CN105095008B - A kind of distributed task scheduling fault redundance method suitable for group system - Google Patents
A kind of distributed task scheduling fault redundance method suitable for group system Download PDFInfo
- Publication number
- CN105095008B CN105095008B CN201510528462.5A CN201510528462A CN105095008B CN 105095008 B CN105095008 B CN 105095008B CN 201510528462 A CN201510528462 A CN 201510528462A CN 105095008 B CN105095008 B CN 105095008B
- Authority
- CN
- China
- Prior art keywords
- node
- task
- cluster
- bit stream
- management
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 18
- 238000007726 management method Methods 0.000 claims description 107
- 230000001360 synchronised effect Effects 0.000 claims description 27
- 238000012423 maintenance Methods 0.000 claims description 6
- 238000011084 recovery Methods 0.000 claims description 6
- 230000000737 periodic effect Effects 0.000 claims description 4
- 239000007858 starting material Substances 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 abstract description 3
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 238000001514 detection method Methods 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 230000007257 malfunction Effects 0.000 description 1
- 230000011664 signaling Effects 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Landscapes
- Hardware Redundancy (AREA)
Abstract
The invention discloses a kind of distributed task scheduling fault redundance method suitable for group system, there is provided a kind of two-stage task failure redundancy feature, achievees the purpose that lifting task high reliability, system high-available and user friendly.The beneficial effect that the present invention is reached：1st, task reliability improve, distributed task scheduling in the cluster operation troubles when can recover in time in node, between node, improve the reliability of cluster distributed task；2nd, the availability of system improves, and management program employs master-slave redundancy technology, and task failure Redundancy Management is transparent, the presence of user's imperceptible task failure redundancy in use for a user；3rd, it is portable good, it is not necessary to carry software by any operating system；4th, there is cross-platform ability, service routine can be deployed on different operating system servers；5th, using simple, user only needs to call several interfaces that fault redundance can be used.
Description
Technical field
The present invention relates to a kind of distributed task scheduling fault redundance method suitable for group system, belong to computer system skill
Art field.
Background technology
With flourishing for cloud computing distributed system, distributed system is just used by more and more users.
And distributed system forms cluster by substantial amounts of computer and provides a user service, with the increase of number of computers, system goes out
The probability of existing mistake greatly increases.Therefore it may cause to run user task error on the nodes, be more likely to user
Bring immeasurable loss.Therefore, distributed task scheduling fault redundance for the high reliability of distributed system, high availability with
And user friendly etc. is indispensable.
The content of the invention
To solve the deficiencies in the prior art, it is an object of the invention to provide a kind of lifting task high reliability, system are high
The distributed task scheduling fault redundance method suitable for group system of availability and user friendly, can be fast and accurately
Detect the fault condition of distributed task scheduling, and it is main syllabus that can restart the failure task in the other nodes of cluster in time
's；It is simple, transparent for secondary objective to use user.
In order to realize above-mentioned target, the present invention adopts the following technical scheme that：
A kind of distributed task scheduling fault redundance method suitable for group system, it is characterized in that, specifically include following steps：
1) fault redundance management is received by external interface：The external interface is supplied to upper strata task call, by task
Information is added in fault redundance task queue, upper strata task and then acquisition fault redundance management；
2) task failure redundancy in node：Management program in cluster on each node is responsible for safeguarding the task on this node
Information, fault redundance is carried out to being run on this node for task, and is responsible for arriving the mission bit stream synchronized update on this node
On cluster management node；
3) mission bit stream is synchronous：The mission bit stream synchronized update on this node is aggregated into cluster pipe by each node in cluster
Manage on node；
4) task failure redundancy between node：Management program on cluster management node is responsible for safeguarding the task in whole cluster
Information, fault redundance carrying out node to the task of failure, by the mission bit stream synchronized update on cluster management node to spare
Node；Between node after fault redundance success, task recovery successful information is synchronous at once from cluster management node to malfunctioning node；
5) cluster management node is elected：There are more cluster management standby nodes in cluster, when cluster management node failure
When, an available node is elected from spare management node immediately and cluster management function is externally provided, reach cluster management section
Point failure redundancy；
6) mission bit stream standby redundancy：Mission bit stream synchronized update on cluster management node is on standby node；
7) fault redundance management is exited by exiting interface：Described exit when interface is supplied to upper strata task to exit is called,
Mission bit stream is deleted out of fault redundance task queue, upper strata task and then exits fault redundance management.
A kind of foregoing distributed task scheduling fault redundance method suitable for group system, it is characterized in that, the step 2)
In, the maintenance to mission bit stream includes the renewal to mission bit stream and delete operation；By polling mode come Detection task whether
Failure, when task failure occurs, to task recovery in node, limits if recovering number and exceeding configured number,
Task failure redundancy fails in node.
A kind of foregoing distributed task scheduling fault redundance method suitable for group system, it is characterized in that, in the step
3) mission bit stream is from every node to cluster management node periodic synchronous in；After the failure of node internal fault redundancy, mission bit stream
It is synchronous at once from malfunctioning node to cluster management node.
A kind of foregoing distributed task scheduling fault redundance method suitable for group system, it is characterized in that, the step 4)
In maintenance to mission bit stream include addition, renewal and delete operation to mission bit stream；After the failure of node internal fault redundancy,
Fault redundance between progress node, cluster management node first looks at whether task has the starter node set specified, if then
The node of a node load minimum is selected to recover the task in node set is specified, if not specifying node set,
The node of a node load minimum is selected to recover the task in the cluster.
A kind of foregoing distributed task scheduling fault redundance method suitable for group system, it is characterized in that, the step 5)
Middle cluster management node periodically sends heartbeat multicast message, and standby node periodically receives the heartbeat message；When spare
After node does not receive heartbeat message more than certain time, cluster management node failure is judged, standby node is upgraded to cluster management section
Point；After a certain cluster management node receives other clustered node management node heartbeat messages, present node can be with sending heartbeat
Message node carries out IP address integer value and compares, and when the machine numerical value is big, then the machine is reduced to standby node and stops sending heartbeat report
Text, when the machine numerical value is small, continues to send heartbeat message until not receiving the heart of other cluster management nodes within a certain period of time
Message is jumped, then the node is saved as new cluster management node and by each of the cluster management node information notification to cluster
Point.
A kind of foregoing distributed task scheduling fault redundance method suitable for group system, it is characterized in that, the step 6)
Middle cluster management node periodically by mission bit stream synchronized update to standby node, and when have task status occur addition or
Exit or during the change of failure at once by job change synchronizing information renewal to standby node.
The beneficial effect that the present invention is reached：1st, the reliability of task improves, distributed task scheduling operation troubles in the cluster
When can recover in time in node, between node, improve the reliability of cluster distributed task；2nd, the availability of system carries
Height, management program employs master-slave redundancy technology, and task failure Redundancy Management is transparent for a user, and user makes
The presence of imperceptible task failure redundancy during；3rd, it is portable good, it is not necessary to be carried by any operating system soft
Part；4th, there is cross-platform ability, service routine can be deployed on different operating system servers；5th, using simple, user
Only need to call several interfaces that fault redundance can be used；6th, deployment is simple, it is only necessary to which disposing management program, dynamic base can transport
OK.
Brief description of the drawings
Fig. 1 is task state transition figure in the present invention；
Fig. 2 is communication scheme between interior joint of the present invention；
Fig. 3 is cluster management node election algorithm flow chart in the present invention；
Embodiment
The invention will be further described below in conjunction with the accompanying drawings.Following embodiments are only used for clearly illustrating the present invention
Technical solution, and be not intended to limit the protection scope of the present invention and limit the scope of the invention.
The present invention relates to a kind of distributed task scheduling fault redundance method suitable for group system, comprise the following steps：
1) fault redundance management is received by external interface：External interface is supplied to upper strata task call, by mission bit stream
It is added in fault redundance task queue, upper strata task and then acquisition fault redundance management.
2) task failure redundancy in node：Management program in cluster on each node is responsible for safeguarding the task on this node
Information, fault redundance is carried out to being run on this node for task, and is responsible for arriving the mission bit stream synchronized update on this node
On cluster management node；Maintenance to mission bit stream includes the renewal to mission bit stream and delete operation；By polling mode come
Detection task whether failure, when task failure occurs, to task recovery in node, if recovering number exceedes what is configured
Number limits, then task failure redundancy fails in node.
3) mission bit stream is synchronous：The mission bit stream synchronized update on this node is aggregated into cluster pipe by each node in cluster
Manage on node；Mission bit stream is from every node to cluster management node periodic synchronous；After the failure of node internal fault redundancy, task
Information is synchronous at once from malfunctioning node to cluster management node.
4) task failure redundancy between node：Management program on cluster management node is responsible for safeguarding the task in whole cluster
Information, fault redundance carrying out node to the task of failure, by the mission bit stream synchronized update on cluster management node to spare
Node；Between node after fault redundance success, task recovery successful information is synchronous at once from cluster management node to malfunctioning node；It is right
The maintenance of mission bit stream includes addition, renewal and delete operation to mission bit stream；After the failure of node internal fault redundancy, carry out
Fault redundance between node, cluster management node first look at whether task has the starter node set specified, if then referring to
Determine to select the node of a node load minimum to recover the task in node set, if not specifying node set, collecting
The node of a node load minimum is selected to recover the task in group.
5) cluster management node is elected：There are more cluster management standby nodes in cluster, when cluster management node failure
When, an available node is elected from spare management node immediately and cluster management function is externally provided, reach cluster management section
Point failure redundancy；Cluster management node periodically sends heartbeat multicast message, and standby node periodically receives the heartbeat report
Text；After standby node does not receive heartbeat message more than certain time, cluster management node failure is judged, standby node is upgraded to collection
Group's management node；After a certain cluster management node receives other clustered node management node heartbeat messages, present node can be with
Send heartbeat message node progress IP address integer value to compare, when the machine numerical value is big, then the machine is reduced to standby node and stops sending out
Heartbeat message is sent, when the machine numerical value is small, continues to send heartbeat message until not receiving other cluster managements within a certain period of time
The heartbeat message of node, then the node is as new cluster management node and by the cluster management node information notification to cluster
Each node.
6) mission bit stream standby redundancy：Mission bit stream synchronized update on cluster management node is on standby node；Cluster
Management node periodically by mission bit stream synchronized update to standby node, and when have task status occur addition or exit or
Job change synchronizing information is updated to standby node at once during the change of failure.
7) fault redundance management is exited by exiting interface：Described exit when interface is supplied to upper strata task to exit is called,
Mission bit stream is deleted out of fault redundance task queue, upper strata task and then exits fault redundance management.
As shown in Figure 1, after calling registration interface to succeed in registration after task start, task is normal condition；This method can week
Phase property carries out fault detect to normal tasks, and task normally then waits next cycle detection；If task is detected as failure, appoint
It is engaged in as malfunction；This method carries out node internal fault redundancy to failure task, if it is successful, task returns to normal condition,
If it fails, fault redundance is carried out then carrying out node to failure task, if it is successful, task returns to normal condition, if
Fail, then mission failure.Called after task run and exit interface, task is to exit state.
As shown in Fig. 2, when there is no management node in cluster, spare management node can send management node EB packet into
The election of row management node.After electing successfully, management node can periodically send heartbeat message, then task run node can be with
Know management node address, spare management node can know the state of management node.Task run node can be to management node
Periodic transmission task collects message, and management node can be to spare management node signalling of bouquet task to received mission bit stream
Backup message, to reach the progress cluster task information backup redundancy purpose in spare management node.When task status occurs
During change, such as：From being normally changed into failure, from normally being changed into exiting.Task run node can send task urgent messages always
To management node, stop transmission task urgent messages after the task emergency answering message of management node return is received；For pipe
Node is managed, after task urgent messages are received, spare management node can be transmitted to, when the receiving spare management node return of the task
Stopped forwarding after emergency answering message.
As shown in figure 3, after management program starts, collection can be received always in 4T (1T cluster management nodes heart beat message cycles)
Group's management node heartbeat message.If receiving message in 4T, the role of the machine is the spare management node of cluster；If in 4T
In do not receive yet, then the machine is upgraded to cluster management node, and sends heartbeat message always in 4T.If receive other cluster pipes
The heartbeat message of node is managed, then judges whether the machine will be reduced to the spare management node of cluster according to certain algorithm；Otherwise continue
Heartbeat message is sent within the time of 4T, until can not receive other cluster management nodes heart beat messages in 4T or the machine is reduced to collection
The spare management node of group.
The above is only the preferred embodiment of the present invention, it is noted that for the ordinary skill people of the art
For member, without departing from the technical principles of the invention, some improvement and deformation can also be made, these are improved and deformation
Also it should be regarded as protection scope of the present invention.
Claims (4)
1. a kind of distributed task scheduling fault redundance method suitable for group system, it is characterized in that, specifically include following steps：
1）Fault redundance management is received by external interface：The external interface is supplied to upper strata task call, by mission bit stream
It is added in fault redundance task queue, upper strata task and then acquisition fault redundance management；
2）Task failure redundancy in node：Management program in cluster on each node is responsible for safeguarding the task letter on this node
Breath, fault redundance is carried out to being run on this node for task, and is responsible for the mission bit stream synchronized update on this node to collection
In group's management node；Maintenance to mission bit stream includes the renewal to mission bit stream and delete operation；Examined by polling mode
Survey task whether failure, when task failure occurs, to task recovery in node, if recovering number exceedes time configured
Number limitation, then task failure redundancy fails in node；
3）Mission bit stream is synchronous：The mission bit stream synchronized update on this node is aggregated into cluster management section by each node in cluster
Point on；
4）Task failure redundancy between node：Management program on cluster management node is responsible for safeguarding the task letter in whole cluster
Breath, fault redundance carrying out node to the task of failure, by the mission bit stream synchronized update on cluster management node to spare section
Point；Between node after fault redundance success, task recovery successful information is synchronous at once from cluster management node to malfunctioning node, right
The maintenance of mission bit stream includes addition, renewal and delete operation to mission bit stream；After the failure of node internal fault redundancy, carry out
Fault redundance between node, cluster management node first look at whether task has the starter node set specified, if then referring to
Determine to select the node of a node load minimum to recover the task in node set, if not specifying node set, collecting
The node of a node load minimum is selected to recover the task in group；
5）Cluster management node is elected：There are more cluster management standby nodes in cluster, when cluster management node failure, stand
An available node is elected from spare management node and cluster management function is externally provided, reach cluster management node failure
Redundancy；
6）Mission bit stream standby redundancy：Mission bit stream synchronized update on cluster management node is on standby node；
7）Fault redundance management is exited by exiting interface：Described exit when interface is supplied to upper strata task to exit is called, and will be appointed
Business information is deleted out of fault redundance task queue, upper strata task and then exits fault redundance management.
2. a kind of distributed task scheduling fault redundance method suitable for group system according to claim 1, it is characterized in that,
In the step 3）Middle mission bit stream is from every node to cluster management node periodic synchronous；Node internal fault redundancy fails
Afterwards, mission bit stream is synchronous at once from malfunctioning node to cluster management node.
3. a kind of distributed task scheduling fault redundance method suitable for group system according to claim 1, it is characterized in that,
The step 5）Middle cluster management node periodically sends heartbeat multicast message, and standby node periodically receives the heartbeat report
Text；After standby node does not receive heartbeat message more than certain time, cluster management node failure is judged, standby node is upgraded to collection
Group's management node；After a certain cluster management node receives other clustered node management node heartbeat messages, present node can be with
Send heartbeat message node progress IP address integer value to compare, when the machine numerical value is big, then the machine is reduced to standby node and stops sending out
Heartbeat message is sent, when the machine numerical value is small, continues to send heartbeat message until not receiving other cluster managements within a certain period of time
The heartbeat message of node, then the node is as new cluster management node and by the cluster management node information notification to cluster
Each node.
4. a kind of distributed task scheduling fault redundance method suitable for group system according to claim 1, it is characterized in that,
The step 6）Middle cluster management node periodically by mission bit stream synchronized update to standby node, and when there is task status
Addition occurs or exits or during the change of failure at once by job change synchronizing information renewal to standby node.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510528462.5A CN105095008B (en) | 2015-08-25 | 2015-08-25 | A kind of distributed task scheduling fault redundance method suitable for group system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510528462.5A CN105095008B (en) | 2015-08-25 | 2015-08-25 | A kind of distributed task scheduling fault redundance method suitable for group system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105095008A CN105095008A (en) | 2015-11-25 |
| CN105095008B true CN105095008B (en) | 2018-04-17 |
Family
ID=54575510
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510528462.5A Active CN105095008B (en) | 2015-08-25 | 2015-08-25 | A kind of distributed task scheduling fault redundance method suitable for group system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105095008B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106789941B (en) * | 2016-11-30 | 2019-12-03 | 国电南瑞科技股份有限公司 | A kind of database and the implementation method of system application heartbeat unified management |
| CN108170375B (en) * | 2017-12-21 | 2020-12-18 | 创新科技术有限公司 | Overrun protection method and device in distributed storage system |
| CN110798339A (en) * | 2019-10-09 | 2020-02-14 | 国电南瑞科技股份有限公司 | Task disaster tolerance method based on distributed task scheduling framework |
| CN112346837A (en) * | 2020-10-28 | 2021-02-09 | 常州微亿智造科技有限公司 | Distributed timer system under industrial Internet of things |
| CN112838965B (en) * | 2021-02-19 | 2023-03-28 | 浪潮云信息技术股份公司 | Method for identifying and recovering strong synchronization role fault |
| CN113794595A (en) * | 2021-09-15 | 2021-12-14 | 领云悠逸(北京)科技有限公司 | IoT (Internet of things) equipment high-availability method based on industrial Internet |
| CN114039978B (en) * | 2022-01-06 | 2022-03-25 | 天津大学四川创新研究院 | Decentralized PoW computing power cluster deployment method |
| CN114968947B (en) * | 2022-03-01 | 2023-05-09 | 华为技术有限公司 | Fault file storage method and related device |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1719831A (en) * | 2005-07-15 | 2006-01-11 | 清华大学 | High-available distributed boundary gateway protocol system based on cluster router structure |
| 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 |
| CN103580915A (en) * | 2013-09-26 | 2014-02-12 | 东软集团股份有限公司 | Method and device for determining main control node of trunking system |
| CN104461752A (en) * | 2014-11-21 | 2015-03-25 | 浙江宇视科技有限公司 | Two-level fault-tolerant multimedia distributed task processing method |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8621463B2 (en) * | 2011-09-09 | 2013-12-31 | GM Global Technology Operations LLC | Distributed computing architecture with dynamically reconfigurable hypervisor nodes |
-
2015
- 2015-08-25 CN CN201510528462.5A patent/CN105095008B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1719831A (en) * | 2005-07-15 | 2006-01-11 | 清华大学 | High-available distributed boundary gateway protocol system based on cluster router structure |
| 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 |
| CN103580915A (en) * | 2013-09-26 | 2014-02-12 | 东软集团股份有限公司 | Method and device for determining main control node of trunking system |
| CN104461752A (en) * | 2014-11-21 | 2015-03-25 | 浙江宇视科技有限公司 | Two-level fault-tolerant multimedia distributed task processing method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105095008A (en) | 2015-11-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105095008B (en) | A kind of distributed task scheduling fault redundance method suitable for group system | |
| CN103744809B (en) | Vehicle information management system double hot standby method based on VRRP | |
| CN104506357B (en) | A kind of high-availability cluster node administration method | |
| CN101150430B (en) | A method for realizing network interface board switching based heartbeat mechanism | |
| CN108712501B (en) | Information sending method and device, computing equipment and storage medium | |
| US20150365320A1 (en) | Method and device for dynamically switching gateway of distributed resilient network interconnect | |
| CN101207517B (en) | Method for reliability maintenance of distributed enterprise service bus node | |
| CN102394914A (en) | Cluster brain-split processing method and device | |
| JP5343436B2 (en) | Information management system | |
| CN104320311A (en) | Heartbeat detection method of SCADA distribution type platform | |
| CN105630589A (en) | Distributed process scheduling system and process scheduling and execution method | |
| CN102025562A (en) | Path detection method and device | |
| CN110677282B (en) | Hot backup method of distributed system and distributed system | |
| CN104980693A (en) | Media service backup method and system | |
| CN104077181A (en) | Status consistent maintaining method applicable to distributed task management system | |
| CN101777020A (en) | Fault tolerance method and system used for distributed program | |
| WO2012012962A1 (en) | Disaster tolerance service system and disaster tolerance method | |
| CN101729426A (en) | Method and system for quickly switching between master device and standby device of virtual router redundancy protocol (VRRP) | |
| CN101610188A (en) | Sip server restoring method of service process fault and sip server | |
| CN107682169A (en) | A kind of method and apparatus using Kafka collection pocket transmission message | |
| CN105959078A (en) | Cluster time synchronization method, cluster and time synchronization system | |
| CN108197222A (en) | A kind of restorative procedure, system and the relevant apparatus of exception flow data | |
| CN104796283B (en) | A kind of method of monitoring alarm | |
| CN110753002B (en) | Traffic scheduling method and device | |
| US10205630B2 (en) | Fault tolerance method for distributed stream processing system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C41 | Transfer of patent application or patent right or utility model | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20161102 Address after: High road high tech Zone of Nanjing City, Jiangsu Province, No. 20 210032 Applicant after: NARI Technology Development Co., Ltd. Applicant after: SGCC NARI Nanjing Control System Co., Ltd. Applicant after: State Grid Corporation of China Applicant after: STATE GRID JIANGSU ELECTRIC POWER COMPANY Address before: High road high tech Zone of Nanjing City, Jiangsu Province, No. 20 210032 Applicant before: NARI Technology Development Co., Ltd. Applicant before: SGCC NARI Nanjing Control System Co., Ltd. |
|
| C41 | Transfer of patent application or patent right or utility model | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20161212 Address after: High road high tech Zone of Nanjing City, Jiangsu Province, No. 20 210032 Applicant after: NARI Technology Development Co., Ltd. Applicant after: SGCC NARI Nanjing Control System Co., Ltd. Applicant after: State Power Networks Co Applicant after: STATE GRID JIANGSU ELECTRIC POWER COMPANY Applicant after: Nanjing Nari Group Corporation Address before: High road high tech Zone of Nanjing City, Jiangsu Province, No. 20 210032 Applicant before: NARI Technology Development Co., Ltd. Applicant before: SGCC NARI Nanjing Control System Co., Ltd. Applicant before: State Power Networks Co Applicant before: STATE GRID JIANGSU ELECTRIC POWER COMPANY |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |