodiment specifically comprises the following steps:
S1. operation arrives overall job queue;
S2. read the end-to-end time delay requirement of this operation, distribute job number, the patient maximum wait time delay of this operation is set, estimates the execution time of this operation;
S3. judged whether that subjob queue remaining cache spatial content is greater than 0, if so, then performed S4, if not, then perform S8;
S4. the previous operations execution time sum of each subjob queue is calculated;
S5. judge whether this operation meets end-to-end time delay requirement, if so, then performs S6, if not, then perform S7;
S6. this operation enters the shortest subjob queue of previous operations execution time sum, terminates this scheduling;
S7. abandon this operation and " time-out " information that sends to job request person, terminate this scheduling;
S8. judge that maximum wait is distributed time delay and whether is greater than the least residue execution time that virtual machine is performing operation: if so, then perform step S9, if not, then perform step S10;
S9., after the least residue execution time is waited in this operation, step S4 is performed;
S10. abandon this operation and " queue full " information that sends to job request person, terminate this scheduling.
Wherein, estimate in step S2 that the method for Job execution time is as follows:
If the size of operation is F, the cloud computing system resource taken is C, then the execution time of this operation is estimated as t, then t meets, and wherein cloud computing system resource C is the virtual center processor of 256,000,000 internal memories.
As preferably, the job number in described step S2 was uniquely constant before this operation completes.
Particularly, in the present embodiment, in described subjob queue remaining cache spatial content, each virtual machine, the previous operations execution time sum of current execution operation residue execution time and each subjob queue is by job scheduler monitoring calculation.
As preferably, described in the present embodiment, the length of overall job queue meets: described job scheduler completes the difference that the little operation of overall job queue required time next operation adjacent with this operation arrives the time of overall job queue.
As preferably, the queue of subjob described in the present embodiment is one or more, and its queue length meets: virtual machine completes subjob queue required time and is not more than the difference that operation next operation adjacent with this operation arrives the time of subjob queue.
Claims (10)
1. the job scheduling method for cloud computing system, the system adopted in method comprises overall job queue, subjob queue, job scheduler and the virtual machine corresponding with subjob queue, described overall job queue receives the operation of arrival system, job scheduling in overall job queue is queued up by described job scheduler to suitable subjob queue, wait virtual machine performs, described job scheduling method is based on first in first out working method, it is characterized in that, scheduling rule is: when operation arrives overall job queue, group job queue remaining cache spatial content is only had to be greater than 0, or operation maximum wait distribution time delay is less than the least residue execution time that virtual machine is performing operation, operation just can be waited for and be scheduled for subjob queue, otherwise, this operation will be dropped.
2. a kind of job scheduling method for cloud computing system according to claim 1, is characterized in that, described maximum wait is distributed time delay and is not more than the difference that operation next operation adjacent with this operation arrives the time of overall job queue.
3. a kind of job scheduling method for cloud computing system according to claim 1 and 2, it is characterized in that, comprise further, when operation arrives overall job queue, read this operation end-to-end time delay requirement, when end-to-end time delay requires to be satisfied, this operation just can be waited for and be scheduled for subjob queue, otherwise this operation will be dropped.
4. a kind of job scheduling method for cloud computing system according to claim 3, it is characterized in that, comprise further, the reason that the different situations be dropped according to operation are dropped to user's backtracking, be specially: group job queue remaining cache spatial content is 0, and operation maximum wait distributes time delay when being less than the least residue execution time performing operation in virtual machine, returns " queue full " information to user; When the end-to-end time delay of operation requires to be satisfied, return " time-out " information to user.
5. a kind of job scheduling method for cloud computing system according to claim 4, it is characterized in that, comprise further, calculate the previous operations execution time sum of each subjob queue, job scheduling in overall job queue is arrived, queues up in the subjob queue that previous operations execution time sum is the shortest.
6. a kind of job scheduling method for cloud computing system according to claim 5, is characterized in that, comprise the following steps:
S1. operation arrives overall job queue;
S2. read the end-to-end time delay requirement of this operation, distribute job number, the patient maximum wait time delay of this operation is set, estimates the execution time of this operation;
S3. judged whether that subjob queue remaining cache spatial content is greater than 0, if so, then performed S4, if not, then perform S8;
S4. the previous operations execution time sum of each subjob queue is calculated;
S5. judge whether this operation meets end-to-end time delay requirement, if so, then performs S6, if not, then perform S7;
S6. this operation enters the shortest subjob queue of previous operations execution time sum, terminates this scheduling;
S7. abandon this operation and " time-out " information that sends to job request person, terminate this scheduling;
S8. judge that maximum wait is distributed time delay and whether is greater than the least residue execution time that virtual machine is performing operation: if so, then perform step S9, if not, then perform step S10;
S9., after the least residue execution time is waited in this operation, step S4 is performed;
S10. abandon this operation and " queue full " information that sends to job request person, terminate this scheduling.
7. a kind of job scheduling method for cloud computing system according to claim 6, is characterized in that, the job number in described step S2 was uniquely constant before this operation completes.
8. a kind of job scheduling method for cloud computing system according to claim 6, it is characterized in that, in described subjob queue remaining cache spatial content, each virtual machine, the previous operations execution time sum of current execution operation residue execution time and each subjob queue is by job scheduler monitoring calculation.
9. a kind of job scheduling method for cloud computing system according to claim 6, it is characterized in that, described overall job queue length meets: described job scheduler completes overall job queue required time and is not more than the difference that operation next operation adjacent with this operation arrives the time of overall job queue.
10. a kind of job scheduling method for cloud computing system according to claim 8 or claim 9, it is characterized in that, described subjob queue is one or more, and its queue length meets: virtual machine completes subjob queue required time and is not more than the difference that operation next operation adjacent with this operation arrives the time of subjob queue.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410774633.8A CN104461722B (en) | 2014-12-16 | 2014-12-16 | A kind of job scheduling method for cloud computing system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410774633.8A CN104461722B (en) | 2014-12-16 | 2014-12-16 | A kind of job scheduling method for cloud computing system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104461722A true CN104461722A (en) | 2015-03-25 |
| CN104461722B CN104461722B (en) | 2017-11-10 |
Family
ID=52907823
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410774633.8A Active CN104461722B (en) | 2014-12-16 | 2014-12-16 | A kind of job scheduling method for cloud computing system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104461722B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105354085A (en) * | 2015-10-30 | 2016-02-24 | 广东石油化工学院 | Scheduling method for cloud workflow jobs |
| CN105389206A (en) * | 2015-11-02 | 2016-03-09 | 广东石油化工学院 | Method for rapidly configuring virtual machine resources in cloud computing data center |
| CN112152938A (en) * | 2020-08-19 | 2020-12-29 | 鹏城实验室 | Method for determining round trip delay in cloud virtual environment |
| CN113448705A (en) * | 2021-06-25 | 2021-09-28 | 皖西学院 | Unbalanced job scheduling algorithm |
| CN114741172A (en) * | 2022-04-06 | 2022-07-12 | 深圳鲲云信息科技有限公司 | Operator scheduling method, device and equipment of artificial intelligence model and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101719931A (en) * | 2009-11-27 | 2010-06-02 | 南京邮电大学 | Multi-intelligent body-based hierarchical cloud computing model construction method |
| US20100287280A1 (en) * | 2009-05-08 | 2010-11-11 | Gal Sivan | System and method for cloud computing based on multiple providers |
| 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 |
| CN103092698A (en) * | 2012-12-24 | 2013-05-08 | 中国科学院深圳先进技术研究院 | System and method of cloud computing application automatic deployment |
-
2014
- 2014-12-16 CN CN201410774633.8A patent/CN104461722B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100287280A1 (en) * | 2009-05-08 | 2010-11-11 | Gal Sivan | System and method for cloud computing based on multiple providers |
| CN101719931A (en) * | 2009-11-27 | 2010-06-02 | 南京邮电大学 | Multi-intelligent body-based hierarchical cloud computing model construction method |
| 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 |
| CN103092698A (en) * | 2012-12-24 | 2013-05-08 | 中国科学院深圳先进技术研究院 | System and method of cloud computing application automatic deployment |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105354085A (en) * | 2015-10-30 | 2016-02-24 | 广东石油化工学院 | Scheduling method for cloud workflow jobs |
| CN105354085B (en) * | 2015-10-30 | 2019-03-15 | 广东石油化工学院 | A kind of cloud workflow job scheduling method |
| CN105389206A (en) * | 2015-11-02 | 2016-03-09 | 广东石油化工学院 | Method for rapidly configuring virtual machine resources in cloud computing data center |
| CN105389206B (en) * | 2015-11-02 | 2019-03-29 | 广东石油化工学院 | A kind of cloud computation data center resources of virtual machine quickly configuration method |
| CN112152938A (en) * | 2020-08-19 | 2020-12-29 | 鹏城实验室 | Method for determining round trip delay in cloud virtual environment |
| CN113448705A (en) * | 2021-06-25 | 2021-09-28 | 皖西学院 | Unbalanced job scheduling algorithm |
| CN113448705B (en) * | 2021-06-25 | 2023-03-28 | 皖西学院 | Unbalanced job scheduling algorithm |
| CN114741172A (en) * | 2022-04-06 | 2022-07-12 | 深圳鲲云信息科技有限公司 | Operator scheduling method, device and equipment of artificial intelligence model and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104461722B (en) | 2017-11-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103324525B (en) | Method for scheduling task under a kind of cloud computing environment | |
| CN101694631B (en) | Real time job dispatching patcher and method | |
| CN104461722A (en) | Job scheduling method used for cloud computing system | |
| CN102915254B (en) | task management method and device | |
| CN103605567B (en) | Cloud computing task scheduling method facing real-time demand change | |
| CN106789726B (en) | A kind of high concurrent service resource allocation method based on service priority | |
| US8812639B2 (en) | Job managing device, job managing method and job managing program | |
| CN102902587B (en) | Distributed task dispatching mthods, systems and devices | |
| CN103365726B (en) | A kind of method for managing resource towards GPU cluster and system | |
| US20070064711A1 (en) | System and method for providing, by a plurality of schedulers, differentiated service to consumers of distributed resources | |
| CN103365711B (en) | It is applied to the task scheduling system and method for internet of things service platform | |
| CN109697122A (en) | Task processing method, equipment and computer storage medium | |
| CN105389206A (en) | Method for rapidly configuring virtual machine resources in cloud computing data center | |
| CN101923487A (en) | Comprehensive embedded type real-time period task scheduling method | |
| CN101145112A (en) | Real-time system task scheduling method | |
| CN111104211A (en) | Task dependency based computation offload method, system, device and medium | |
| CN102999377A (en) | Service concurrent access control method and device | |
| WO2015100995A1 (en) | Intelligent service scheduling method | |
| CN103294548A (en) | Distributed file system based IO (input output) request dispatching method and system | |
| CN101452519A (en) | Data scheduling method for identifying middleware by radio frequency | |
| Squillante et al. | Threshold-based priority policies for parallel-server systems with affinity scheduling | |
| CN101374109B (en) | Method and apparatus for scheduling packets | |
| Li et al. | Task scheduling algorithm for heterogeneous real-time systems based on deadline constraints | |
| CN102609307A (en) | Multi-core multi-thread dual-operating system network equipment and control method thereof | |
| CN103455375A (en) | Load-monitoring-based hybrid scheduling method under Hadoop cloud platform |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |