CN101374152B - Whole exchange storage control system and management method thereof as well as service node - Google Patents

Whole exchange storage control system and management method thereof as well as service node Download PDF

Info

Publication number
CN101374152B
CN101374152B CN200710120636XA CN200710120636A CN101374152B CN 101374152 B CN101374152 B CN 101374152B CN 200710120636X A CN200710120636X A CN 200710120636XA CN 200710120636 A CN200710120636 A CN 200710120636A CN 101374152 B CN101374152 B CN 101374152B
Authority
CN
China
Prior art keywords
node
service node
management
service
signal
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.)
Expired - Fee Related
Application number
CN200710120636XA
Other languages
Chinese (zh)
Other versions
CN101374152A (en
Inventor
安辉
赖守锋
刘介良
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
New H3C Technologies Co Ltd
Original Assignee
Hangzhou H3C Technologies Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Hangzhou H3C Technologies Co Ltd filed Critical Hangzhou H3C Technologies Co Ltd
Priority to CN200710120636XA priority Critical patent/CN101374152B/en
Publication of CN101374152A publication Critical patent/CN101374152A/en
Application granted granted Critical
Publication of CN101374152B publication Critical patent/CN101374152B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Landscapes

  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

The invention discloses a management method for a full-interchange storage control system, a service node and a full-interchange storage control system. The management method comprises the following steps: one arbitrary service node in the full-interchange storage control system negotiates with other nodes so as to determine the service node as a management node, and the management operation of the full-interchange storage control system is carried out by the service node as the management node. The service node comprises a negotiating unit, a management processing unit and a service processing unit. The management method improves the reliability of the full-interchange storage control system.

Description

Whole exchange storage control system and its management method and a kind of service node
Technical field
The present invention relates to Network storage technology, particularly relate to whole exchange storage control system and its management method and a kind of service node.
Background technology
Because the quick increase of information data amount and the requirements at the higher level of reliability at present, the full switching network storage control system between applied host machine and the disk array occurred being coupled in.Fig. 1 is the structural representation of whole exchange storage control system in the prior art.Referring to Fig. 1, comprise management node and a plurality of service node in the present whole exchange storage control system, and preferably, as shown in Figure 1, management node can comprise a plurality of management nodes that can back up mutually.
Referring to Fig. 1, in present full switching network storage system, management node is not managed business, and only be in charge of the management, as the configuration information that issues whole exchange storage control system gives each service node, and whether the service node of hot plug etc. is arranged in the detection system, and management node can also be born the task of handling request message certainly, such as be write request distribution spatial cache, even be included as tasks such as whether read request inquiry Cache hits; The business datum between processing and other service nodes only is responsible in the then not responsible management work of each service node.Wherein, the communication between each service node need be transmitted by management node, such as needs to send mutually message such as some read-write requests, in addition as previously mentioned in the part system read-write requests message of service node also need management node to handle.
By above description as can be seen, in the prior art, whole exchange storage control system inside must be provided with special-purpose management node and finish all management work in the whole exchange storage control system, perhaps also bears the processing of request message in the part system.Like this, after the management node single-point of special use broke down, the service node of One's name is legion then can't be communicated by letter normally in the whole exchange storage control system, thereby greatly reduced the reliability of whole exchange storage control system.
Summary of the invention
In view of this, one object of the present invention is to provide a kind of management method of whole exchange storage control system, another purpose of the present invention is to provide a kind of service node, a further object of the present invention is to provide a kind of whole exchange storage control system, so that improve the reliability of whole exchange storage control system.
In order to achieve the above object, technical scheme of the present invention is achieved in that
A kind of management method of whole exchange storage control system, described whole exchange storage control system are between applied host machine and disk array, and in order to handle the read-write operation that applied host machine is initiated to disk array, this method comprises:
Any one service node in the whole exchange storage control system and other service nodes are consulted, and determine the service node as management node, carry out the management work of whole exchange storage control system as the service node of management node.
A kind of service node is used for whole exchange storage control system, and wherein this whole exchange storage control system is between applied host machine and disk array, and in order to handle the read-write operation that applied host machine is initiated to disk array, this service node comprises:
Negotiation element is used for holding consultation with outside other service nodes, when negotiation result is the service node at described negotiation element place when being management node, management node role notice is sent to the management processing unit;
The management processing unit is used for carrying out the management work of whole exchange storage control system after receiving management node role notice;
Service Processing Unit is used for carrying out data interaction with outside other service nodes.
This shows, in the present invention, need not proprietary management node, any one service node all can be through consultation as management node in the whole exchange storage control system, carry out the management work of whole exchange storage control system, like this, even have only a service node to exist in the whole exchange storage control system, this service node also can be simultaneously as management node, the regular traffic that guarantees whole exchange storage control system is handled and management work, therefore, improved the reliability of whole exchange storage control system greatly, and the flexibility of management work realization.
Description of drawings
Fig. 1 is the structural representation of whole exchange storage control system in the prior art.
Fig. 2 is the structural representation of the whole exchange storage control system that proposes in one embodiment of the invention.
Fig. 3 is the flow chart of electing management node in one embodiment of the invention from service node.
Fig. 4 is the flow chart that issues configuration management information in one embodiment of the invention as the service node of management node.
Fig. 5 is according to the signal monitoring that produces as the service node of management node and the flow chart of handover management node in a preferred embodiment of the present invention.
Fig. 6 is that hot drawing goes out process chart in one embodiment of the invention.
Fig. 7 is that heat is inserted process chart in one embodiment of the invention.
Fig. 8 is the structural representation of message packet format in one embodiment of the invention.
Fig. 9 is the structural representation of the service node that proposes in one embodiment of the invention.
Figure 10 is the structural representation of management processing unit in one embodiment of the invention.
Figure 11 is the concrete structure schematic diagram of signal controlling subelement in one embodiment of the invention.
Figure 12 realizes the processing of signal controlling subelement and a kind of particular circuit configurations figure that Service Processing Unit detects the Locked signal processing in one embodiment of the invention.
Embodiment
For making the purpose, technical solutions and advantages of the present invention clearer, the present invention is described in further detail below in conjunction with drawings and the specific embodiments.
The present invention has at first proposed a kind of whole exchange storage control system.Fig. 2 is the structural representation of the whole exchange storage control system that proposes in one embodiment of the invention.Referring to Fig. 2, in one embodiment of the invention, whole exchange storage control system includes only a plurality of service nodes, and no longer comprise proprietary management node, and, any one service node, be used for consulting with other service nodes, determine service node as management node, when self during as management node, carry out the management work of whole exchange storage control system.Those skilled in the art know that application of the present invention is not limited to the system architecture of Fig. 2.
Wherein, the mode that each service node is held consultation is various, such as, can adopt precedence information such as slot number etc. with self, send to other service nodes, and with the service node of priority the highest (minimum or maximum) as slot number as management node; For another example, self disposal ability information is sent to other service nodes, and the service node that disposal ability is the highest is as management node.
Then describe the management process of the whole exchange storage control system of the present invention's proposition below in detail in conjunction with each concrete business realizing flow process.
Fig. 3 is the flow chart of electing management node in one embodiment of the invention from service node.Referring to Fig. 3, in one embodiment of the invention, when the whole exchange storage control system initial start, be example to determine management node by the priority negotiation mode, the process of electing management node from each service node specifically may further comprise the steps:
Step 301: after each service node in the whole exchange storage control system starts, the broadcast packet that carries self precedence information is sent to other service nodes.
For ease of describing, in the following steps with the processing that example illustrates that each service node is performed that is treated to of service node in the whole exchange storage control system 1.
Step 302: service node 1 receives the broadcast packet that carries precedence information that other service nodes are sent.
Here, service node 1 can also be safeguarded a priority query according to precedence information in each broadcast packet and the priority of self in self.
Step 303: service node 1 is the priority of self and the precedence information in each broadcast packet of receiving relatively, and whether the priority of judging self for the highest, if then execution in step 304, otherwise, execution in step 305.
Step 304: service node 1 self-role is set to service node and adds management node, finishes current flow process.
Step 305: service node 1 self-role is set to service node.
Like this, after each service node in whole exchange storage control system is all carried out the processing of above-mentioned flow process shown in Figure 3, then can from each service node, put forward management node.
In above-mentioned flow process shown in Figure 3, precedence information can be the slot number of service node, the little then priority of slot number height, and slot number greatly then priority is low.
After the service node that puts forward as management node, this then needs to carry out the management work of whole exchange storage control system as the service node of management node, such as, issue configuration information such as ID and management information to other all service nodes, and need to characterize self whether signal of operate as normal to other service nodes output.
Fig. 4 is the flow chart that issues configuration management information in one embodiment of the invention as the service node of management node.Referring to Fig. 4, this flow process specifically may further comprise the steps:
Step 401: obtain the essential information of service node in the whole exchange storage control system as the service node of management node, information such as the number of the service node that inserts such as current system and slot number thereof.
Step 402: judge the configuration mode of the whole exchange storage control system that sets in advance as the service node of management node, if the dynamic-configuration mode, then execution in step 403, if the static configuration mode, then execution in step 404.
Here, if static configuration mode, so, the sign of each service node (ID) has been pre-configured in each service node in the whole exchange storage control system, therefore, then need not again promptly can directly to carry out the processing of subsequent step 404 as the service node of management node for each service node distributes ID, if the dynamic-configuration mode then also needs to carry out the processing of the distribution ID in the subsequent step 403.
Step 403: as the service node of management node is that other all service nodes distribute ID, and with the ID that distributes with after the slot number of service node is related, by sideband the ID of distribution is sent to the corresponding service node respectively.
Step 404: the service node as management node is handed down to other all service nodes by in-band channel with management information.
Step 405: according to the information that sideband or in-band channel receive self is not configured as other all service nodes of management node.
Behind the service node that puts forward as management node, this service node as management node also needs to produce in real time and characterizes self whether signal of operate as normal, so that when it can't operate as normal, other service nodes can in time consult to determine a new management node.Fig. 5 is according to the signal monitoring that produces as the service node of management node and the flow chart of handover management node in a preferred embodiment of the present invention.Referring to Fig. 5, this flow process specifically may further comprise the steps:
Step 501: when its operate as normal, produce the Locked signal of impulse form as the service node of management node in real time, the Locked signal of this impulse form is carried out non-locking processing back output.
Working condition as the service node of management node in the whole exchange storage control system is analyzed as can be known, and its working condition comprises: this service node does not exist, and has promptly withdrawed from whole exchange storage control system; Or this service node exists but because fault can't operate as normal; This service node operate as normal.Wherein, when the service node as management node does not exist or this service node exists but because fault can't operate as normal the time, it all may externally export high level signal or low level signal, therefore, in order to be distinguished, the signal of being exported when the operate as normal as the service node of management node then should not use fixing high level or low level signal.And because the signal of impulse form has certain frequency and high-low level changes, could produce when having only the node operate as normal, therefore, in this step, the Locked signal that has adopted impulse form is as the signal that characterizes the node operate as normal.
Further, in this step, current service node as management node carries out non-locking processing to the Locked signal, be in order to guarantee after it breaks down, can be to the Locked semaphore lock, the role who makes other service nodes can take over management node continues to produce and export the Locked signal of impulse form.
Step 502: whether ineffective as the service node detection Locked signal of management node, if, then return step 502, otherwise, execution in step 503.
In whole exchange storage control system, can in advance all service nodes be linked to each other by sharing holding wire, like this, in above-mentioned steps 501, the Locked signal of the impulse form of its generation can be exported on the shared holding wire of whole exchange storage control system as the service node of management node, in this step 502, the service node as management node does not obtain the Locked signal by driver from sharing holding wire, and detects.
In addition, whether effective method is for detecting whether the Locked signal is pulse signal to detect the Locked signal, if pulse signal, then the Locked signal is effective, thereby can determine current service node operate as normal as management node, if not pulse signal, Locked invalidating signal then, thus can determine that current service node as management node can't operate as normal.
Step 503: the not priority of other each service nodes that from broadcast packet, obtain during according to the whole exchange storage control system initial start as each service node of management node and the height of the priority of self, determine whether self can be used as management node, if, then execution in step 505, otherwise, execution in step 504.
Step 504: this service node self-role is set to service node, with other service node interaction datas, finishes current flow process.
Step 505: this service node self-role is set to service node and adds management node.
Step 506: this service node produces the Locked signal of impulse form, carries out non-locking processing back output.
Step 507: the former management information that this service node deletion is preserved.
Step 508: this service node upgrades the priority query that safeguards in self.
Here, renewal process is: before the deletion as the priority of the service node of management node, the whole reach of priority query.
Step 509: this service node produces new management information, and is handed down to other all service nodes in the whole exchange storage control system respectively.
The implementation procedure of this step is identical with the process of above-mentioned steps 404.
Step 510: this service node is the service node of fault, reports by alarm notification as the slot number and the id information of the service node of management node before promptly.
Step 511: this service node and other service node interaction datas.
In the business realizing of reality, any one service node in the whole exchange storage control system all might be pulled out, and causes this service node to withdraw from whole exchange storage control system.For this kind hot drawing artificial situation, concrete processing procedure may further comprise the steps as shown in Figure 6:
Step 601: the service node as management node is received the request of extracting.
Here, the request of extracting can be that other one produces and send to the service node as management node will not be pulled out as the service node of management node the time, also can be the own generation will be pulled out the time of service node as management node.
Step 602: judge that as the service node of management node the request of extracting is to be produced by self or other service nodes generations, if self produces, then execution in step 603, if other service nodes produce execution in step 608.
Step 603: this service node is judged the current service node that can take over that whether exists, if then execution in step 605, otherwise, execution in step 604.
Step 604: stop to drive the Locked signal, node is on the throne etc., and information setting is invalid, extracts as the service node of management node, finishes current flow process.
Step 605: according to the priority query that safeguards in self, from other all service nodes, select one as the service node of management node, the management node notice is sent to selected service node as management node.
Selected service node returns and prepares to take over response after receiving the management node notice, prepares self to be set to new management node.
Step 606: judge before adapter is overtime, whether to receive the preparation adapter response that selected service node is sent as the service node of management node, if then execution in step 607, otherwise, return step 603.
Here, can comprise management information after the current renewal that issues as the service node of management node in the described adapter response.
Step 607: the service node as management node is extracted, and finishes current flow process.
Step 608: the service node as management node upgrades the priority query that safeguards in self.
Here, renewal process is: the priority of the service node of the request of extracting is sent in deletion from priority query, the whole reach of priority query.
Step 609: the information that will send the service node of the request of extracting as the service node of management node is deleted from management information, and the management information after will upgrading is handed down to other all service nodes in the whole exchange storage control system respectively.
In the business realizing of reality, might there be a new service node need add whole exchange storage control system at any time.For the hot insertion situation of this kind, concrete processing procedure may further comprise the steps as shown in Figure 7:
Step 701: when a service node, when being inserted into as service node 0, service node 0 each service node in whole exchange storage control system sends the request of insertion.
Step 702: after receiving the request of insertion as the service node of management node, will be sent to service node 0 for the configuration information that service node 0 distributes.
Here, if be the dynamic-configuration mode, then be included as the ID that this service node 0 distributes in the configuration information.
Step 703: service node 0 carries out initialization procedure according to the configuration information that receives.
Step 704: judged before the initialization time-out time arrives as the service node of management node whether receive initialization that service node 0 the sends notice that finishes, if then execution in step 706, otherwise, execution in step 705.
Step 705: will cancel the adding supervisory packet as the service node of management node and be sent to service node 0, the adding of refusal service node 0 finishes current flow process.
Step 706: the service node as management node joins the last of priority query with service node 0.
Step 707: the service node as management node produces new management information, and is handed down to other all service nodes in the whole exchange storage control system respectively.
In the whole exchange storage control system that the present invention proposes, when issuing configuration information and management information, can adopt self-defining message packet format to carry configuration information and management information as the service node of management node.Wherein, a kind of preferred construction of message packet format can comprise command code (OPCODE) field, type (TYPE) field, source node sign (SRC.ID) field, destination node sign (TAR.ID) field, length (LENGTH) field and a plurality of data (DATA) field referring to shown in Figure 8.Wherein, the OPCODE field is used to represent the concrete operations that need carry out, the corresponding concrete command code of each concrete operations, and its corresponding relation can be referring to as shown in table 1 below; The TYPE field is used to represent the type of the concrete operations that need carry out, the corresponding concrete action type sign indicating number of each action type, and its corresponding relation can be referring to as shown in table 2 below; The SRC.ID field is used to represent to send the source service node of this message bag; The TAR.ID field is used to represent to receive the purpose service node of this message bag; The LENGTH field is used to represent the length of this message bag; It is management information content perhaps in the concrete configuration information that each DATA field is used to carry concrete message bag content.Like this, when need be when other service nodes issue configuration information or management information, the various information of this configuration information or management information correspondence be carried at respectively in each field of message bag as the service node of management node.
Command code OPCODE Implication
?00001 Insert Node
?00010 Extract Node
?00011 Configuration Node
?00100 Management Node is used to issue configuration management information
?00101 Take over notice
?00110 System broadcasts
Other Keep
Table 1
Bag type TYPE Implication
?001 Bag request REQUEST
?010 Bag response RESPONSE
?100 The Node that notice request is extracted can extract safely now
?011 Cancellation current request or notice; Request such as cancellation Node adds or cancellation adapter notice
Other Keep
Table 2
More than describe in the whole exchange storage control system that the present invention proposes the process of collaborative work between each service node in detail.
In addition, the invention allows for a kind of service node.This service node can be applied in the whole exchange storage control system of the present invention's proposition.Fig. 9 is the structural representation of the service node that proposes in one embodiment of the invention.Referring to Fig. 9, in one embodiment of the invention, this service node inside can comprise:
Negotiation element is used for holding consultation with outside other service nodes, when negotiation result is management node, management node role notice is sent to the management processing unit;
The management processing unit is used for carrying out the management work of whole exchange storage control system after receiving management node role notice;
Service Processing Unit is used for carrying out data interaction with outside other service nodes.
As seen, when service node is consulted to be defined as management node, then it not only needs to finish original data interaction business, but also to finish the management work of whole exchange storage control system, when service node was consulted to determine not as management node, then it only needed to finish original data interaction business.
Figure 10 is the structural representation of management processing unit in one embodiment of the invention.Referring to Figure 10, in the service node that one embodiment of the invention proposes, when this service node during as management node, it not only needs to issue the configuration management information of whole exchange storage control system, characterize self whether signal of operate as normal but also need to produce in real time, therefore, its inner management processing unit specifically can the information of being divided into issue subelement and signal controlling subelement, wherein, information issues subelement, be used for after receiving management node role notice, configuration management information being sent to each outside service node respectively.And for the signal controlling subelement, based on as can be known in the analysis at step 501 place, in order to ensure under any circumstance, all can be by signal indication as the service node of management node whether in operate as normal, and guarantee after it breaks down, the role that other service nodes can be taken over management node continues to produce and output signal, preferably, the signal controlling subelement is used for after receiving management node role notice, produce the Locked signal of impulse form, after this Locked signal carried out non-locking processing, send to outside other service node.
Figure 11 is the concrete structure schematic diagram of signal controlling subelement in one embodiment of the invention.Referring to Figure 11, the concrete structure of signal controlling subelement inside can comprise:
Signal generator module is used for when the service node operate as normal of place, produces the Locked signal of impulse form, exports the Locked signal of this impulse form to replication module;
Replication module, the Locked signal that is used for the paired pulses form duplicates, and exports a road after duplicating to triple gate, and another road exports the delayed backward module to;
The delayed backward module is carried out after delayed backward handles the Locked signal of the impulse form that receives, exports the Enable Pin of triple gate to;
Triple gate is used for when Enable Pin is effective, exports the Locked signal of the impulse form that receives to the OD output module;
The OD output module is to the Locked signal of outside other service node output impulse forms.
After guaranteeing service node fault as management node, can not lock the Locked signal, referring to Figure 11, described signal controlling subelement can further include: the power supply signal module, like this, after its assurance is broken down, the Locked signal is not locked, the role who makes other service nodes can take over management node continues to produce and the cooperation of output signal realizes comprising:
Described signal generator module is further used for when the place service node can't operate as normal, with the fixed level signal, can be high level signal or low level signal, exports described replication module to;
Described replication module is further used for the fixed level signal that receives is duplicated, and exports the Yi Lumai after duplicating to described triple gate, and another road exports described delayed backward module to;
Described delayed backward module is further used for the fixed level signal that receives is carried out after delayed backward handles, and exports the Enable Pin of described triple gate to;
Described triple gate, being further used for detecting Enable Pin is effective low level signal, and when receiving high level signal, export high level signal to described OD output module, detecting Enable Pin is invalid high level signal, and when receiving low level signal, quits work;
Described power supply signal module is used for exporting power supply signal to described OD output module by pull-up resistor;
Described OD output module is further used at the high level signal that receives the output of described triple gate or when receiving power supply signal, stops to drive the Locked signal.
This shows, when can't operate as normal as the service node of management node, no matter it exports high level signal or low level signal, the OD output module finally all can stop to drive the Locked signal, thereby guarantee that other service nodes can continue to drive the Locked signal after taking over the management node role.
Referring to Figure 10, when service node during not as management node, its inner Service Processing Unit then is further used for receiving the outside signal of exporting as the service node of management node by driver, whether detect this signal is the Locked signal of impulse form, if not, can determine that then current service node as management node can't operate as normal, then will consult the negotiation element that notice is sent to service node inside, thereby the triggering negotiation element is timely and other service nodes are consulted definite service node as management node again.Wherein, detection signal whether be impulse form the Locked signal realization can for: Service Processing Unit detects to the received signal, if detect a pulse signal, then the value with counter adds 1, if in a setting-up time length, the value of this counter is compared with initial value variation has been taken place, and can determine that then signal is the Locked signal of impulse form, otherwise, can determine that signal is not the Locked signal of impulse form.
Be used to realize the processing of signal controlling subelement and a kind of particular circuit configurations that Service Processing Unit detects the Locked Signal Processing in the service node as shown in figure 12.Referring to Figure 12, service node 11 (Node 11) produces the Locked signal as management node, service node 21 (Node 21) does not produce the Locked signal as the non-management node, in this circuit, CMM101 is as the signal generator module among Figure 11, Mul102 is as the replication module among Figure 11, RC circuit and reverser 105 are as the delayed backward module among Figure 11, the triple gate of triple gate 103 for pointing out among Figure 11, OD door 104 is the OD output module among Figure 11, be connected to the pull-up resistor of OD door 104 and power supply VCC as the power supply signal module among Figure 11, driver 106 is the driver when service node reception Locked signal during as the non-management node.
In a word, the above is preferred embodiment of the present invention only, is not to be used to limit protection scope of the present invention.Within the spirit and principles in the present invention all, any modification of being done, be equal to replacement, improvement etc., all should be included within protection scope of the present invention.

Claims (13)

1. the management method of a whole exchange storage control system, described whole exchange storage control system in order to handle the read-write operation that applied host machine is initiated to disk array, is characterized in that this method comprises between applied host machine and disk array:
Any one service node in the whole exchange storage control system and other service nodes are consulted, and determine the service node as management node, carry out the management work of whole exchange storage control system as the service node of management node.
2. method according to claim 1, it is characterized in that, the step that described any one service node and other service nodes are consulted to determine as the service node of management node comprises: the broadcast packet that described any one service node will carry self precedence information sends to described other service nodes, and receive the broadcast packet that carries precedence information that described other service nodes are sent, and, by relatively priority of self and the precedence information in each broadcast packet of receiving, determine service node as management node.
3. method according to claim 1 is characterized in that, the step that described service node as management node is carried out the management work of whole exchange storage control system comprises:
Described service node as management node obtains the configuration mode of the whole exchange storage control system that sets in advance, if dynamic-configuration mode, then be that other all service nodes distribute ID, by sideband the ID that distributes is sent to the corresponding service node respectively, and management information is handed down to other all service nodes by in-band channel, if the static configuration mode then directly is handed down to other all service nodes by in-band channel with management information.
4. according to claim 1,2 or 3 described methods, it is characterized in that, the step that described service node as management node is carried out the management work of whole exchange storage control system comprises: described service node as management node when self operate as normal, the Locked signal and the output that produce impulse form;
This method further comprises: any one service node as management node does not detect whether received Locked signal is pulse signal, if not, then turn to and carry out described and other service nodes and consult to determine step as the service node of management node.
5. method according to claim 4 is characterized in that, after the Locked signal that produces impulse form, and before output, further comprises: the Locked signal that produces is carried out non-locking processing.
6. according to claim 1,2 or 3 described methods, it is characterized in that, the step that described service node as management node is carried out the management work of whole exchange storage control system comprises: described service node as management node receives the request of extracting that other service nodes are sent, the information of sending the service node of the request of extracting is deleted from management information, management information after upgrading is handed down to other all service nodes, and, when self need extract, priority according to other all service nodes, from other all service nodes, select one as management node, the management node notice is sent to selected service node.
7. according to claim 1,2 or 3 described methods, it is characterized in that, the step that described service node as management node is carried out the management work of whole exchange storage control system comprises: described service node as management node receives the insertion request that other service nodes are sent, the configuration information of correspondence is sent to the service node of sending the request of insertion, and new administration information more, the management information after upgrading is handed down to other all service nodes.
8. service node, be used for whole exchange storage control system, wherein this whole exchange storage control system is between applied host machine and disk array, in order to handle the read-write operation that applied host machine is initiated to disk array, it is characterized in that this service node comprises:
Negotiation element is used for holding consultation with outside other service nodes, when negotiation result is the service node at described negotiation element place when being management node, management node role notice is sent to the management processing unit;
The management processing unit is used for carrying out the management work of whole exchange storage control system after receiving management node role notice;
Service Processing Unit is used for carrying out data interaction with outside other service nodes.
9. service node according to claim 8 is characterized in that, described management processing unit comprises:
Information issues subelement, is used for after receiving management node role notice configuration management information being sent to each outside service node respectively;
The signal controlling subelement is used for producing the Locked signal of impulse form after receiving management node role notice, this Locked signal is carried out non-locking processing after, send to outside other service node.
10. service node according to claim 9 is characterized in that, described signal controlling subelement comprises:
Signal generator module is used for when the service node operate as normal of place, produces the Locked signal of impulse form, exports the Locked signal of this impulse form to replication module;
Replication module, the Locked signal that is used for the paired pulses form duplicates, and exports a road after duplicating to triple gate, and another road exports the delayed backward module to;
The delayed backward module is carried out after delayed backward handles the Locked signal of the impulse form that receives, exports the Enable Pin of triple gate to;
Triple gate is used for when Enable Pin is effective, exports the Locked signal of the impulse form that receives to the OD output module;
The OD output module is to the Locked signal of outside other service node output impulse forms.
11. service node according to claim 10 is characterized in that, described signal controlling subelement further comprises: the power supply signal module, wherein,
Described signal generator module is further used for exporting the fixed level signal to described replication module when the place service node can't operate as normal;
Described replication module is further used for the fixed level signal that receives is duplicated, and exports a road after duplicating to described triple gate, and another road exports described delayed backward module to;
Described delayed backward module is further used for the fixed level signal that receives is carried out after delayed backward handles, and exports the Enable Pin of described triple gate to;
Described triple gate, being further used for detecting Enable Pin is effective low level signal, and when receiving high level signal, export high level signal to described OD output module, detecting Enable Pin is invalid high level signal, and when receiving low level signal, quits work;
Described power supply signal module is used for exporting power supply signal to described OD output module by pull-up resistor;
Described OD output module is further used at the high level signal that receives the output of described triple gate or when receiving power supply signal, stops to drive the Locked signal.
12. according to any described service node in the claim 9 to 11, it is characterized in that, described Service Processing Unit, be further used for receiving the outside signal of exporting as the service node of management node by driver, whether detect this signal is the Locked signal of impulse form, if not, then will consult notice and be sent to described negotiation element.
13. a whole exchange storage control system is characterized in that, comprises more than two as any described service node in the claim 8 to 12.
CN200710120636XA 2007-08-22 2007-08-22 Whole exchange storage control system and management method thereof as well as service node Expired - Fee Related CN101374152B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN200710120636XA CN101374152B (en) 2007-08-22 2007-08-22 Whole exchange storage control system and management method thereof as well as service node

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN200710120636XA CN101374152B (en) 2007-08-22 2007-08-22 Whole exchange storage control system and management method thereof as well as service node

Publications (2)

Publication Number Publication Date
CN101374152A CN101374152A (en) 2009-02-25
CN101374152B true CN101374152B (en) 2011-08-31

Family

ID=40448075

Family Applications (1)

Application Number Title Priority Date Filing Date
CN200710120636XA Expired - Fee Related CN101374152B (en) 2007-08-22 2007-08-22 Whole exchange storage control system and management method thereof as well as service node

Country Status (1)

Country Link
CN (1) CN101374152B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106452886A (en) * 2016-10-31 2017-02-22 中国联合网络通信集团有限公司 Super node selection method and apparatus
CN111629013B (en) * 2019-02-27 2024-06-18 北京奇虎科技有限公司 Service node management method, device and node network

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1673975A (en) * 2004-03-25 2005-09-28 株式会社日立制作所 Storage system

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1673975A (en) * 2004-03-25 2005-09-28 株式会社日立制作所 Storage system

Also Published As

Publication number Publication date
CN101374152A (en) 2009-02-25

Similar Documents

Publication Publication Date Title
CN101022363B (en) Network storage equipment fault protecting method and device
CN101027873B (en) Systems and methods for network management
US6272500B1 (en) Object-oriented device management system and method
CN100454817C (en) Computer remote wakeup method and device based on network
CN103995868A (en) Distributed system oriented global transaction manager and transaction handling method
CN104656574A (en) Time domain analysis and control-based automatic checking and monitoring method
CN108551356A (en) A kind of method and system that centralized meter-reading system power failure reports
CN101374152B (en) Whole exchange storage control system and management method thereof as well as service node
CN1926825B (en) Terminal capable of substituting frames for a control station
CN102929224A (en) Condition acquisition controller (CAC) for power transformer equipment and system control method
CN102075341A (en) Method and system for main-standby synchronization
CN109936532A (en) A kind of data bus transmission security protection system
CN109213066A (en) PLC redundant control data backup method and system based on block chain technology
CN105138417B (en) A kind of asynchronous replication method based on zfs send
CN108791372A (en) The distributed system and its upgrade method of single face input based on automatic Pilot
CN107257373A (en) A kind of grid model data maintenance management method based on CIM/E files
CN113240945B (en) Flight telegraph automatic processing fault-tolerant system and method in main and standby civil aviation system
CN203101932U (en) Power transformation equipment condition acquisition controller
DE112010003684B4 (en) System and method for responding to error detection
Cisco The DIST Component
Cisco The DIST Component
Cisco The DIST Component
Cisco The DIST Component
CN114666195A (en) Multi-level safety protection data exchange sharing system and method thereof
CN101997672A (en) Information security transmission method and 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
C14 Grant of patent or utility model
GR01 Patent grant
CP03 Change of name, title or address

Address after: 310052 Binjiang District Changhe Road, Zhejiang, China, No. 466, No.

Patentee after: Xinhua three Technology Co., Ltd.

Address before: 310053 Hangzhou hi tech Industrial Development Zone, Zhejiang province science and Technology Industrial Park, No. 310 and No. six road, HUAWEI, Hangzhou production base

Patentee before: Huasan Communication Technology Co., Ltd.

CP03 Change of name, title or address
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20110831

Termination date: 20200822

CF01 Termination of patent right due to non-payment of annual fee