CN1512374A - Method for node load information transfer and node survival detection in machine group - Google Patents
Method for node load information transfer and node survival detection in machine group Download PDFInfo
- Publication number
- CN1512374A CN1512374A CNA021594724A CN02159472A CN1512374A CN 1512374 A CN1512374 A CN 1512374A CN A021594724 A CNA021594724 A CN A021594724A CN 02159472 A CN02159472 A CN 02159472A CN 1512374 A CN1512374 A CN 1512374A
- Authority
- CN
- China
- Prior art keywords
- node
- load information
- group
- host
- planes
- 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
Landscapes
- Small-Scale Networks (AREA)
Abstract
In the process of transferring node information and detecting node survival in computer group, the node information is transferred to the group leader node via the load information transferring ring; the main node controls the grouping of nodes inside the computer group and processes fault node; and the auxiliary main node replaces the faulty main node while selects new auxiliary node. The present invention collects the load information of nodes through grading and grouping to realize the load information monitoring and node survival monitoring in large scale computer group. The main node receives only the information the group leader nodes send and the load information inside the load information transferring ring is transferred only between adjacent nodes, resulting in reduced communication overhead, better expanding performance, timely collection of node load information, equalized load, no overload of node, fast finding and processing of fault and raised overall system availability.
Description
Technical field
The present invention relates to the method for node detection in a kind of group of planes, relate in particular to the method that the transmission of node load information and node survival detects in a kind of group of planes, belong to computer software fields.
Background technology
Along with development of internet technology, the demand of a commerce services group of planes constantly enlarges, the scale of a commerce services group of planes is with rapid expanding, some application can't be finished by the server of a platform independent, need bear jointly by multiple servers, and between these servers, carry out load balancing, select appropriate nodes to come the request of process user according to each node actual loading situation.
High availability is more and more important to present business service, and of short duration shutdown outside the plan all can bring tremendous loss, not only reduces transaction and brings loss to company, also can increase client's dissatisfaction, even lose the part client.Large-scale Network of Workstation is because interstitial content is many, and the probability that node breaks down in the group of planes just increases relatively, in order to guarantee the high availability of group of planes integral body, need carry out monitoring survive to all nodes in the system.When node broke down, selection adapter node also needed the load state according to each backup node, selects the lighter node of load to take over the work of malfunctioning node.
At present node in the group of planes is survived and detect the main mode sending node information that adopts broadcast/group broadcast to every other node, will increase the burden of all node machines in the group of planes, when a group of planes is larger, the performance of each node will be had a strong impact on.In addition, survival information aggregating to a node of all nodes causes this node overload even the machine of delaying easily in the group of planes.This stability to a group of planes is very unfavorable.
Summary of the invention
Fundamental purpose of the present invention is to provide at the deficiency of prior art the method that the transmission of node load information and node survival detect in a kind of group of planes, in a large-scale group of planes, transmit, collect the load information of all nodes accurately, select the lighter backup node of load to take over the work of malfunctioning node when taking place as the foundation of the selection service node of load equalizer and fault.
The object of the present invention is achieved like this:
The method that node load information is transmitted and the node survival detects in a kind of group of planes,
The load information transmission ring of nodal information in the group of planes by the node place is delivered to the group leader's node in this load information transmission ring, and this group leader's node sends the nodal information of place load information transmission ring to host node;
The load transfer grouping of the host node control group of planes interior nodes in the group of planes and the processing of malfunctioning node make the load of each node be tending towards balanced.
Auxiliary host node in the group of planes is taken over host node when host node breaks down, and will select new node as auxiliary host node.
Described load information transmission ring is at least by a node, and a load information transmission ring only is provided with group leader's node.
The transmission method of above-mentioned nodal information is:
Step 10: the group leader's node in the load information transmission ring sends load information to its downstream node;
Step 11: each ordinary node in the load information transmission ring receives the load information that its upstream node sends, and self load information is carried in the back of former load information, transmits to its downstream node again; Transmit successively, transfer back to group leader's node up to this load information;
Step 12: this group leader's node is reported the load information of all nodes in the load information transmission ring of place according to the load information of passing back to host node.
The load transfer grouping of described host node control group of planes interior nodes is specially: when new node starts, host node is assigned to this node in the corresponding load information transmission ring, will start the positional information of node in the load information transmission ring then and send to this node; Concrete step is as follows:
Step 20: host node checks whether the node number has reached the upper limit that the load information transmission link is counted out in the existing load information transmission ring when receiving the joining request of new startup node;
Step 21: if the node number in the load information transmission ring does not reach the upper limit, then this node is joined this load information transmission ring, end of packet operation then;
Step 22: if all load information transmission rings have all reached the upper limit, host node is created new load information transmission ring, and this node is made as group leader's node of this load information transmission ring, end of packet operation.
Described positional information is the information of respective nodes adjacent node, i.e. the upstream node of this node in the load information transmission ring and the information of downstream node.
The load transfer grouping of described host node control group of planes interior nodes also further comprises: when first node of a group of planes started, this node was set at host node with self.
Described malfunctioning node is handled and is comprised at least:
Step 30: when downstream node is not received the load information of upstream node in the time interval of setting, send confirmation to this upstream node;
Step 31: if this upstream node can correctly be replied, after then this upstream node was received and confirmed bag, self-verifying was also repaired all service processess and the managing process of a group of planes; If can repair, then execution in step 35;
Step 32: if can't repair, determine that this upstream node is a malfunctioning node, its downstream node is given auxiliary host node with the report information of this malfunctioning node, and then reports this malfunctioning node to host node;
Step 33: host node selects new node to take over this malfunctioning node, sends then to take over message to this selected node;
Step 34: host node sends to the node that need change with new grouping information, makes this group reconstitute the load information transmission ring;
Step 35: wait for receiving the transmission of load information next time.
Described malfunctioning node is handled and is also further comprised:
When group leader's node broke down, its downstream node reported to auxiliary host node with the failure message of this group leader's node, and then reports to host node;
Host node selects new node as new group leader's node, and new group leader's node is taken over the service of former group leader's node.
Described malfunctioning node is handled and is also further comprised:
When auxiliary host node node broke down, host node selected new node as new auxiliary host node, and new auxiliary host node is taken over the service of former auxiliary host node.
Described malfunctioning node is handled and is also further comprised:
When host node broke down, auxiliary host node was taken over host node automatically;
New host node selects new node as new auxiliary host node, and the service of taking over former auxiliary host node.
The present invention collects the load information of each node by classification, grouping, and has improved the mode of the broadcast/group broadcast that adopts when load information transmits, and has realized load information monitoring and node monitoring survive to extensive cluster nodes; Because host node only need receive the information that all group leader's nodes send, the load information transmission of each load information transmission ring interior nodes is only carried out between adjacent node, has significantly reduced the expense of communication; Therefore, compare with traditional heartbeat detecting method, the present invention has better extensibility and more performance.
Employing the present invention can in time collect the load information of each node in the group of planes, and equally loaded between node corresponding, and the overload phenomenon appears in the node of anti-locking system.
Adopt the present invention to find malfunctioning node fast, especially when a plurality of nodes broke down simultaneously, system can detect malfunctioning node simultaneously, made system's handling failure in time, improved the availability of total system.
Description of drawings
Fig. 1 is that the load information transmission system of a group of planes is formed synoptic diagram;
Fig. 2 is the processing procedure synoptic diagram during two node simultaneous faultss in the same load information transmission ring;
Fig. 3 is the processing procedure synoptic diagram during all node simultaneous faultss in the load information transmission ring.
Embodiment
In a group of planes,, need the load information of all nodes in the monitoring group of planes in order to improve the availability of system.For a large-scale group of planes,, need a kind of reliable technique to guarantee the extensibility of system because the number of node is very many.
Network of Workstation of the present invention, the grouping of a plurality of annulars is adopted in its group of planes load information transmission, to reduce the expense of network communication.
As shown in Figure 1: in the load information transmission system of a group of planes, there are four category nodes, host node P, auxiliary host node S, group leader's node L and ordinary node N, some nodes constitute a loop in logic, are called the load information transmission ring.
Host node P will gather the information of all nodes in the whole Network of Workstation, and the shielding fault that takes appropriate measures when node breaks down, and in time adjust load equalizer according to load information, make the load of each node be tending towards balanced.
Auxiliary host node S is the backup node of host node P, and the node of in time taking over the work of host node P and reformulate other when host node P breaking down is auxiliary host node S.
Group leader's node is respectively arranged in each load information transmission ring, load information is at first sent by group leader's node L, after next node receives the load information of group leader's node L, the load information of self is loaded on the back, pass to next node again, the rest may be inferred, and the information of all nodes all has been aggregated into group leader's node L place in the terminal ring, and group leader's node L reports the load information of all nodes in this group again to host node P.
All nodes except host node P and auxiliary host node S all are ordinary node N.
Thin directed line is the direction of transfer of load information in each grouping among Fig. 1, and thick dashed line is represented to send to host node P after each group information gathers.If do not adopt the mode of grouping, all nodes are directly to host node P parcel load information, when the interstitial content of a group of planes is more greatly to cause host node P overload.
The load information of all nodes is safeguarded by host node P in the group of planes, host node P is according to the service on the operating load of each node load situation distribution node or the migration overload node where necessary, so the load information of all nodes finally is to be aggregated into host node P place.Because host node P has collected the multidate information of a large amount of Network of Workstation operations, in case fault has taken place in host node P, can influence the normal operation of whole Network of Workstation, so system is provided with auxiliary host node S, all send to the group of planes multidate information of host node P, as the load information of each node and malfunctioning node information etc., sending to host node P after all sending each auxiliary host node S earlier, it is the working mechanism that example illustrates the heartbeat ring that Fig. 1 is aggregated into host node P with the load information of each node, and the flow process that sends to auxiliary host node S is identical therewith.
The grouping of group of planes load transfer is controlled by host node P, when first node of a group of planes starts, owing to also do not have host node P in the Network of Workstation, therefore will self be upgraded to host node P.When other nodes start, add a group of planes to the host node P request of sending, host node P is after confirming that this node can add a group of planes, this node is assigned in corresponding group, then this is started the position of node in grouping and send to this node, positional information is the adjacent nodal information of this node, i.e. the upstream node of this node in the heartbeat ring and the information of downstream node.First node of every group is group leader's node L, and host node P and other service nodes also constitute a heartbeat ring.The load information transmission system of service node can be made of a plurality of rings, the information that each node is preserved a upstream and downstream node in the ring.
The big I of each grouping is adjusted voluntarily by the user, be generally 8 nodes, host node P is when receiving the joining request an of node, check whether node number in the existing grouping has reached the upper limit of packet node number, if grouping also has remaining sum, this node is added this grouping, if all groupings have all reached the upper limit, host node P creates new grouping, and this node is made as group leader's node L of this grouping.When packet node broke down, host node P reselected the group leader node L of a node as this grouping in this grouping, if be somebody's turn to do when dividing into groups not had the node of operation, host node P will cancel this grouping.
When the node in the Network of Workstation broke down, its adjacent node can detect found fault rapidly, and Fig. 2 has illustrated when two nodes break down simultaneously, the treatment scheme of system.Individual node fault or more node simultaneous faults treatment scheme are similarly.
Treatment step is as follows:
1, when node breaks down, this node can't send load information to its adjacent node.
2, adjacent node is not received the load information of front node in the time interval of setting; therefore think that the node of front may locate fault; so; send confirmation to this node; if this node can correctly be replied; do not receive that so packets of information just may be because unusual or other reasons appear in load information transmission process; after " abnormal nodes " received and confirmed bag; self-verifying is also repaired all service processess and the managing process of a group of planes; at this moment, should " abnormal nodes " can detect the another one abnormal nodes and take same method to send the affirmation bag.
If the system of 3 " abnormal nodes " is unavailable or can't repair, this node just is identified as malfunctioning node, adjacent node is given auxiliary host node S with the report information of malfunctioning node, if when host node P delays machine with assurance this moment, auxiliary host node S has the most complete up-to-date group of planes information.
4, adjacent node is carried out next step processing to the host node P of group of planes report malfunctioning node by host node P.
5, after host node P selects appropriate nodes taking over fault node, send and take over message to selected node.When the node report fault, there is a more than node to break down in very possible this group, in order to handle the situation that the large tracts of land malfunctioning node occurs, host node P sends the affirmation bag to all nodes of malfunctioning node place grouping, and what can not correctly reply promptly is the fault node.Occurred two malfunctioning nodes in this example simultaneously, first malfunctioning node has adjacent node to detect and reports to host node P, and the second failure node is detected and taked measures such as taking over fault node service in time to shield malfunctioning node by host node P.After malfunctioning node is identified, host node P will send to new grouping information the node that need change, and make this group reconstitute the load information transmission path of annular.Fine dotted line among Fig. 2 is revises the new loop that the posterior nodal point load information transmits.
When group leader's node L of each group breaks down, use above mechanism also can correctly detect fault.When group leader's node L of each group breaks down, adjacent node can't be received the load information that it sends, adjacent node can be given auxiliary host node S with the report information of malfunctioning node, and then report to host node P, host node P will select appropriate nodes as new group leader's node L, and the service of taking over former group leader's node L.
As shown in Figure 3: when all nodes all fault have taken place simultaneously in certain group in the group of planes, need to adopt following mechanism to monitor so that system can in time find.
Because the load information of all nodes is aggregated into group leader's node L place, and sends to host node P by group leader's node L.The node that occurs whole group in a group of planes all breaks down simultaneously, during as indivedual switch power down of system or indivedual rack powered-off fault, just the whole group node failure might appear, in order in time to detect malfunctioning node, when host node P does not receive the information of certain group that group leader's node L sends in official hour, host node P will send from all nodes that trend should be organized and confirm bag, and the node that can not correctly reply just is considered as malfunctioning node.
It should be noted that at last: above embodiment is the unrestricted technical scheme of the present invention in order to explanation only, although the present invention is had been described in detail with reference to the foregoing description, those of ordinary skill in the art is to be understood that: still can make amendment or be equal to replacement the present invention, and not breaking away from any modification or partial replacement of the spirit and scope of the present invention, it all should be encompassed in the middle of the claim scope of the present invention.
Claims (10)
1, the method that node load information is transmitted and the node survival detects in a kind of group of planes is characterized in that:
The load information transmission ring of nodal information in the group of planes by the node place is delivered to the group leader's node in this load information transmission ring, and this group leader's node sends the nodal information of place load information transmission ring to host node;
The load transfer grouping of the host node control group of planes interior nodes in the group of planes and the processing of malfunctioning node make the load of each node be tending towards balanced;
Auxiliary host node in the group of planes is taken over host node when host node breaks down, and will select new node as auxiliary host node.
2, the method that node load information is transmitted and the node survival detects in the group of planes according to claim 1, it is characterized in that: described load information transmission ring is at least by a node, and a load information transmission ring only is provided with group leader's node.
3, the method that node load information is transmitted and the node survival detects in the group of planes according to claim 1, it is characterized in that: the transmission method of described nodal information is:
Step 10: the group leader's node in the load information transmission ring sends load information to its downstream node;
Step 11: each ordinary node in the load information transmission ring receives the load information that its upstream node sends, and self load information is carried in the back of former load information, transmits to its downstream node again; Transmit successively, transfer back to group leader's node up to this load information;
Step 12: this group leader's node is reported the load information of all nodes in the load information transmission ring of place according to the load information of passing back to host node.
4, the method that node load information is transmitted and the node survival detects in the group of planes according to claim 1, it is characterized in that: the load transfer grouping of described host node control group of planes interior nodes is specially: when new node starts, host node is assigned to this node in the corresponding load information transmission ring, will start the positional information of node in the load information transmission ring then and send to this node; Concrete step is as follows:
Step 20: host node checks whether the node number has reached the upper limit that the load information transmission link is counted out in the existing load information transmission ring when receiving the joining request an of node;
Step 21: if the node number in the load information transmission ring does not reach the upper limit, then this node is joined this load information transmission ring, end of packet operation then;
Step 22: if all load information transmission rings have all reached the upper limit, host node is created new load information transmission ring, and this node is made as group leader's node of this load information transmission ring, end of packet operation.
5, the method that node load information is transmitted and the node survival detects in the group of planes according to claim 4, it is characterized in that: described positional information is the information of respective nodes adjacent node, i.e. the upstream node of this node in the load information transmission ring and the information of downstream node.
6, the method for surviving and detecting according to transmission of node load information and node in a claim 1 or the 4 described group of planes, it is characterized in that: the load transfer grouping of described host node control group of planes interior nodes also further comprises: when first node of a group of planes started, this node was set at host node with self.
7, the method that node load information is transmitted and the node survival detects in the group of planes according to claim 1 is characterized in that: described malfunctioning node is handled and is comprised at least:
Step 30: when downstream node is not received the load information of upstream node in the time interval of setting, send confirmation to this upstream node;
Step 31: if this upstream node can correctly be replied, after then this upstream node was received and confirmed bag, self-verifying was also repaired all service processess and the managing process of a group of planes; If can repair, then execution in step 35;
Step 32: if can't repair, determine that this upstream node is a malfunctioning node, its downstream node is given auxiliary host node with the report information of this malfunctioning node, and then reports this malfunctioning node to host node;
Step 33: host node selects new node to take over this malfunctioning node, sends then to take over message to this selected node;
Step 34: host node sends to the node that need change with new grouping information, makes this group reconstitute the load information transmission ring;
Step 35: wait for receiving the transmission of load information next time.
8, the method that node load information is transmitted and the node survival detects in the group of planes according to claim 7 is characterized in that: described malfunctioning node is handled and is also further comprised:
When group leader's node broke down, its downstream node reported to auxiliary host node with the failure message of this group leader's node, and then reports to host node;
Host node selects new node as new group leader's node, and new group leader's node is taken over the service of former group leader's node.
9, the method that node load information is transmitted and the node survival detects in the group of planes according to claim 7 is characterized in that: described malfunctioning node is handled and is also further comprised:
When auxiliary host node broke down, host node selected new node as new auxiliary host node, and new auxiliary host node is taken over the service of former auxiliary host node.
10, the method that node load information is transmitted and the node survival detects in the group of planes according to claim 7 is characterized in that: described malfunctioning node is handled and is also further comprised:
When host node broke down, auxiliary host node was taken over host node automatically;
New host node selects new node as new auxiliary host node, and the service of taking over former auxiliary host node.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 02159472 CN1251110C (en) | 2002-12-31 | 2002-12-31 | Method for node load information transfer and node survival detection in machine group |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 02159472 CN1251110C (en) | 2002-12-31 | 2002-12-31 | Method for node load information transfer and node survival detection in machine group |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1512374A true CN1512374A (en) | 2004-07-14 |
| CN1251110C CN1251110C (en) | 2006-04-12 |
Family
ID=34237486
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 02159472 Expired - Fee Related CN1251110C (en) | 2002-12-31 | 2002-12-31 | Method for node load information transfer and node survival detection in machine group |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1251110C (en) |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101488966A (en) * | 2009-01-14 | 2009-07-22 | 深圳市同洲电子股份有限公司 | Video service system |
| CN101452406B (en) * | 2008-12-23 | 2011-05-18 | 北京航空航天大学 | Cluster load balance method transparent for operating system |
| CN102292963A (en) * | 2009-01-22 | 2011-12-21 | 瑞典爱立信有限公司 | Address allocation in a network |
| CN103298012A (en) * | 2013-06-09 | 2013-09-11 | 杭州华三通信技术有限公司 | AP failure detection method and device |
| CN103491018A (en) * | 2013-09-04 | 2014-01-01 | 华中科技大学 | Load balancing method in large-scale behavior modeling application |
| CN103533567A (en) * | 2012-09-29 | 2014-01-22 | 深圳市友讯达科技发展有限公司 | Determination method of standby node, and nodes |
| CN103581971A (en) * | 2012-07-23 | 2014-02-12 | 中兴通讯股份有限公司 | Method, device and access points for detecting abnormity of centralized wireless local area network |
| CN104852857A (en) * | 2014-02-14 | 2015-08-19 | 航天信息股份有限公司 | Distributed data transmission method and system based on load balancing |
| CN105207818A (en) * | 2015-09-23 | 2015-12-30 | 杭州数梦工场科技有限公司 | Node control method and device |
| CN105306545A (en) * | 2015-09-28 | 2016-02-03 | 浪潮(北京)电子信息产业有限公司 | Failover method and system for external service node of cluster |
| CN105721233A (en) * | 2014-12-03 | 2016-06-29 | 北京奇虎科技有限公司 | Website survival detection method, apparatus and system |
| CN107743618A (en) * | 2015-06-24 | 2018-02-27 | 英特尔公司 | Technology for data center environment checkpointing |
| CN108989082A (en) * | 2018-06-04 | 2018-12-11 | 曙光信息产业(北京)有限公司 | The reliability estimation method and assessment device of High Performance Computing Cluster system |
| CN109391507A (en) * | 2017-08-02 | 2019-02-26 | 南宁富桂精密工业有限公司 | Network collocating method, computer readable storage medium and network equipment |
| CN110430141A (en) * | 2019-08-08 | 2019-11-08 | 北京字节跳动网络技术有限公司 | Current-limiting method and device |
-
2002
- 2002-12-31 CN CN 02159472 patent/CN1251110C/en not_active Expired - Fee Related
Cited By (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101452406B (en) * | 2008-12-23 | 2011-05-18 | 北京航空航天大学 | Cluster load balance method transparent for operating system |
| CN101488966A (en) * | 2009-01-14 | 2009-07-22 | 深圳市同洲电子股份有限公司 | Video service system |
| CN102292963A (en) * | 2009-01-22 | 2011-12-21 | 瑞典爱立信有限公司 | Address allocation in a network |
| US9577870B2 (en) | 2009-01-22 | 2017-02-21 | Telefonaktiebolaget Lm Ericsson (Publ) | Address allocation in a network |
| CN103581971A (en) * | 2012-07-23 | 2014-02-12 | 中兴通讯股份有限公司 | Method, device and access points for detecting abnormity of centralized wireless local area network |
| CN103533567B (en) * | 2012-09-29 | 2016-08-03 | 深圳友讯达科技股份有限公司 | The determination method of secondary node and node |
| CN103533567A (en) * | 2012-09-29 | 2014-01-22 | 深圳市友讯达科技发展有限公司 | Determination method of standby node, and nodes |
| CN103298012B (en) * | 2013-06-09 | 2016-05-04 | 杭州华三通信技术有限公司 | A kind of AP fault detection method and equipment |
| CN103298012A (en) * | 2013-06-09 | 2013-09-11 | 杭州华三通信技术有限公司 | AP failure detection method and device |
| CN103491018A (en) * | 2013-09-04 | 2014-01-01 | 华中科技大学 | Load balancing method in large-scale behavior modeling application |
| CN104852857A (en) * | 2014-02-14 | 2015-08-19 | 航天信息股份有限公司 | Distributed data transmission method and system based on load balancing |
| CN105721233B (en) * | 2014-12-03 | 2020-10-27 | 北京奇虎科技有限公司 | Website survival detection method, device and system |
| CN105721233A (en) * | 2014-12-03 | 2016-06-29 | 北京奇虎科技有限公司 | Website survival detection method, apparatus and system |
| CN107743618A (en) * | 2015-06-24 | 2018-02-27 | 英特尔公司 | Technology for data center environment checkpointing |
| CN105207818A (en) * | 2015-09-23 | 2015-12-30 | 杭州数梦工场科技有限公司 | Node control method and device |
| CN105306545B (en) * | 2015-09-28 | 2018-09-07 | 浪潮(北京)电子信息产业有限公司 | A kind of method and system of the external service node Takeover of cluster |
| CN105306545A (en) * | 2015-09-28 | 2016-02-03 | 浪潮(北京)电子信息产业有限公司 | Failover method and system for external service node of cluster |
| CN109391507A (en) * | 2017-08-02 | 2019-02-26 | 南宁富桂精密工业有限公司 | Network collocating method, computer readable storage medium and network equipment |
| CN109391507B (en) * | 2017-08-02 | 2021-11-02 | 南宁富桂精密工业有限公司 | Network configuration method, computer readable storage medium and network device |
| CN108989082A (en) * | 2018-06-04 | 2018-12-11 | 曙光信息产业(北京)有限公司 | The reliability estimation method and assessment device of High Performance Computing Cluster system |
| CN110430141A (en) * | 2019-08-08 | 2019-11-08 | 北京字节跳动网络技术有限公司 | Current-limiting method and device |
| CN110430141B (en) * | 2019-08-08 | 2022-08-09 | 北京字节跳动网络技术有限公司 | Current limiting method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1251110C (en) | 2006-04-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1251110C (en) | Method for node load information transfer and node survival detection in machine group | |
| US8885658B2 (en) | Distributed router application serialization | |
| US20080189352A1 (en) | Complex event processing system having multiple redundant event processing engines | |
| CN1503513A (en) | Syste mand method for detecting lost messages transmitted between modules in a communication device | |
| CN1869932A (en) | Computer processing system for implementing data update and data update method | |
| CN1317658C (en) | Fault-tolerance approach using machine group node interacting buckup | |
| CN1805408A (en) | Quick route switching method and apparatus for network node devices | |
| CN1533095A (en) | Agent responding device, | |
| CN1531262A (en) | Network communication method for carrying out load division | |
| CN1842020A (en) | Method for realizing data service backup | |
| CN101360005B (en) | Intelligent failover in a load-balanced network environment | |
| CN1859219A (en) | Service take-over method based on device disaster tolerance, service switching device and backup machine | |
| CN101571813A (en) | Master/slave scheduling method in multimachine assembly | |
| CN101079795A (en) | Data forwarding method and forwarding device | |
| US8799382B2 (en) | Multi-network multi-plane structure short message centre system and implementation method thereof | |
| CN1992707A (en) | Fast restoration method of multicast service and network apparatus | |
| CN1259786C (en) | Method for detecting problem in route protocol module of distribution router | |
| CN1941683A (en) | Method and system for re-distributing triggered link aggregate service | |
| CN106385334A (en) | Call-center system and abnormality detection and self-recovery method therefor | |
| CN112346925A (en) | Process-level dual-computer hot standby redundancy system and method | |
| CN101056254A (en) | An expansion method, system and device of network storage device | |
| CN1127242C (en) | On-line message accounting device and method for mobile communication network | |
| CN1750419A (en) | Method and device for realizing single board backup in communication network device | |
| CN110557369A (en) | high-speed data processing platform based on domestic operating system kernel mode | |
| CN1728609A (en) | Method for detecting connection mode of devices with multiple frames |
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 |
Granted publication date: 20060412 Termination date: 20201231 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |