CN1568467B - Exactly once cache framework - Google Patents
Exactly once cache framework Download PDFInfo
- Publication number
- CN1568467B CN1568467B CN02820026.8A CN02820026A CN1568467B CN 1568467 B CN1568467 B CN 1568467B CN 02820026 A CN02820026 A CN 02820026A CN 1568467 B CN1568467 B CN 1568467B
- Authority
- CN
- China
- Prior art keywords
- server
- master server
- copy
- data item
- cluster
- 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 - Lifetime
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1095—Replication or mirroring of data, e.g. scheduling or transport for data synchronisation between network nodes
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/56—Provisioning of proxy services
- H04L67/566—Grouping or aggregating service requests, e.g. for unified processing
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/56—Provisioning of proxy services
- H04L67/568—Storing data temporarily at an intermediate stage, e.g. caching
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L69/00—Network arrangements, protocols or services independent of the application payload and not provided for in the other groups of this subclass
- H04L69/30—Definitions, standards or architectural aspects of layered protocol stacks
- H04L69/32—Architecture of open systems interconnection [OSI] 7-layer type protocol stacks, e.g. the interfaces between the data link level and the physical level
- H04L69/322—Intralayer communication protocols among peer entities or protocol data unit [PDU] definitions
- H04L69/329—Intralayer communication protocols among peer entities or protocol data unit [PDU] definitions in the application layer [OSI layer 7]
Abstract
A system for managing objects in a clustered network includes a file system containing at least one copy of a data object (214). The system can include several clustered servers in communication with the file system (212). A lead server is selected, which contains a distributed consensus algorithm for selecting a host server (206), and which utilizes multicasting while executing rounds of the algorithm. The selected host server (206) can contain a copy of the data object (208), such as in local cache, providing access to the local copy (208) to any other server in the cluster. Any change made to an item hosted by the host server (206) can also be updated in the file system (212). If the host server (206) becomes unable to host the object, a new host can be chosen using the distributed consensus a algorithm. The other severs (216, 218) are then notified ot the new host by multicast messaging.
Description
The right of priority of the following application that the application's requirement is introduced here:
The application number of being applied for September 6 calendar year 2001 by Dean Bernard Jacobs and EricHalpern is No.60/317,718, denomination of invention is the U.S. Provisional Patent Application of " EXACTLY ONCE CACHE FRAMEWORK ".The denomination of invention of being applied on September 4th, 2002 by Dean Bernard Jacobs and EricHalpern is " EXACTLY ONCE CACHE FRAMEWORK " U.S. Patent application.
The application number of being applied for September 6 calendar year 2001 by Dean Bernard Jacobs and EricHalpern is No.60/317,566, denomination of invention is the U.S. Provisional Patent Application of " EXACTLY ONCE JMSCOMMUNICATION ".
The denomination of invention of being applied on September 4th, 2002 by Dean Bernard Jacobs and EricHalpern is the U.S. Provisional Patent Application of " EXACTLY ONCE JMS COMMUNICATION ".
The copyright notice
The disclosed part of this patent documentation comprises material protected by copyright.In the time of in it appears at the patent document of Patent﹠Trademark Office or writes down, the copyright owner does not oppose anyone the disclosed patent documentation of this patent of reproduction of faxing, otherwise, in any case then all keep all copyrights.
The case of cross reference:
Following U.S. Patent application is the cross reference case and here introduces to do reference:
The application number of being applied for July 16 calendar year 2001 by Dean Bernard Jacobs, Reto Kramer and Ananthan Bala Srinivasan is No.60/305,986, denomination of invention is the U.S. Patent application of " DATAREPLICATION PROTOCOL ".
Technical field
The present invention relates to be used for the technology of distribute objects in the middle of the server of network cluster device.
Background technology
In Distributed Computer System, often there is such situation, promptly several servers and/or network node must be worked together.When in the middle of described a plurality of equipment, having the representative network information that to share so that allowing them to be used as single entities, must coordinate these servers and node.With regard to resource and efficient, the method that can coordinate equipment is very expensive usually.
Usually, owing between a plurality of nodes, exist some information to transmit, therefore,, need certain synchronous in order to make these node unanimities.Yet this synchronous requirement is undesirable in the cluster networks environment.Many cluster environment have simply avoided utilizing any this synchronous requirement.Yet, in some applications, this consistent be essential.
Need to exist a cluster to attempt to get rid of the equipment that it is conducted interviews under the consistent situation at some.A kind of such equipment is an item register-file system.As long as it is underway that item is handled, just deposit the object that some must be preserved with continuous fashion, thereby if fault, then can recover to continue the object of preserving.
For it need be stored in a locational object, generally exist it to run on the item monitor on each server in cluster or the territory, after this this item monitor uses local file system to visit object.Each server can have its oneself transaction management device, so as on continuation existing problems hardly.Because each server all has its own transaction management device, so, do not need coordination yet.
For example, may have the cluster that comprises three servers, each server has a transaction management device.In these servers one may run into fault or because this server is not useable for the caused other problem of cluster.Because out of order server is the server of unique addressable specific transaction record, so, before this server can be used for this cluster once more, can not recover any item on the specific record.Because server must spend a large amount of time and solve these problems, so to recover this record be very difficult or efficient is very low at least.Important server problem may comprise such as motherboard short circuit on the server or power supply and burnt out such incident.
Summary of the invention
The present invention includes such system, this system is used for managing such as the object that is stored in the server on network or the cluster.This system comprises a data source, application or such as the such service within the cluster or outside the cluster of being positioned at of file system or Java messenger service parts.This system comprises several servers, and these servers are such as connecting by express network and communicating with this document system or application.
This system comprises a Boot Server (1ead server) of agreeing such as another server.This Boot Server is contained in hardware or the software bundling storage.This system comprises an algorithm that is used for selecting Boot Server in the middle of server, and this algorithm for example can be the algorithm that is built in the hardware cluster equipment.This Boot Server comprises that successively such as the Paxos algorithm such being used to select a distributed consistency algorithm of master server.Be used to select Boot Server algorithm can be used to select the algorithm of master server different or identical with it.
This master server comprises such as being stored in the item in the local cache buffer memory device or the copy of object.Master server provides the local copy that can conduct interviews to any server in network or the cluster.Master server can also provide unique access point that the object that is stored in the file system is conducted interviews, and perhaps can provide unique access point of using or service conducts interviews.Can also in file system, application or service, upgrade any variation master server institute speed buffering, that item that admitted or that had is made.
Can not admit described object if master server becomes, use the distribution consistency algorithm to select new main frame so, after this from file system or service, take out the required data of this object.The notified new server of another server in the cluster is being admitted this object.By connecting such as point-to-point or coming announcement server by the so suitable means of multileaving.
Description of drawings
Fig. 1 has provided the synoptic diagram of distributed object system according to an embodiment of the invention;
Fig. 2 has provided the synoptic diagram of another distributed object system according to an embodiment of the invention;
Fig. 3 has provided the process flow diagram that is used to select a method of master server according to of the present invention;
Fig. 4 has provided the process flow diagram that is used to select a method of new master server according to of the present invention;
Fig. 5 has provided the process flow diagram that is used to use a method of Boot Server according to of the present invention;
Fig. 6 has provided the synoptic diagram of JMS message stores system according to an embodiment of the invention;
Fig. 7 has provided the block scheme according to the parts of computer system used in the present invention.
Embodiment
The solution of issue validity when system according to the present invention provides and has been not useable for the server zone storage such as becoming when the server that has data object.A solution can make another server in this cluster have the entitlement of this data object like this.But, such problem has appearred, promptly do not needing to arrive this two described data objects of server access under the situation that the data object on two servers is duplicated.
If cluster uses file system, data-carrier store or database (being generically and collectively referred to as " file system " hereinafter) the storage data to continue, and the more than addressable this document of server system, encounter problems if having first server of that object so, second server can automatically be taken over the task of visit data object.In addition, can utilize the employed algorithm of server in cluster or the cluster to come commander server to take over described entitlement.Yet another basic problem comprises makes cluster agree which server has resource or object at present, perhaps realizes in the middle of server " agreeing unanimously ".
Fig. 1 has provided an example according to cluster of the present invention system 100, will register 114 such object storage in file system 112 such as item in this example.The all addressable this document of Servers-all 106,116,118 in the cluster 110 system 112, but have only an all addressable at every turn registration 114 in these servers.Such as the copy 108 by storage registration 114 or by the registration 114 in accessible file systems 112 all, the master server 106 in the middle of the server in the cluster 110 will " have " or " admittance " this registration 114.Any server 116,118 of in the cluster 110 other can be visited the copy 108 of this record, and/or can visit registration 114 by master server 106.For example, a client or browser 102 can be asked its network 104 that refers to the server 116 in the cluster 110.This server can visit the copy 108 of the transaction record on the master server 106 by network 104.If must the update transaction record, copy 108 former registrations 114 on file system 112 so and be updated.
For example when server during as the storer of object, such as being stored in the local cache buffer memory device by copy and making other server in this cluster can use this copy with data object, but perhaps by making the object in the unique service device random access file system, this server then can " have " or " admittance " this data object, so that all other servers in this cluster must visit those objects by this master server.This has guaranteed that object " just in time once " is present in the server zone storage.
Fig. 3 has provided one and has handled 300, and this processing 300 is used to set up the admittance to an object.The so distributed consistency algorithm 302 of utilization such as Paxos algorithm is selected master server.Because the server in the cluster usually must with regard to how in the cluster server distribute objects and reach general agreement or agree unanimously, so this algorithm is called as " distributed consistance " algorithm.
If by the object admitted for example is to be cached in the master server, so, the copy of the data object that from file system, takes out be transferred into master server and as an object storage in local cache buffer memory device 304.After this such as by master server to network or suitably the local copy of other this object of server notification in the cluster be present in the master server and local copy will be used in the processing that network requirement 306 is in the future carried out.
In the Paxos algorithm of an example that is distributed consistency algorithm, select a server with as master server or Boot Server by the webserver, this webserver has guided a series of " the consistent circulation (consensus rounds) ".In each consistent circulation, new master server or Boot Server have been advised.Circulation continues to accept the server of being advised up to the server of most or legal number always.Although this system is configured to always start a circulation so that select a master server by Boot Server,, any server all can come suggestion master server or Boot Server by starting a circulation.Can be used for simultaneously the circulation of different choice.Therefore, discern circulation by period or a pair of like this value and select, as for this a pair of value, one of them and described circular correlation, and another with guide described round-robin server relevant.
It is as follows to be used for such round-robin step, although other step and/or method can be suitable for some situation or application.At first, by Boot Server other server that one " concentrating " message is sent in the cluster is started a circulation.Concentrate message set suffered from server in the cluster with the relevant information of circulation that participates at preceding these servers carrying out.If handle consistent the preceding circulation of existence for a specific selection, so described concentrated message also announcement server does not submit to comfortable preceding round-robin to select.For example, in case Boot Server has gathered the response from least one semigroup storage server, Boot Server just can determine described value to advise next circulation and this suggestion is sent to the cluster server with as " startup " message so.In order in this method, to make Boot Server select a value, must receive initial value information from server to offer suggestions.
In case server receives the initiation message from Boot Server, it makes response by transmitting one " acceptance " message, and this shows master/Boot Server that this server acceptance is advised.If this Boot Server receives the message of accepting of most or legal number server, its output valve of Boot Server value of being set in circulation to be advised so.If Boot Server does not receive the acceptance (" agreeing unanimously ") of most or legal number in the period of regulation, Boot Server can start new circulation so., Boot Server agrees unanimously that Boot Server can notify described cluster or the described server of the webserver will be designated as selected server so if receiving.Can this notification broadcast be arrived the webserver by any suitable broadcast technology, for example can connect or multileaving by point-to-point.
Utilize the condition of agreeing unanimously that can guarantee the method for agreeing unanimously with selection in the relevant information of preceding circulation by suggestion.Require this information from the most at least webserver, so that for any two circulations, exist at least one to participate in two round-robin servers.
Described Boot Server can be by accepting up-to-date numbering cycle from a value to each server interrogates server, may to inquire also that the value of being accepted selects a value.In case Boot Server has obtained this information from the server of most or legal number, it can select to be used for new round-robin value, and this value equals the central up-to-date round-robin value of response.If the neither one server relates to circulation the preceding, Boot Server can also be selected initial value so.For example, if Boot Server receives the response that the circulation of accepting last time is x, and current circulation is y, and server represents not accept any circulation between x and the y so, so that keep consistency.
Sampling between the circulation Boot Server and the webserver comprises following message alternately:
(1) " Collect "---a message is sent to the server that starts a new circulation " r ".This message can be taked m=(" Collect ", form r).
(2) " Last "---will send Boot Server to from the message of a webserver, this webserver provides " a " and this round-robin value " v " that last time, circulation was accepted.This message can be taked m=(" Last ", r, a, form v).
(3) " Begin "---a message is sent to the server of the issue described value relevant with circulation r.This message can be taked m=(" Begin ", r, form v).
(4) " Accept "---send a message to described Boot Server from the server that is used to accept the described value relevant with the r that circulates.This message can be taked m=(" Accept ", form r).
(5) " Success "---an information is sent to the server of the selection that is used to issue the described value v relevant with the r that circulates.This information can be taked m=(" Success ", r, form v).
(6) " Ack "---send a message to Boot Server from a server, this server admits to receive the decision relevant with the r that circulates.This information can be taked m=(" Ack ", form r).
Existence is positioned at the file system that hardware cluster or software bundling storage inside or external server are separated with it.This document system is such as the storage transaction record by recording storage is also continued this record copies at first disk to the second disk that is positioned within the file system.If first disk crash, file system can make described cluster and/or server can't discover described stroke to bump and can obtain recorded information from second disk so.This document system can also select so that this record copies is used as on the 3rd disk of second disk backup to it.
The angle of the server from cluster, this document system can be single resources.In one embodiment, this server can only be concerned about that single server locates to have this document system at any time.
Another example according to the present invention comprises the Cache that is arranged in the server zone storage.For example, therefore wish in cluster environment, to make the server in the single Cache cluster show data object because of the reason of network performance.It may be favourable in the single Cache that a plurality of items are kept at, and need not constantly to get back to permanent memory because of the addressable Cache of the server in the cluster.Taking out a plurality of items that have been arranged in storer can increase the utilization factor of this system greatly, because the time of hiting data storehouse or file system is concentrated relatively.
Yet the problem that single Cache had is exactly that the object that must guarantee to be stored in the storer is identical with object in being stored in file system one disk.Needing this conforming reason is to guarantee that any operation or the calculating carried out on institute's cache object have produced correct result.Another reason is exactly to bump or otherwise infect or the Cache of the unavailable file system that causes is repaired drawing owing to Cache.
Have at least two basic modes to handle this class high-speed cache in the cluster, though alternate manner works to some application at least.A kind of mode is duplicated this Cache in a plurality of places exactly.This processing is problematic, because just the Servers-all that is all required to copy described Cache by any variation of cache entry is agreed unanimously this variation, perhaps knows this variation at least.With regard to resource and performance, verified this is very expensive.
Distribute specific server so that it is the owner of cluster high speed buffer according to of the present invention one replaceable method, and by all addressable Cache of these specific servers.Any server in the cluster is admitted this Cache.Each server can admit one, several or do not admit Cache.This Cache of host on single server perhaps spreads to it in cluster in the middle of some or the Servers-all.Cluster itself can be any suitable cluster, for example the one group of specified server of software application in hardware cluster or given by being in " software " cluster.
It is also conceivable that as an item login that is arranged in a kind of object in somewhere in the system and/or an example of Cache.Can look to guaranteeing that any one this object only exists once in a cluster, and this object is always available.Can also look to guaranteeing if admit the server of this object fault to occur then described object can be restored on another server, and this object can be used for this cluster.
Fig. 4 has provided a method 400 that is used to recover.In the method, determine whether master server can continue to admit object 402, for example determine whether server still can be used for this network.If not, utilize distributed consistency algorithm to select new main frame.This selection can be according to selecting original host 404 employed methods to carry out.The copy that takes out data object from file system is provided for new main frame, and it is stored in the local cache buffer memory device 406.On the network or suitably the notified new master server of other server in the cluster comprises the local copy of described object, and local copy will be used in any network requirement 408 in the future handled in the middle of.
System and a method according to the invention can define the object that is present in the cluster in what a position just, and can guarantee what these objects always existed.From the angle of server, whether utilize such as a file system come mirror image or the copy such as transaction record such to as if inessential.Angle from server, always have one can be by the permanent memory of any one server access in the cluster, the existence of object is checked on this system cycle ground, or object entitlement is appointed as the very short time cycle, so that frequent appointed object again is to guarantee existing on certain equipment on the network or in the cluster.
The hardware cluster comprises one group of equipment, and each equipment can move a plurality of servers.After each equipment, also there is a file system.Server in the hardware cluster normally is made of hardware, so they can be made decision faster and the server failure of hardware cluster inside is handled.The size of hardware cluster is restricted to the actual hardware of the equipment that includes described server.Server in the hardware cluster can be used as the server in the software bundling storage, and because the independent server on this equipment can be used for this network, so can also comprise the webserver.
One the shared-file system that is used for these equipment is such as the Servers-all that can be used for cluster by express network.File system can also be redundant.In one embodiment, realize this redundancy by a plurality of data disks that use file system.In a kind of like this realization of redundancy, whenever all can copy object by a plurality of disks what object is written to this document system.When regarding this document system as " black box ", this document system can bear the fault in the middle of arbitrary disk and still can provide in the cluster any server to the visit of data item.
These objects of supposing to be kept in the storer always recover by reliable, nonvolatil storage mechanism, then can set up according to of the present invention its to be called as the structure of " just in time once " structure.For example, have an object of expression transaction record.As long as produced, then upgrade corresponding transaction record to the calling of this object.This comprises from database being read or of being written in the database calls.The object of expression transaction record is arranged on the server such as the such cluster of master server.As long as at least one startup of server (up) and operation in the cluster, so just in time primary structure can guarantee that this server can be taken over the entitlement of this record if another server has fault.
The object that may have an expression Cache.When upgrading Cache, this renewal can also be write back in the permanent storage.When a server in a plurality of servers need use data item, require this server to study (go through) this object comprehensively.If be used to admit the described server of the object of representing Cache to break down, then this object recovered on another server.The object that is recovered can take out all necessary informations from permanent storage.
Just in time primary structure can be used as the employed memory buffer unit of cluster.This structure provides its single Cache that data in system can be provided, and this system is recovered by reliable, a permanent storer.As long as reading of data from Cache need not to visit permanent storer and can carry out this and read.Yet, when being updated/written to this in Cache, must write back, if so that exist a fault can make this system recovery by permanent storer.
Just in time primary structure importance comprises a kind of mode, can take out a kind of according to the method for using and/or realizing changing in this mode.Created such as the new distribution type object that is called as " just in time object ".Object local cache copy that for example is data item in the file system just in time, or unique access point that this data item of server in the cluster is conducted interviews.Realize that this abstract essential technique is also very important.
System of the present invention can utilize it to can be used in distributed conforming a plurality of methods any one, such as a method of utilizing above-mentioned Paxos algorithm.Select a kind of like this algorithm, effective method is provided this algorithm so that multinode and/or distributed node are agreed a value of object.Even node has fault and/or returns, also can select this algorithm so that it works in handling process through consultation.
The conventional method of network cluster device is to utilize reliable broadcast, guarantees in the method each message is sent to its predetermined acceptance person, perhaps sends it to the server of each predetermined action at least.This method is difficult to make system's parallelization, because reliable broadcast need the recipient want affirmative acknowledgement to receive a message before moving on on next message or the recipient.Utilize the distributed algorithm of multileaving can reduce the number of times of assurance, receive a message because multileaving can not guarantee Servers-all.Yet multileaving can make this methods simplification so that this system can participate in parallel processing, because single message is by concurrent all cluster servers that are multi-cast to, and need not to wait for reaction from each server.The server that does not receive multicast message at last can be from Boot Server or another cluster server or the webserver taking-up information.As used herein, the webserver can be with reference to any server on the network, and no matter this webserver is in the hardware cluster, in the software bundling storage or outside any cluster.
Just in time primary structure importance is to have reduced consistent difficulty.According to the present invention, come multicast message can improve the performance that distributed consistance realizes by utilizing distributed consistency algorithm.This method can be and makes Servers-all agreement consistent with each other and required message and/or Internet traffic minimizes.
When multileaving, can adopt in the Several Methods.Be called as in first method of " single-phase distribution " at it, Boot Server is multi-cast to a message on all other servers that are positioned on the network, for example being used in the circulation of Paxos algorithm or being used for a kind of like this state, promptly has been that an object has been selected new main frame.In the method, master server need only transmit a message, and this message is sent in the network on any available server.If this network of disconnection that a server is interim, this server requirement is discerned new main frame after on getting back to this network so.
Utilize another its be called as the multileaving method of " two-phase distribution ", Boot Server can utilize a next preselected master server of suitable algorithm.Yet before distributing an object to this main frame, this Boot Server can be connected with all other servers in the cluster to determine whether server agrees selected new master server.Boot Server can and be connected with each server by the point-to-point connection, perhaps can transmit the multileaving requirement and wait for that after this each server is to reply.If server is disagreed with selected main frame, Boot Server utilizes this algorithm to come preselected new main frame so.Boot Server after this in another circulation, transmit another multileaving request and the sign of preselected again main frame.
If each server is all agreed preselected main frame, Boot Server is given master server with this object distribution so.After this Boot Server multileaving one submission information with announcement server new variation has been come into force and therefore server will upgrade its information.
Just in time primary structure can also utilize " taxi " mechanism.In utilizing the process of this mechanism, use a kind of algorithm so that the cluster server is consistent with Boot Server, for example by utilizing distributed consistency algorithm.In case selected, this Boot Server is undertaken object is just in time once distributed to the various servers in the cluster.If set up this system so that fault has appearred in existing master server, the cluster server is always consistent with new Boot Server so.
When described Boot Server was activated, this Boot Server can be known all objects just in time that need be present among this system.Which server this Boot Server can determine to admit each object, and therefore can give selected server with this object " taxi ".When an object was hired out to a server, this server can have or admit a certain period of this object, for example hires out the time period.This Boot Server is configured to these taxis of regular update.If this method can provide a mode with guarantee a server fault arranged or disconnected in any mode or otherwise can not operation normally in cluster, this server can not make its taxi be updated so.
Be to be difficult to tell server and only be not make different between the server of response in the greatest problem that distributed system had under the situation about breaking down with fault.No longer admit object with any server that this network disconnects for a certain reason.Even this server cannot be used for this cluster, but it knows still that after the taxi time it will finish admittance to all objects.When this server can not be used to this cluster, its taxi will can not be updated.
Boot Server knows that also if Boot Server can not arrive master server within a certain amount of time, this master server will be abandoned the entitlement of this object so.The taxi cycle can be arbitrary reasonable time, for example several seconds.The taxi cycle is all identical to all objects in the cluster, perhaps can change between to picture.
Utilize the system of primary structure just in time can also be compact more.Operating system often provides the specific mechanism that more controls also can be provided near hardware more.Yet a problem of this method is exactly the restriction that it is subjected to available hardware.For example, the hardware of server cluster has 16 servers in order.Because these systems need certain hardware coupling closely, therefore the number of servers that can be included in the cluster have been made restriction.
On the other hand, compare with handled the trooping of specialized hardware cluster, just in time primary structure can be handled more and troop.Certain leverage of the service quality that this structure permission is obtained from a specific clusters device, thus allow bigger cluster.Different service quality for example can comprise whether transmitting message by the reliable protocol that connects such as point-to-point, perhaps by such as the less reliable of multileaving but the agreement of how friendly resource transmits message.Utilizing just in time the advantage of primary structure is can balance scalability and fault-tolerance, so that the user can make this system be adapted to the needs of application-specific.
Attempt the high practicability solution such as the existing system of hardware cluster mechanism by the single mechanism that has (will be provided for described cluster) and support by second mechanism.If the first mechanism failure then exists and carries out " partner " who takes over, and any software that ran on first mechanism originally is transferred to second mechanism.
Boot Server can be distributed to one the server that is arranged in these hardware clusters according to just in time primary structure of the present invention, handle faster so that the Boot Server fault is handled the fault of comparing in the software bundling storage that becomes.Yet this Boot Server is issued to server with a spot of taxi, and no matter whether those servers are arranged in hardware cluster or software bundling storage.This configuration can make Boot Server recover quickly, though allow the software bundling storage greater than the hardware cluster, the software bundling storage still comprises the hardware cluster.
A this system 200 as shown in Figure 2.Hardware cluster 218 comprises that it includes the single institution of a plurality of servers 220,222,224.For example for raising the efficiency, this hardware cluster can be used to select a Boot Server 220 the server from this mechanism.In case selected Boot Server 220, this Boot Server just can select to be used for the main frame 206 of object 214 in file system 212, and it is inner or outside that this document system 212 may be positioned at software bundling storage 210.File system 214 itself can be second object 216 on another disk that is positioned at file system with object 214 copies, thus the persistence of providing.New main frame 206 utilizes the copy 208 that is cached at the described object on the described main frame 206 to take out this object 214 from file system 212.When through network 204 when browser or client computer 202 receive request such as service 206,216 and 220, if this server needs the cache copies of access object 208, this server is known and will be connected with master server 206 so.
Utilize a kind of method 500 of this system to be shown in Fig. 5.Use the algorithm of hardware cluster 502 to select Boot Server.This algorithm for example can be the tailor-made algorithm of hardware cluster equipment, perhaps can be its distributed consistency algorithm of only requiring hardware cluster server unanimity.That utilizes after this that Boot Server 504 had comes preselected master server such as the so distributed consistency algorithm of Paxos algorithm.After this with the sign of preselected main frame be multi-cast on another server that is arranged in its software bundling storage that includes the hardware cluster.This Boot Server receives from the approval of each server or does not approve of, and these servers will be operated soon and link to each other with cluster 508.If server with preselected master server consistent, so a submission information is multi-cast to the cluster server with to this server notification preselected main frame admit this object now; Otherwise, if not approve of new main frame be preselected to server and start this processing once more.
Just in time primary structure for example is used for transaction record is handled or high-speed cache.This structure example is defined as management server just in time an object and hires out management server so that this management server will never quit work as being used for.
Fig. 6 has provided another example according to cluster system 600 of the present invention, and wherein, object 608 is used as and the relevant message memory of Java messenger service 612 (JMS).Servers-all 606,614,616 in the cluster 610 uses JMS, but they must transmit the message to message memory 608 and obtain any message from this message memory 608 by network 604.The master server 606 of server will " have " or " admittance " message memory 608 in the cluster 610.Client or browser 602 can propose directly to point to a request of server 616 in the cluster 610 to network 604.Server 616 only visits JMS by through network 604 message being sent to the message memory 608 that is arranged on the master server 606.
Fig. 7 has provided the block scheme 700 of the computer system that can be used as parts of the present invention or can realize the inventive method.The computer system of Fig. 7 comprises a processing unit 704 and primary memory 702.Processing unit 704 comprises single microprocessor, can comprise that perhaps a plurality of microprocessors are being multicomputer system with computer system configurations.The storage of primary memory 702 parts is by the performed instruction and data of processor unit 704.If the present invention is whole or part realizes that with software primary memory 702 can be stored executable program code when operating so.Primary memory 702 comprises the storer of other type of dynamic RAM group (DRAM), high speed and ultrahigh speed buffer and institute of the prior art called optical imaging.
System shown in Figure 7 comprises that further a mass storage 706, a peripherals 708, a user input apparatus 712, a portable storage medium drives 714, a graphics subsystem 718 and an output show 716.For simplicity, parts shown in Figure 7 link to each other by unified bus 720.Yet, it will be apparent to those skilled in the art that these parts can link to each other by one or more data links.For example, processor unit 704 and primary memory 702 are connected by local micro-processor bus, and mass storage 706, peripherals 708, portable storage medium drives 714 and graphics subsystem 718 link to each other by one or more I/O (I/O) bus.The mass storage of being realized by a disc driver, CD drive and other driver known in the field 706 is that a non-volatile memory device is to be used for 704 employed data and the instructions of storage of processor unit.In one embodiment, mass storage 706 storages are used for realizing software of the present invention so that with its primary memory 702 of packing into.
Portable storage medium drives 714 link to each other with portable non-volatile memory medium such as floppy disk so as to/from computer system I/O data and program code shown in Figure 7.In one embodiment, be used to realize that system software of the present invention is stored in this portable medium, and input to described computer system through this portable storage medium drives 714.Peripherals 708 can comprise such as any of I/O (I/O) interface and adds subsidiary function on the described computing machine computer support equipment.For example, peripherals 708 can comprise that it can make link to each other with a network network interface and such as modulator-demodular unit, router or other such network hardware of other hardware known in the field of computer system.
User input apparatus 712 provides certain customers' interface.User input apparatus 712 can comprise Alpha's numeric keypad of being used to import Alpha's numeral and other information or such as mouse, follow the tracks of pen, stylus or the such indicating equipment of cursor direction key.For videotex and graphical information, the computer system of Fig. 7 comprises graphics subsystem 718 and Output Display Unit 716.Output shows that 716 comprise a cathode ray tube (CRT) display, LCD (LCD) or other suitable display device.Graphics subsystem 718 receives text and graphical information, and this information is handled to output to demonstration 716 devices.In addition, the system of Fig. 7 comprises output unit 710.The example of suitable output unit comprises loudspeaker, printer, network interface, monitor, reaches other output unit known in the field.
The general parts that computer system comprised that in the computer system that is applicable to some specific embodiment of the present invention, can find Fig. 7, and it represents this computer components of known in the field one big class.Therefore, the computer system of Fig. 7 can be a personal computer, workstation, server, microcomputer, mainframe computer or any other computing machine.Computer system 700 can also adopt different bus structure, the network platform, multi processor platform or the like.Can use any operating system that comprises Unix, Linux, Windows, Macintosh OS, Palm OS and other suitable operating system.
In order to illustrate and to describe and proposed the preferred embodiments of the present invention.This is not to the detailed description of disclosed precision architecture or restriction.Obviously, it is evident that to those skilled in the art and can make various modifications and variations.For better explanation principle of the present invention and its practical application, can select and describe this embodiment, thereby make those of ordinary skills know that the present invention can be used for various embodiment and can make the modification that it is suitable for specific application target.Scope of the present invention by following claim with and the equivalent defined.
Claims (51)
1. system that is used for object on the managerial grid comprises:
A plurality of webservers, each webserver are used for and the network data sources traffic; And
Be arranged in a Boot Server of described a plurality of webservers, this Boot Server comprises a distributed consistency algorithm that is used for selecting from described a plurality of webservers a master server, this master server comprises an object relevant with a data item in the network data source, thereby makes in the described a plurality of webservers that need the visit data item any one can visit described object on the described master server.
2. according to the system of claim 1, wherein, the described webserver is to choose from a group of being made up of hardware cluster server and software bundling storage server.
3. according to the system of claim 1, wherein, described distributed consistency algorithm comprises the message circulation between described Boot Server and the described a plurality of server, and this circulation continues always, till the great majority in described a plurality of webservers are agreed described master server.
4. according to the system of claim 1, wherein, described master server comprises a data object, and this data object comprises the copy from the data in described network data source.
5. according to the system of claim 1, wherein, described master server comprises a data object, unique access point that this data object conducts interviews to the data item in the network data source with work.
6. according to the system of claim 1, wherein, described data item is a transaction record.
7. according to the system of claim 1, wherein, described distributed consistency algorithm is the Paxos algorithm.
8. system that is used for object on the managerial grid comprises:
A plurality of webservers, each webserver are used for and the network data sources traffic; And
Be arranged in a Boot Server of described a plurality of webservers, this Boot Server comprises a distributed consistency algorithm that is used for selecting from described a plurality of webservers a master server, this master server comprises a copy of a data item that is arranged in the network data source, thereby makes in the described a plurality of webservers that need the described data item of visit any one can visit described copy on the described master server.
9. system that is used for object on the managerial grid comprises:
A plurality of webservers, each webserver are used for and the network data sources traffic; And
Boot Server in described a plurality of webserver, this Boot Server comprises a distributed consistency algorithm that is used for selecting from described a plurality of webservers a master server, this master server comprises unique access point that the data item that is arranged in the network data source is conducted interviews, thus make described a plurality of webservers of needing this data item of visit any one all must pass through this master server and visit this data item.
10. system that is used for object on the managerial grid comprises:
One file system, this document system comprises the copy of at least one data item;
The a plurality of servers that communicate with this document system; And
Be arranged in a Boot Server of described a plurality of servers, this Boot Server comprises a distributed consistency algorithm that is used for selecting from described a plurality of servers a master server; And
Be arranged in a master server of described a plurality of servers, described master server comprises the local copy of described data item, described master server can make in described a plurality of server any one visit this local copy, and upgrades the described copy of the described data item in the described file system in any time of upgrading local copy.
11. according to the system of claim 10, wherein, described master server is further used for local copy is stored in the local cache buffer memory device.
12. according to the system of claim 10, wherein, described a plurality of servers comprise a cluster.
13. according to the system of claim 10, wherein, described file system is replicated in data item on a plurality of disks.
14. a system that is used for object on the managerial grid comprises:
One file system, this document system comprises the copy of at least one data item;
The a plurality of servers that communicate with this document system;
One hardware cluster, this hardware cluster comprise the hardware cluster server that is arranged in described a plurality of servers, and described hardware cluster comprises an algorithm that is used for selecting from described a plurality of hardware clusters a Boot Server;
Be arranged in a Boot Server of described hardware cluster server, this Boot Server comprises a distributed consistency algorithm that is used for selecting from described a plurality of servers a master server; And
Be arranged in a master server of described a plurality of servers, described master server comprises the local copy of data item, if when described master server can make any one addressable this local copy in described a plurality of server and upgrade local copy with regard to the copy of the data item in the updating file system.
15. according to the system of claim 14, wherein said master server is arranged in described hardware cluster.
16. a method that is used for object on the managerial grid comprises:
Utilize distributed consistency algorithm from a plurality of webservers, to select a master server;
Send the copy of a data item to described master server from file system; And
The described master server of notifying other webserver to include a data item copy will be used in the processing of network requests.
17., further comprise when the copy on the master server is modified the step of the data item in the updating file system according to the method for claim 16.
18., further comprise and limit other webserver visits file system through described master server step according to the method for claim 16.
19., comprise that further a copy that guarantees to have only data item is present in the step of the outside of this document system according to the method for claim 16.
20., comprise that further a copy that guarantees data item always is present in the step of this document system outside according to the method for claim 16.
21. according to the method for claim 16, further comprise, then utilize distributed consistency algorithm from a plurality of webservers, to select the step of new master server if master server is no longer admitted this object.
22. method according to claim 16, further comprise if master server is no longer admitted this object, then send the copy of a data item to the step of a new master server, the master server that this new master server is to use described distributed consistency algorithm to select from described file system.
23., comprise further that the new master server of notifying one of other webserver to comprise described data item so will be used in the step in the processing that network requests is carried out according to the method for claim 16 if master server is no longer admitted this object.
24. a structural system that is used for object on the managerial grid comprises:
A plurality of servers, all cacheable data object of each server;
One file system, this document system comprise at least one copy of data item;
One distributed consistency algorithm is used for selecting a master server from described a plurality of servers, and this master server is used for the copy of cached data object; And
One compartment system is used for the copy that this principal computer of server on the informing network comprises described data object.
25. one kind is used for object is hired out method to server on the network, comprises:
Utilize distributed consistency algorithm from a plurality of webservers, to select a master server;
Data object is hired out to described master server, and wherein this master server has in the time period or admits this data object hiring out;
From file system, take out the copy of data item to deliver to master server; And
The copy that comprises data item to other webserver notice master server is handled network requests being used for.
26., further comprise step: in another hires out the time period, upgrade termly this data object is hired out to master server according to the method for claim 25.
27., further comprise step:, then this data object is hired out to one of new master server and hired out the time period in case the taxi time period on the master server has expired according to the method for claim 25.
28. one kind is used for object is hired out method to the server on the network, comprises:
Select a Boot Server in a plurality of hardware cluster servers from the hardware cluster;
Utilize the distributed consistency algorithm on the described master server from a plurality of webservers, to select a master server;
Data object is assigned to master server, and appointed master server can carry out independent visit at an official hour to data item in the cycle;
From file system, take out the copy of data item to deliver to master server; And
The copy that comprises data item to other webserver notice master server is handled network requests being used for.
29. the method that the entitlement that is used for the object on the network distributes comprises:
Select a Boot Server in a plurality of hardware cluster servers from the hardware cluster;
Utilize the distributed consistency algorithm on the described Boot Server from a plurality of webservers, to select a master server;
Data object is assigned to this master server, and appointed master server can be visited separately the data object on the network;
From file system, take out the copy of data item to deliver to master server; And
The copy that comprises data item to other webserver notice master server is handled network requests being used for.
30. a method that is used for guaranteeing to exist at cluster object comprises:
The master server that use is arranged in a plurality of servers provides the visit to the data object;
If described master server can not provide the visit to described data object, so, utilize distributed consistency algorithm from a plurality of servers, to select a new master server;
With information send to need provide to described data item provide visit new master server and
Notify other server in described a plurality of server new addressable this data object of master server.
31. a method that is used at the cluster distribute objects comprises:
Utilize distributed consistency algorithm to come from a plurality of webservers, to select a master server;
One data object is assigned to this master server, and appointed this master server can be visited separately a data item; And
The one described master server of notifying other server that is multi-cast in the cluster to comprise the copy of described data item with notice will be used in the processing of network requests.
32. a method that is used at the cluster distribute objects comprises:
Utilize distributed consistency algorithm to come from a plurality of webservers, to select a master server;
Whether each server that connects in the cluster can be accepted by those servers with definite selected master server; And
If it is acceptable that the Servers-all in the cluster is all agreed selected master server, notifies other server that is multi-cast in the cluster to transfer to selected new master server with one so and be responsible for notice.
33. a system that is used at the cluster distribute objects comprises:
Device is used for utilizing distributed consistency algorithm to select a master server from a plurality of webservers;
Device is used for taking out the copy of data item to deliver to master server from a file system; And
Device, this master server that is used for notifying other webserver to comprise this data item copy will be used in the processing that network requests is carried out.
34. a method that is used for object on the managerial grid comprises:
Utilize and select a master server in a plurality of webservers of Paxos algorithm from the software bundling storage; And
Data object is assigned to described master server, and wherein this master server is provided in the network the independent visit to this data object.
35., further comprise the step of from a file system, taking out these data object data according to the method for claim 34.
36., comprise that further this master server of notifying other webserver to comprise the object relevant with described data item will be used in the step in the processing that network requests is carried out according to the method for claim 34.
37., further comprise with the sign multileaving of new master server step to another webserver according to the method for claim 34.
38., wherein utilize the Paxos algorithm to come to select in a plurality of webservers from the software bundling storage the described step of a master server to comprise cyclical information be multi-cast to another webserver according to the method for claim 34.
39. the system that the object that is used for cluster distributes comprises:
Device is used for utilizing distributed consistency algorithm to select a master server from a plurality of webservers;
Device is used for giving master server with a JMS object distribution, and this master server that is distributed provides the independent visit to the JMS object; And
Device is used to notify this master server of its webserver that independent visit to JMS is provided.
40. a method that is used for object on the managerial grid comprises:
Utilize and select a master server in a plurality of webservers of Paxos algorithm from the software bundling storage; And
The JMS message memory is assigned to master server, and this JMS message memory exists only on the described master server.
41., further comprise and notify other webserver step that this master server is admitted described JMS message memory according to the method for claim 40.
42., further comprise step according to the method for claim 40: with the sign multileaving of new master server to another webserver.
43., wherein utilize the Paxos algorithm to come to select in a plurality of webservers from the software bundling storage the described step of a master server to comprise cyclical information be multi-cast to another webserver according to the method for claim 40.
44. a system that is used for JMS object on the managerial grid comprises:
A plurality of webservers; And
Boot Server in described a plurality of webserver, this Boot Server comprises a distributed consistency algorithm that is used for selecting from the described a plurality of webservers master servers, and this master server comprises a JMS object so that its any one of a plurality of webservers that need visit JMS must conduct interviews to the JMS object on the master server.
45. according to the system of claim 44, wherein, the described webserver is to choose from be made up of hardware cluster server and software bundling storage server such one group.
46. according to the system of claim 44, wherein, described distributed consistency algorithm comprises the circulation message between described master server and the described a plurality of server, this circulation continues to agree master server up to the great majority of described a plurality of webservers always.
47. one kind is used for the JMS object is hired out method to the server on the network, comprises:
Utilize distributed consistency algorithm to come from a plurality of hardware cluster servers, to select a master server;
The JMS object is hired out to this master server, and wherein this master server has in the time period or admits this JMS object hiring out; And
Notify this master server that this JMS object is admitted to other webserver.
48., further comprise step: in another hires out the time period, upgrade termly this JMS object is hired out to master server according to the method for claim 47.
49., further comprise step:, then described JMS object is hired out to one of new master server and hired out the time period in case the taxi time period on the described master server has expired according to the method for claim 48.
50. one kind is used for the JMS object is hired out method to the server on the network, comprises:
Select a Boot Server in a plurality of hardware cluster servers from the hardware cluster;
Utilize the distributed consistency algorithm on the described Boot Server to come from a plurality of webservers, to select a master server;
The JMS object is assigned to this master server, and appointed master server can provide the independent visit to described JMS object in a time cycle of regulation; And
Notify this master server of other webserver to admit described JMS object.
51. one kind is used for method that the object on the network is distributed, comprises:
Select a Boot Server in a plurality of hardware cluster servers from the hardware cluster;
Utilize the distributed consistency algorithm on the described Boot Server to come from a plurality of webservers, to select a master server;
The JMS object is assigned to this master server, and appointed master server can provide the independent visit to described JMS object; And
Notify this master server of other webserver to admit described JMS object.
Applications Claiming Priority (9)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US31771801P | 2001-09-06 | 2001-09-06 | |
| US31756601P | 2001-09-06 | 2001-09-06 | |
| US60/317,566 | 2001-09-06 | ||
| US60/317,718 | 2001-09-06 | ||
| US10/234,597 | 2002-09-04 | ||
| US10/234,693 | 2002-09-04 | ||
| US10/234,597 US7113980B2 (en) | 2001-09-06 | 2002-09-04 | Exactly once JMS communication |
| US10/234,693 US6826601B2 (en) | 2001-09-06 | 2002-09-04 | Exactly one cache framework |
| PCT/US2002/028199 WO2003023633A1 (en) | 2001-09-06 | 2002-09-05 | Exactly once cache framework |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1568467A CN1568467A (en) | 2005-01-19 |
| CN1568467B true CN1568467B (en) | 2010-06-16 |
Family
ID=27499750
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN02820026.8A Expired - Lifetime CN1568467B (en) | 2001-09-06 | 2002-09-05 | Exactly once cache framework |
Country Status (5)
| Country | Link |
|---|---|
| EP (1) | EP1433073A4 (en) |
| JP (1) | JP2005502957A (en) |
| CN (1) | CN1568467B (en) |
| AU (1) | AU2002332845B2 (en) |
| WO (1) | WO2003023633A1 (en) |
Families Citing this family (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7698465B2 (en) * | 2004-11-23 | 2010-04-13 | Microsoft Corporation | Generalized Paxos |
| JP4707477B2 (en) * | 2005-06-23 | 2011-06-22 | 富士通株式会社 | File sharing program and file sharing apparatus |
| CN103327066B (en) * | 2006-05-16 | 2016-08-17 | 甲骨文国际公司 | Method and system for schedule job in cluster |
| JP5395517B2 (en) * | 2009-05-29 | 2014-01-22 | 日本電信電話株式会社 | Distributed data management system, data management apparatus, data management method, and program |
| JP5123961B2 (en) * | 2010-02-04 | 2013-01-23 | 株式会社トライテック | Distributed computing system, distributed computing method, and distributed computing program |
| JP5292351B2 (en) * | 2010-03-30 | 2013-09-18 | 日本電信電話株式会社 | Message queue management system, lock server, message queue management method, and message queue management program |
| JP5292350B2 (en) * | 2010-03-30 | 2013-09-18 | 日本電信電話株式会社 | Message queue management system, lock server, message queue management method, and message queue management program |
| US8595546B2 (en) * | 2011-10-28 | 2013-11-26 | Zettaset, Inc. | Split brain resistant failover in high availability clusters |
| CN107181608B (en) * | 2016-03-11 | 2020-06-09 | 阿里巴巴集团控股有限公司 | Method for recovering service and improving performance and operation and maintenance management system |
| CN106899648B (en) * | 2016-06-20 | 2020-02-14 | 阿里巴巴集团控股有限公司 | Data processing method and equipment |
| JP7099305B2 (en) | 2018-12-20 | 2022-07-12 | 富士通株式会社 | Communication equipment, communication methods, and communication programs |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5802291A (en) * | 1995-03-30 | 1998-09-01 | Sun Microsystems, Inc. | System and method to control and administer distributed object servers using first class distributed objects |
| US6067477A (en) * | 1998-01-15 | 2000-05-23 | Eutech Cybernetics Pte Ltd. | Method and apparatus for the creation of personalized supervisory and control data acquisition systems for the management and integration of real-time enterprise-wide applications and systems |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6122629A (en) * | 1998-04-30 | 2000-09-19 | Compaq Computer Corporation | Filesystem data integrity in a single system image environment |
| US6367029B1 (en) * | 1998-11-03 | 2002-04-02 | Sun Microsystems, Inc. | File server system tolerant to software and hardware failures |
| US6389462B1 (en) * | 1998-12-16 | 2002-05-14 | Lucent Technologies Inc. | Method and apparatus for transparently directing requests for web objects to proxy caches |
| US6826601B2 (en) * | 2001-09-06 | 2004-11-30 | Bea Systems, Inc. | Exactly one cache framework |
-
2002
- 2002-09-05 CN CN02820026.8A patent/CN1568467B/en not_active Expired - Lifetime
- 2002-09-05 JP JP2003527615A patent/JP2005502957A/en active Pending
- 2002-09-05 AU AU2002332845A patent/AU2002332845B2/en not_active Expired
- 2002-09-05 EP EP02798120A patent/EP1433073A4/en not_active Ceased
- 2002-09-05 WO PCT/US2002/028199 patent/WO2003023633A1/en active Application Filing
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5802291A (en) * | 1995-03-30 | 1998-09-01 | Sun Microsystems, Inc. | System and method to control and administer distributed object servers using first class distributed objects |
| US6067477A (en) * | 1998-01-15 | 2000-05-23 | Eutech Cybernetics Pte Ltd. | Method and apparatus for the creation of personalized supervisory and control data acquisition systems for the management and integration of real-time enterprise-wide applications and systems |
Non-Patent Citations (1)
| Title |
|---|
| 全文. |
Also Published As
| Publication number | Publication date |
|---|---|
| AU2002332845B2 (en) | 2008-06-12 |
| CN1568467A (en) | 2005-01-19 |
| WO2003023633A1 (en) | 2003-03-20 |
| EP1433073A1 (en) | 2004-06-30 |
| JP2005502957A (en) | 2005-01-27 |
| EP1433073A4 (en) | 2009-11-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7660949B2 (en) | Exactly once cache framework | |
| US7921169B2 (en) | System and method for exactly once message store communication | |
| US10491560B2 (en) | Message delivery in messaging networks | |
| US8196150B2 (en) | Event locality using queue services | |
| US5987502A (en) | Workload management in an asynchronous client/server computer system | |
| US6996548B2 (en) | Method and apparatus for providing a reward for the use of a processor in a parallel processing environment | |
| CN1568467B (en) | Exactly once cache framework | |
| CN103049565B (en) | Application Instance and inquiry storage | |
| US6928477B1 (en) | Availability and scalability in clustered application servers by transmitting expected loads of clients to load balancer | |
| JP2004318744A (en) | Database processing method providing high availability | |
| CN100390776C (en) | Group access privatization in clustered computer system | |
| AU2002332845A1 (en) | Exactly once cache framework | |
| JP4224037B2 (en) | Service providing method and data processing apparatus | |
| JP2924779B2 (en) | Data hierarchy distribution processing method | |
| Snyder et al. | Cougaar agent communities | |
| Steiner et al. | Network services in the Athena environment | |
| JPH11298880A (en) | Shared storage type distributed video server system | |
| Videira et al. | Efficient communication in a design environment | |
| Nie et al. | An MAS Framework for Speculative Trading Research in Stock Index Futures Market | |
| JPH0934770A (en) | Distributed data base system | |
| JPH09160888A (en) | System for distributing transaction load in loosely-coupled multiplex computer 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 | ||
| ASS | Succession or assignment of patent right |
Owner name: ORACLE INT CORP Free format text: FORMER OWNER: BEA SYSTEMS CORP. Effective date: 20110418 |
|
| C41 | Transfer of patent application or patent right or utility model | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20110418 Address after: California, USA Patentee after: ORACLE INTERNATIONAL Corp. Address before: California, USA Patentee before: BEA SYSTEMS, Inc. |
|
| CX01 | Expiry of patent term | ||
| CX01 | Expiry of patent term |
Granted publication date: 20100616 |