Patent Application
20070061380
The present invention relates to propagating and managing data, transactions and events either within a database, or from one database to another.
Complex database systems often need to propagate data, transactions and events within the database system's own database and also from one database to another. For example, a bank may need to transmit a financial transaction across several databases in order to maintain integrity in their records. As another example, a shipping company may need to send package tracking information across several systems in order to make this data available to its customers.
One method employed to transfer information in database systems is to use a data stream infrastructure. One such data stream infrastructure is described in U.S. Pat. No. 6,889,231, entitled “Asynchronous Information Sharing System”, the entire contents of which are incorporated herein. Such data stream architectures use the publisher/subscriber model to propagate data and events in a data stream from sources to subscribed destinations (“targets”). The source and targets of a stream may be within the same database, or in different databases.
Typically, propagating information from a source to a target takes place via queues. Queues act as “post boxes” where different applications can look for “mail” in the form of messages. When a publisher (such as a database server or other application) wants to contact subscribers (such as other database servers or applications), the publisher can leave messages in these queues so that the subscribers will be able to find and retrieve these messages. Queuing also provides a unified framework for processing events.
Propagation of events between queues on different databases happens through database links (“dblinks”). When a dblink is initialized, the user specifies a database. After the dblink has been set up, the dblink may then be used to establish connections with the specified database.
The same database link may be used to propagate messages from several source queues in one database to several target queues in another database. For example, a dblink may be established between a database X and a database Y. The dblink may be used to propagate messages from a source queue X1 in database X to queues Y1, Y2 and Y3 in database Y. The same dblink may be used to propagate messages from another source queue X2 in database X to queues Y4, Y5, and Y6 in database Y.
The task of propagating messages from a source queue to one or more target queues is encapsulated in a “job”. When a job is executed, the corresponding task is performed. For any given source queue, the task of propagating messages from that source queue over a particular dblink is encapsulated in a single job, regardless of how many target queues will be receiving messages from the source queue over that particular dblink. Thus, a single job would be used to propagate messages from source queue X1 to queues Y1, Y2 and Y3. A separate job would be used to propagate messages from source queue X2 to queues Y4, Y5 and Y6. The same dblink (between database X and database Y) may be used to propagate the messages of both jobs.
Because propagation tasks are encapsulated into jobs on a queue-to-dblink basis, the jobs are referred to herein as “queue-to-dblink” jobs. Queue-to-dblink jobs involve multiplexing when the messages from the source queue are propagated to multiple destination queues within the same destination database.
To perform the queue-to-dblink job of propagating messages from source queue 101 to target queues 151, 152 and 153, a sending process 102 located on the source database and a receiving process 154 located on the target database make the connections that allow for the delivery of messages. For each message in queue 101, the sending process 102 determines which of the target queues 151, 152, 153 are to receive the message, inserts into the message an indication of the appropriate destination queue(s), and sends the message to the receiving process 154 over dblink 130. The receiving process 154 reads the destination queue information from the message, and sends the message to the specified destination queue(s).
One consequence of managing the propagation at the queue-to-dblink level of granularity is that situations may arise in which, within the same queue-to-dblink job, (1) a problem prevents the delivery messages to one destination queue, but (2) messages continue to be successfully delivered to other destination queues. For example, assume that target queue 152 has failed. When messages are propagated from the source queue 101 using a queue-to-dblink job, the messages destined for target queue 152 will return as undeliverable. Even though target queue 152 is unable to accept messages, the delivery of messages to target queues 151 and 153 within the same schedule will not be affected.
Under these circumstances, if the queue-to-dblink job is not disabled, subsequent propagations will result in further undelivered messages destined for target queue 152, but target queues 151 and 153 will receive their messages. Resending such undelivered messages is not a straight-forward process, and may require the manual intervention of a database administrator. On the other hand, if the queue-to-dblink job is disabled, no further undelivered messages will result for target queue 152. However, after disabling the queue-to-dblink job, target queues 151 and 153 will cease to receive their messages.
A “cluster” is a system in which multiple database servers have access to the same database. The database to which the multiple servers have access is referred to herein as a “cluster database”. The persistent storage that stores a cluster database is accessible by all database servers in the cluster. In contrast, each database server has its own volatile memory that is not directly accessible to the other database servers in the cluster.
Queues can be persistent or buffered. A persistent queue is stored persistently. Since persistent storage is shared by all database servers in a cluster, persistent queues may be directly accessed by all database servers in a cluster. A buffered queue is stored in volatile memory. Consequently, even though a buffered queue is associated with a cluster database that is accessible to many database servers, the buffered queue itself may only be directly accessible to the database server in whose volatile memory the buffered queue resides (the “owner instance” of the buffered queue).
Propagating information to a buffered queue in a cluster environment presents a unique challenge because the messages destined for a buffered queue of a cluster database do not merely have to be delivered to the right database, but have to be delivered to the right database server instance (the database server-in whose volatile memory they reside).
When the target is a clustered database system, dblinks are allocated on a source-to-target-instance basis. For example, dblink 206 is associated with the database 200 and instance 1 of the clustered database system. Thus, dblink 206 may be used to propagate messages from source queue 202 to one or more target buffer queues located on instance 1 of the target database 250, but cannot be used to propagate messages from source queue 202 to buffered queues in any other instance of cluster database system 250.
When events are propagated from a source queue to buffered queues in a clustered database using queue-to-dblink granularity, the instance that owns the buffered queues must be identified. The dblink is configured for connection from the source to the target buffer queue(s) by populating the dblink with various pieces of information, such as the hostname, port, username, and password, service name, and instance.
The service name is the name of the database, and the instance is the instance which owns the buffer queues that subscribe to the messages in the source queue. This information is typically entered by the user, and statically maintained by the dblink. In the system illustrated in
Problems arise when events are propagated from a source to multiple buffer queues in a clustered database. Specifically, because each dblink is associated with a single database server instance, and the propagation job is allocated on a source-to-dblink basis, all of the target buffer queues of any propagation task must be located on the same database server instance. This limitation precludes effective load balancing between the database instances of the clustered database system. Further, if the database instance specified in a dblink configuration fails, then the buffer queues are automatically migrated to another alive instance in the cluster database. The user then needs to manually change the configuration of the dblink to the other instance.
The present invention is illustrated by way of example, and not by way of limitation, in the figures of the accompanying drawings and in which like reference numerals refer to similar elements and in which:
A method for propagating information in a queue-to-queue propagation mode is described. In the following description, for the purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the present invention. It will be apparent, however, that the present invention may be practiced without these specific details. In other instances, well-known structures and devices are shown in block diagram form in order to avoid unnecessarily obscuring the present invention.
Techniques are provided for finer-grained point-to-point event propagation. Specifically, queue-to-queue jobs are used to propagate messages from a particular source (such as a source queue) to a particular destination queue. Each queue-to-queue job deals solely with propagating events between one source and one target queue. When messages are to be propagated to multiple destination queues, a separate queue-to-queue job is scheduled for each destination queue. Because each propagation job has its own propagation schedule, the propagation schedule of each queue-to-queue propagation can be managed separately. Even when multiple queue-to-queue propagations use the same dblink, each propagation job can be enabled, disabled, or altered independently of the other propagation jobs.
As mentioned above, a queue-to-queue job is a finer grained point-to-point propagation job than queue-to-dblink jobs previously employed for message propagation. A queue-to-queue job deals solely with propagating events between a source (such as a source queue) and a target queue. Because each source and destination queue pair has its own propagation schedule, the propagation schedule of each queue-to-queue propagation can be managed separately.
By managing the propagation of messages on a per-destination-queue basis, it is possible to disable propagation on a source-queue by destination-queue basis. Referring again to
Employing queue-to-queue propagation jobs allows a user to enable a connection after a target queue recovers from failure. Specifically, when a message is undeliverable using queue-to-dblink scheduling, a database administrator must manually handle the undeliverable messages once the queue is again enabled. However, in queue-to-queue propagation the job can be disabled before undeliverable messages accumulate. Consequently, no intervention is required by the database administrator in order for the messages to properly travel to the queue when the job is re-enabled after the problem is fixed.
When propagation tasks are encapsulated at the queue-to-queue level of granularity, the properties associated with those propagation tasks can also be specified at the per-destination-queue level of granularity. Those properties may include, for example, the priority assigned to the schedule, the latency assigned to the schedule, etc.
With respect to scheduling, different latency periods may be imposed on messages destined for different target queues. Referring again to
As mentioned above, in a cluster database system, each buffer queue is only maintained in the volatile memory of a single database instance. By scheduling propagation jobs on a queue-to-queue basis, a single source is able to propagate messages to multiple buffered queues that reside in several instances of the same database cluster.
For the purpose of illustration, it will be assumed that a message 1 must go to target buffer queue 1 owned by database server instance 1, a message 2 must go to target buffer queue 2 owned by database server instance 2, and a message 3 must go to target buffer queue 3 owned by database server instance 3.
In one embodiment, connections from a particular source to different database instances use separate dblinks. In such an embodiment, a first dblink would be used to propagate messages from source queue 402 to target queue 1. A second dblink would be used to propagate messages from source queue 402 to target queue 2. A third dblink would be used to propagate messages from source queue 402 to target queue 3.
In the embodiment described above, connections from a particular source to different database instances use separate dblinks. In an alternative embodiment, dblinks are not tied to specific database instances. Consequently, the same dblink may propagate messages to buffer queues in different instances. Such an embodiment is illustrated in
According to one embodiment, queue-to-instance mapping information is maintained. For example, cluster database system 404 may maintain queue-to-instance mapping information that indicates that target queue 1 is owned by instance 1, target queue 2 is owned by instance 2, and target queue 3 is owned by instance 3.
When a connection is created for propagating messages from a source to a particular destination queue, data that identifies the destination queue is stored in association with the connection. For example, connection 1 is the connection created for sending messages from source queue 402 to target queue 1. Thus, the metadata associated with connection 1 identifies target queue 1 as the destination queue for the connection. Connection 2 is the connection created for sending messages from source queue 402 to target queue 2. Thus, the metadata associated with connection 2 identifies target queue 2 as the destination queue for the connection. Connection 3 is the connection created for sending messages from source queue 402 to target queue 3. Thus, the metadata associated with connection 3 identifies target queue 3 as the destination queue for the connection.
A global listener 406 may search the queue-to-instance mapping information based on the destination queue information associated with a connection to determine the specific database instance to which messages on the connection should be sent. In an implementation of this method, the user still configures the dblink 408, but does not specifying any particular instance. When a connection is initiated, the global listener 406 communicates to the dblink which instance owns the queue. Thus, locating the correct instance does not require any user intervention.
For example, the metadata associated with connection 2 identifies target queue 2 as the destination queue of connection 2. The queue-to-instance mapping may be searched to determine that target queue 2 is currently located on instance 2. Therefore, messages sent over dblink 408 on connection 2 are sent to instance 2. In contrast, the metadata associated with connection 3 identifies target queue 3 as the destination queue of connection 3. The queue-to-instance mapping may be searched to determine that target queue 3 is currently located on instance 3. Therefore, messages sent over dblink 408 on connection 3 are sent to instance 3.
By no longer requiring all non-persistent events to be relegated to queues located only on one database server instance, the database system can more effectively load balance the clustered database. For instance, assume there are 500 buffered queues on the clustered database with five database server instances. Queue-to-dblink propagation requires that all 500 buffered queues be owned by only a single database server instance. That one database server instance would have a significantly heavier workload with respect to the queues than the other four instances. Allowing queues to be placed on multiple instances, such as 100 queues on each of the five servers, would alleviate or eliminate the problem above and result in more efficient uses of system resources.
As explained herein, encapsulating propagation tasks at the finer queue-to-queue level of granularity insulates destination queues from any problems that may occur relative to other destination queues. The finer granularity of scheduling also allows for greater load balancing, by eliminating the need to assign all destination buffer queues within the same database instance. However, the coarser queue-to-dblink granularity may produce better results under certain conditions, such as when each message of the source queue is to be sent to several persistent queues within a single destination database.
Therefore, in one embodiment, the user is given the option of encapsulating propagation tasks on a queue-to-dblink basis, or on a queue-to-queue basis. The user may specify queue-to-dblink granularity for some propagation tasks, and queue-to-queue granularity for other propagation tasks. Thus, a database system can concurrently use both queue-to-queue propagation and queue-to-dblink propagation.
Computer system 500 may be coupled via bus 502 to a display 512, such as a cathode ray tube (CRT), for displaying information to a computer user. An input device 514, including alphanumeric and other keys, is coupled to bus 502 for communicating information and command selections to processor 504. Another type of user input device is cursor control 516, such as a mouse, a trackball, or cursor direction keys for communicating direction information and command selections to processor 504 and for controlling cursor movement on display 512. This input device typically has two degrees of freedom in two axes, a first axis (e.g., x) and a second axis (e.g., y), that allows the device to specify positions in a plane.
The invention is related to the use of computer system 500 for implementing the techniques described herein. According to one embodiment of the invention, those techniques are performed by computer system 500 in response to processor 504 executing one or more sequences of one or more instructions contained in main memory 506. Such instructions may be read into main memory 506 from another machine-readable medium, such as storage device 510. Execution of the sequences of instructions contained in main memory 506 causes processor 504 to perform the process steps described herein. In alternative embodiments, hard-wired circuitry may be used in place of or in combination with software instructions to implement the invention. Thus, embodiments of the invention are not limited to any specific combination of hardware circuitry and software.
The term “machine-readable medium” as used herein refers to any medium that participates in providing data that causes a machine to operation in a specific fashion. In an embodiment implemented using computer system 500, various machine-readable media are involved, for example, in providing instructions to processor 504 for execution. Such a medium may take many forms, including but not limited to, non-volatile media, volatile media, and transmission media. Non-volatile media includes, for example, optical or magnetic disks, such as storage device 510. Volatile media includes dynamic memory, such as main memory 506. Transmission media includes coaxial cables, copper wire and fiber optics, including the wires that comprise bus 502. Transmission media can also take the form of acoustic or light waves, such as those generated during radio-wave and infra-red data communications.
Common forms of machine-readable media include, for example, a floppy disk, a flexible disk, hard disk, magnetic tape, or any other magnetic medium, a CD-ROM, any other optical medium, punchcards, papertape, any other physical medium with patterns of holes, a RAM, a PROM, and EPROM, a FLASH-EPROM, any other memory chip or cartridge, a carrier wave as described hereinafter, or any other medium from which a computer can read.
Various forms of machine-readable media may be involved in carrying one or more sequences of one or more instructions to processor 504 for execution. For example, the instructions may initially be carried on a magnetic disk of a remote computer. The remote computer can load the instructions into its dynamic memory and send the instructions over a telephone line using a modem. A modem local to computer system 500 can receive the data on the telephone line and use an infra-red transmitter to convert the data to an infra-red signal. An infra-red detector can receive the data carried in the infra-red signal and appropriate circuitry can place the data on bus 502. Bus 502 carries the data to main memory 506, from which processor 504 retrieves and executes the instructions. The instructions received by main memory 506 may optionally be stored on storage device 510 either before or after execution by processor 504.
Computer system 500 also includes a communication interface 518 coupled to bus 502. Communication interface 518 provides a two-way data communication coupling to a network link 520 that is connected to a local network 522. For example, communication interface 518 may be an integrated services digital network (ISDN) card or a modem to provide a data communication connection to a corresponding type of telephone line. As another example, communication interface 518 may be a local area network (LAN) card to provide a data communication connection to a compatible LAN. Wireless links may also be implemented. In any such implementation, communication interface 518 sends and receives electrical, electromagnetic or optical signals that carry digital data streams representing various types of information.
Network link 520 typically provides data communication through-one or more networks to other data devices. For example, network link 520 may provide a connection through local network 522 to a host computer 524 or to data equipment operated by an Internet Service Provider (ISP) 526. ISP 526 in turn provides data communication services through the world wide packet data communication network now commonly referred to as the “Internet” 528. Local network 522 and Internet 528 both use electrical, electromagnetic or optical signals that carry digital data streams. The signals through the various networks and the signals on network link 520 and through communication interface 518, which carry the digital data to and from computer system 500, are exemplary forms of carrier waves transporting the information.
Computer system 500 can send messages and receive data, including program code, through the network(s), network link 520 and communication interface 518. In the Internet example, a server 530 might transmit a requested code for an application program through Internet 528, ISP 526, local network 522 and communication interface 518.
The received code may be executed by processor 504 as it is received, and/or stored in storage device 510, or other non-volatile storage for later execution. In this manner, computer system 500 may obtain application code in the form of a carrier wave.
In the foregoing specification, embodiments of the invention have been described with reference to numerous specific details that may vary from implementation to implementation. Thus, the sole and exclusive indicator of what is the invention, and is intended by the applicants to be the invention, is the set of claims that issue from this application, in the specific form in which such claims issue, including any subsequent correction. Any definitions expressly set forth herein for terms contained in such claims shall govern the meaning of such terms as used in the claims. Hence, no limitation, element, property, feature, advantage or attribute that is not expressly recited in a claim should limit the scope of such claim in any way. The specification and drawings are, accordingly, to be regarded in an illustrative rather than a restrictive sense.
The present application is related to the following U.S. Patent Application, the entire contents of which are incorporated herein by reference for all purposes: U.S. patent application Ser. No. ______, filed on equal day herewith, entitled HIGH AVAILABILITY FOR DISTRIBUTED NON-PERSISTENT EVENT PROPAGATION, by Ramkumar Venkatesan, Krishnan Meiyyappan, Alan Downing and John W. Stamos (Attorney Docket No. 50277-2801).
