Method for high speed control over blocks for instant copying in data storage systems with joint usage of memory by n units

FIELD: engineering of computers for controlling memory, in particular, external memory controllers.

SUBSTANCE: memory control device for operation in memory controller network contains memory controller being an owner unit, capable of controlling the blocking of certain data area during execution of input-output outputs, and component for exchanging messages, providing for transmission of at least one message with blocking request, permission of blocking, blocking removal request and blocking removal signal, and also input-output component, while any image of aforementioned data area, received by instant copying thereof, is maintained as coherent relatively to data area itself, and input-output component may position previous direct confirmation, that this data area remains coherent to any such image, to cash-memory, and may perform input-output operations on basis of aforementioned previous direct confirmation. Method describes operation of aforementioned device. Software product for computer is realized on machine-readable carrier and contains a program recorded thereon, realizing operations of aforementioned method.

EFFECT: expanded functional capabilities.

3 cl, 3 dwg

 

The technical field to which the invention relates.

The present invention relates to computing devices for memory management, in particular to the storage controllers with advanced features in storage systems with shared memory n nodes with instant copy of the data.

The level of technology

In the field of computer data storage systems become more and more popular so-called "advanced features". These functions go beyond simple control I / o implemented conventional systems of memory controllers. Advanced features are widely known and are based on management metadata that is used to store information about the state of the real, or "custom" data stored in the system. Advanced functions allow you to quickly perform various actions with virtual images, or snapshots of data, while the actual data remains available for use by user applications. One such well-known advanced features is instant, or snapshot, copying the data.

At the highest level instant copy is a function that creates a secondary image of some data. This feature is sometimes referred to as the copy of the data at a control point, or t0 to the Pia. The contents of the secondary original image is identical to the primary content. The secondary image is created "instantly". In practice, this means that the secondary image is created for much less time than would be required to create this separate physical copies, and that there is no unwanted disruption to programs that use the data copied.

The thus created a secondary copy can be used for various purposes, including backup, scanning and extracting information from data. The primary copy is still used by the original application for its original purpose. In contrast to this function to create a backup without instant copy, you must first close the application, make a backup copy, and then you restart the application. Find time Windows, when used you can exit the application without prejudice to the workflow, it is becoming increasingly difficult. Thus, the backup costs increase. For enterprises the ability to create backups using instant copy of data without stopping the workflow has a significant and growing value.

Solutions that implement the instant backup is created and Luzia the existence of a data image, what is achieved by redirecting read requests, addressable secondary image data (hereinafter recipient), the original image (hereinafter - the source), if only in the data region does not write. When a specific data region (source or destination) must be made in the record, to preserve the illusion that the source and destination have their own copy of the data, it triggers a process of suspending the execution of the write command, and while the record is not held, issued the command to read a rewritable data area of the source, the read data are written to the destination, and then (and only after successful execution of all phases) occurs unlock write operations. At the subsequent write commands in the same area, their execution can not pause, because the recipient already has its own copy of the data. This technique is copy-on-write is well known and is used in many operating environments.

The basis of all decisions, implementing the function of a snapshot is the data structure that manages the adoption of the above solutions, namely decisions about sent if received by the addressee read requests to the source or the destination, and regarding whether to suspend the write operation for the implementation of copying the Oia when recording. The above data structure is essentially monitors the areas or portions of the data copied from the source to the destination, in contrast to data that were not copied.

Maintain a data structure (hereinafter referred to as metadata) is key to the implementation of the algorithm underlying the instant copy operation.

Instant backup is relatively easy to implement in a single processor complex (possibly with symmetric multiprocessing modules), as it is often used in modern memory controllers. The price is slightly more effort, you can implement failover instant copy, when access to copies of the metadata are two (at least two) of the processor complex. In case of failure of the first complex to continue can be used in the second set without losing access to the destination.

However, the performance of a single processor complex operations are I / o limited. Even with the increase of the performance of one processor complex, expressed in number of disk I / o per second or throughput (Mbit/s), it has a finite limit, which ultimately imposes a limitation on system performance when working with applications. This limit exists in many applications, technologies MH the public copy, but the most obvious example is the memory controller. A typical memory controller has one processor complex (or perhaps a pair of overlapping processor complexes), which defines the limit of the performance of such a controller.

You can install additional memory controllers. But such a separate memory controllers do not have joint access to the metadata and, therefore, do not interact in the management of instant image data. Memory becomes fragmented, and the scope of the instant copy is limited to the framework of a single controller. The source disk and the disk-the recipient must be running the same memory controller. Disk space is one of such a memory controller may be filled while another disk controller may be free space, but the source disk and the disk addressee cannot be separated by passing the disc to the destination under the control of the other controller. (This is particularly unfortunate in the case of the new instant copy when moving the addressee does not require significant resources, because the addressee is not mapped to any physical data.)

In addition to limiting the potential effectiveness of a pair of "source-destination", limited storage features Yes the data with a single controller additionally complicate the administration of the means of data storage.

In modern systems memory management usually does not attempt to solve the problem. These systems implement the methods of instant copy, allow you to use only one controller, and therefore, their functionality is limited by the performance of this controller.

A simple way to enable multiple controllers sharing instant copy is that one of the controllers is assigned ownership of the metadata, while the other domain controllers should send him all requests read and write. Using the above algorithm, owns the data controller handles requests I / o as if they were received directly from the fixed host servers, and returns each executed query I / o sent it to the controller.

The main drawback of this system and the constraint distribution, is that the cost of sending each request I / o so big that you can even double the consumption of resources at the level of the whole system and, therefore, about twice reduce the system performance.

It is known, for example, in the field of distributed database systems with means of parallel data processing using patterns races is a certain lock control, which Protocol is used two-phase locking to lock data to ensure the coherence of any copies of the data. However, two-phase locking is typically very time consuming and leads to reduced system performance due to the necessity of messaging. Known from the prior art Protocol two-phase locking as such, in its initial view, it is impractical to use in systems of lower level hardware-software complex, such as a storage area network with distributed memory controllers, in which the effect of the passage of governors lock messages on network performance is even more important than at the level of database management.

Thus, it is desirable to use the advantages of the distributed lock management in a computing environment implementing the instant copy with minimal computing resources on the transmission locking messages.

Summary of the invention

To solve the above problem the present invention proposes a control device memory network controller memory, including the memory controller, which is the site-owner is able to control the blocking of a specific area Yes the data when performing I / o, and a component for messaging transmitting at least one message with the lock request, the lock resolution, request the unlock signal unlock containing the component I / o, providing operations input-output relation data belonging to any unit owner, provided that this component's locking protocols managed by the site owner, and any image of the specified area data obtained by an instantaneous copy of the latter, supported by coherent with respect to the data region, and the I / o can be placed in the cache memory of the previous direct confirmation that this data area remained coherent in any such manner, and to perform I / o on the basis of this previous direct confirmation.

Preferably, component I / o could drop placed in the cache memory direct evidence of coherence and subsequently again to request the lock.

By such selective drop placed in the cache memory direct confirmation of coherence can be managed in a limited area of the cache memory.

Preferably also, that direct evidence of coherence in relation to one area of the data so the e was a direct confirmation of coherence in relation to other adjacent to it, the field data.

The object of the present invention is also a method for managing memory in a network storage controller, including a memory controller, which is the site-owner is able to control the blocking of a specific data region when performing I / o, and a component for messaging transmitting at least one message with the lock request, the lock resolution, request the unlock signal unlock, namely, that perform I / o with respect to the data belonging to any unit owner, provided that the locking protocols managed by the site owner, support image data region, obtained by instant copy last coherent with respect to the data area, and placed in the cache memory of the previous direct evidence that the specified data area remains coherent in any such way, and the operation of I / o performed on the basis of this previous direct confirmation.

In the implementation proposed in the invention method, it is preferable to discard the cached memory direct confirmation of coherence with subsequent re-lock request.

In this case, by election Ambra is ivania placed in the cache memory direct confirmation of coherence can be controlled in a limited area of the cache memory.

Preferably also, that direct evidence of coherence in relation to one area of the data was a direct confirmation of coherence in relation to other, related, data region.

Another object of the invention is a software product for PC, material implemented in a machine-readable medium containing a program which when loaded into a computing system and run it controls the control unit memory network memory controllers, which includes the controller, which is the owning node data managing block data region when performing I / o, and a component for messaging transmitting at least one message with the lock request, the lock resolution, request the unlock signal unlock with software execution device memory management the following actions: perform I / o-output relation data belonging to any unit owner, provided that the locking protocols managed by the site owner; maintaining the image data region, obtained by instant copy last coherent with respect to the data region; space in the cache memory of the previous direct evidence that the decree of the cm data area remains coherent in any such way, the operations of I / o performed on the basis of this previous direct confirmation.

In a preferred embodiment, the present invention uses a message exchange pattern for a two-phase locking, which allows to coordinate the work of several memory controllers (or nodes) in a system with n nodes (n is a third-party system). Messaging provides the coordination of work of system nodes, but each node is still responsible for their own operations I / o. Each node in the network has a means of caching to cache the results of previous queries lock that allows this node to reuse some of the results, without the need of queries regarding the status of a particular area data, if such condition has been directly reflected previous results placed in the cache.

Brief description of drawings

Further, in the example described the preferred implementation of the present invention with reference to the attached drawings on which is shown:

figure 1 is a block diagram illustrating a variant of the scheme of two-phase blocking using blocking messages to control the coherence of the data region and its image received instant copy,

on Fig - the components proposed in the invention of the system in its preferred version,

figure 3 - additional operations to be performed when carrying out the invention in its preferred form.

A detailed description of the preferred option implementation

For a better understanding of the preferred alternative implementation of the present invention should be considered messaging for two-phase locking to coordinate the operation of multiple memory controllers (or nodes) in the data storage system having n nodes (n-way system).

As an example, let us consider a system with n nodes, in which the function of an instantaneous copy of the data region. Assume that each node has access to memory that is managed by a group of n interacting nodes. In step 102 one of the nodes is assigned ownership of the metadata relating to all relationships input-output in a particular field of data. The other nodes are assigned to clients. In addition, in the most preferred at present embodiment of the invention one of the nodes client identifies the owner and maintains a copy of the metadata, in the case of a node failure, the owner provides continuous availability.

Consider the case when the request is complete I / o, which in step 104 is transported is t from the host system (from the English. "host" - in this case a stand-alone computer, such as a server or workstation) in a specific node, the client C. it is Assumed that the request I / o from the host system refers to reading or writing on the disk, the destination, or perhaps a write operation on the source drive. Node client begins processing data by pausing execution of requests I / o at step 106. Then in step 108, the host client sends to the node owner Of the REQ message with the question, copied whether the corresponding fragment data scope.

After receipt of the request REQ to the owner node checks On the structure of the native metadata. If the site owner About determines that the data area is already copied, at step 110 the owner node Of passes in the response message NACK (negative acknowledgement). If the site owner About determines that the data area has not yet been copied, it is placed opposite the corresponding metadata related to the above-mentioned data area and its own metadata structures, locking the account, and in step 112 will respond to the source node to the client, sending the message GNT (lock request is made). Locking the record is necessary for compatibility between just accepted and executed by the incoming request and subsequently requests that may affect the same metadata as the process continued the I data processing node, the client C. Procedure maintain the locking entries in the power and limitations of compatibility similar procedure and limitations in the case, if the query is I / o was obtained locally by the node On that is well known to specialists in this field of technology.

After receiving the NACK message, the node client at step 114 resumes the suspended execution of the original query I / o.

After receiving the message GNT node client at step 116 continues to work, performing one or more data transfer operations required by the algorithm instant copy. In the case of read operations of the recipient, this means the implementation of reading in relation to the original disk. Some time later, in step 118, the host client has finished processing the request to read, and at step 120 transmits to the node Of the message UNL (lock capture) to simultaneously send messages about the processing of the initial request I / o the host that sent the query.

After receiving the message UNL site owner About to step 122 removes the locking entry from its table metadata, thereby resuming the processing of other requests I / o, the execution of which was suspended because of this lockout. In this most preferred embodiment of the invention in step 124, the site was owned by the C O node transmits to the client With the message UNLD (block removed), allowing the host to the client to reuse the resources specified in the original request. However, the algorithm itself snapshot does not require it.

In the case of a write operation (destination or source) node client needs to step 127 to perform copy-on-write. After completion of all steps copy-on-write and as long as the original write request I / o is suspended, at step 126 the node client node transmits On the message UNLC (request unlock copy).

After receiving the message UNLC site owner About to step 128 marks in the metadata corresponding area as copied in step 130 releases the write lock on the step 132 is responsible for all pending requests, indicating that the data area is copied, and then at step 134 passes the node to the client With the message UNLD.

After receiving the message UNLD node client at step 136 resumes a suspended write operation that is performed after a certain time, and then at step 138 informs the host system about the completion of the write operation.

In case of error I / o on disk failure messaging system or node failure required the restoration of data, however, the requirements for such paths and their implementation are well known from the technique.

The above operation is described with respect to one request I / o is from the point of view of one of a client node C. However, it questions the action described scheme when there are many requests I / o coming from multiple nodes clients that site owner About processes using the above algorithm.

Figure 2 shows proposed in the invention device in preferred at present embodiment, execution, implemented in the network memory controllers, including the owner node 202, the node-client 204, the component performing the operation, I / o, patch metadata 206 related to data 208, which manages a network of memory controllers, a copy of the 209 data 208 and communications. The device has component 210 that controls the distribution of ownership of data and fixing at site 202 owner rights metadata, and component 212 for lock control, able to manage a lock on the metadata layer 206 in the operations of the I / o to ensure their coherence with any copy 209. The device also includes a component 214 for messaging, which is located at the owning node 202 and is used to exchange between the host client 204 and the owning node 202 by one or more request messages about the state of the metadata, the permission to install a lock request lock, and signal lock. Node-client 204 performs the surgery the input-output data, metadata which are owned by the owner node 202, subject to the execution node, the client 204 of the locking protocols at the level of metadata managed by such owner node 202.

The above described system and method allow for distributed lock management in containing n nodes in the network controller shared memory, however, the disadvantage of this system are significant overhead costs associated with the messaging system. In systems with a relatively small number of controllers or a relatively small activity such disadvantage is not so significant, which, however, cannot be said about modern data storage systems, such as a large network of data storage devices with a large number of controllers and a very high activity of memory accesses. In such conditions it is advisable to eliminate unwanted messaging.

Therefore, to improve the process of using the system for processing data in the most preferred embodiment of the present invention for each node client provides the ability to store information in which is recorded the last response received from the site owner. In particular (see figure 3, supplementary to the block diagram in figure 1), node-enabled client at step 308 to place in the cache memory of the details is teaching them the NACK message after step 114, shown in figure 1, or on the transfer and to confirm them a couple of messages UNLC/UNLD at step 126 and after step 134, shown in figure 1.

After receiving from the host system in step 302 request I / o, similar to the one shown in figure 1 step 104, the node client in this case uses described later modified control algorithm locks.

First, in step 303, the node client checks whether it is placed in the cache memory, data indicating that the affected region is data already copied. If so, then in step 304, the node client continues processing the request I / o, not passing the site Of Protocol messages.

If the cache memory does not specify, use the above Protocol without modification. Node client performs step 106 and the subsequent steps shown in figure 1. In the case of receiving in step 306, the NACK message or a couple of messages UNLC/UNLD in step 308 is updated in the cache information, and once detected in step 303 of such information in the cache memory, the processing subsequent requests I / o, which affect this area may continue at step 304 without transfer Protocol messages.

When describing the technology necessary for the operation of the present invention in its most preferred at this time variant sometimes use the term "pessimistic the cue cache". This means that the node client may not have a complete updated copy of the metadata of the site owner: site-the client may assume that a certain memory area needs to be copied, and the site owner will correct it by notifying the NACK message that copying is not required. But the site-the client should never assume that a certain area is copied, if the site owner is known to the contrary.

For proper functioning of the system for caching information about locks in accordance with the preferred embodiment of the invention in the operation of a client node you want to make some changes. First, each time at the beginning of the interaction to create instant copies (in step 300A) in step 301 it is necessary to initialize the cache memory (to indicate that you want to copy all fields). This can be done in various ways, but the most simple is the message the site owner node client. Secondly, each time a node is the client in step 300b could miss the message that the cache memory has been initialized again (possibly due to disturbances in the power system), the host customer must assume the worst and at step 301 re-initialize its cache memory or to double-check the correctness of the data contained in it.

In the proposed invention the method can be made is by further additions and changes, obvious to a person skilled in the art. For example, the cached information can be discarded, i.e. not taken into account, since it can always be recovered by using the site-owner, which contains the only and true with the updated copy. Thus, in the node client for caching information may be allocated a smaller portion allocated to the metadata space than would be required to store all metadata available in all network nodes. In this case, the nodes clients could rely on the location of access in respect of processed their operations, I / o, to be able to continue to use the caching of the information contained in messages about locks.

In another embodiment of the invention with extended functionality the NACK message (or messages GNT or UNLD) may contain additional information beyond that which relates directly to the field, processed using message REQ/GNT/UNLC/UNLD. Nodes owners can also be transferred to the nodes client information relating to neighboring areas, which were also cleared.

A typical embodiment of the above method is implemented in software running on one or more processors (not shown), and this software obespechennoy be an element of a computer program, recorded on a suitable data carrier (also not shown)such as magnetic or optical disk. Similarly, the data channels may include any kind of storage medium for the information and the medium of signal transmission, such as a wired or wireless transmission channels of signals.

The present invention can be appropriately implemented in the form of a software product for use in computing systems. The invention can be implemented as a sequence of machine-readable commands, stored on a tangible medium, such as computer-readable media, such as floppy disk, ROM, CD-ROM (CD-ROM), ROM or hard disk, or transmitted to a computer system via a modem or other interface devices using the material environment, including, but not limited to optical or analog communications lines, or intangible way with the use of wireless technologies, including (and not only) microwave technology, infrared or other signaling technology. In the sequence of machine-readable commands partially or fully implemented the above functionality.

For professionals it is clear that such machine-readable commands can be written in several programming languages, applicable in a variety of computer architectures is whether operating systems. In addition, these commands can be stored in any storage technology, existing or future, including (and not only) solid-state, magnetic or optical technology, or transmitted using any communications technology, existing or future, including (and not only) optical, infrared or microwave equipment. It is assumed that such software for computers may be distributed as a removable media with accompanying documentation in printed or electronic form, for example, in the form of closed software (software package without revealing the internal structure), pre-loaded in the computing system, e.g., on system ROM or fixed disk, or distributed from a server or electronic Bulletin Board over a network, such as the Internet or world wide web.

For professionals it is obvious that in the above-described variant of implementation of the present invention can be made various changes.

1. Device control memory network controller memory, including the memory controller, which is the site-owner is able to control the blocking of a specific data region when performing I / o, and a component for messaging transmitting at least one message with C the millet lock the lock resolution, request the unlock signal unlock containing the component I / o, providing operations input-output relation data belonging to any unit owner, provided that this component's locking protocols managed by the site owner, and any image of the specified area data obtained by an instantaneous copy of the latter, supported by coherent with respect to the data region, and the I / o can be placed in the cache memory of the previous direct confirmation that this data area remained coherent in any such manner, and to perform I / o based on this previous direct confirmation.

2. The device according to claim 1, in which component I / o can be placed in the discard cache memory direct evidence of coherence and subsequently again to request the lock.

3. The device according to claim 2, in which through selective drop placed in the cache memory direct confirmation coherence control in a limited area of the cache memory.

4. The device according to claim 1, in which direct evidence of coherence in relation to one data region is a direct confirmation of coherence in relation to other, related, data region.

5. Method for managing memory in a network storage controller, including a memory controller, which is the site-owner is able to control the blocking of a specific data region when performing I / o, and a component for messaging transmitting at least one message with the lock request, the lock resolution, request the unlock signal unlock, namely, that

perform I / o with respect to the data belonging to any unit owner, provided that the locking protocols managed by the site owner,

support image region data, obtained by instant copy last coherent with respect to the data region and

placed in the cache memory of the previous direct evidence that the specified data area remains coherent in any such way, and the operation of I / o performed on the basis of this previous direct confirmation.

6. The method according to claim 5, in which is placed in the cache memory direct confirmation of coherence drop out and re-lock request.

7. The method according to claim 6, in which through selective drop placed in the cache memory direct confirmation of coherence exercise control is of limited area of the cache memory.

8. The method according to claim 5, in which direct evidence of coherence in relation to one data region is a direct confirmation of coherence in relation to other, related, data region.

9. Software product for PC, material implemented in a machine-readable medium containing a program which when loaded into a computing system and run it controls the control unit memory network memory controllers, which includes the controller, which is the owning node data managing block data region when performing I / o, and a component for messaging transmitting at least one message with the lock request, the lock resolution, request the unlock signal unlock with software execution device memory management the following actions:

the operations of the input / output relation data belonging to any unit owner, provided that the locking protocols managed by the site owner,

maintaining image region data, obtained by instant copy last coherent with respect to the data region,

the location in the cache memory of the previous direct evidence that the specified area is given the s remains coherent in any such way, the operations of I / o performed on the basis of this previous direct confirmation.



 

Same patents:

FIELD: computer engineering, possible use in data exchange systems and local computing networks.

SUBSTANCE: device contains N≥2 client blocks, clock impulse generator, N client time controllers, OR element, AND-NOT element, selector-multiplexer, two N-input AND-NOT elements, two priority encoders, main wait time controller.

EFFECT: increased probability of timely servicing of clients under conditions of real functioning process of data exchange systems, with continuous dynamics of change of modes of different priority requests from clients.

4 cl, 7 dwg

FIELD: engineering of interrupt processing mechanisms in computer systems.

SUBSTANCE: system contains processor with multiple contexts for execution of commands stored in memory. In response to common interrupt logical processors of processor with multiple contexts compete for receiving access to jointly utilized register. First logical processor gaining access to aforementioned jointly utilized register processes common interrupt. Remaining logical processors return from interrupt.

EFFECT: increased productiveness of system.

4 cl, 5 dwg

FIELD: engineering of information processing systems.

SUBSTANCE: system contains master-system for processing information, interface, central communication device, client system for processing information, object model. In accordance to method each master system sends to central communication device elements of its data array determined in appropriate master-representation, while in master-representation of connected master system elements of data array are contained, for which system has data priority.

EFFECT: simplified specification and development of interfaces between technical applications.

2 cl, 6 dwg

FIELD: method and device for processing data for preserving recurrent status in data processing device.

SUBSTANCE: device has data processing block, having multiple functioning modes, for each of which special memory stack is present. Method describes operation of this device. Data carrier contains program, configuring data processing device for performing stages of method.

EFFECT: decreased size of code and decreased interruption processing delay.

3 cl, 16 dwg

FIELD: computer science.

SUBSTANCE: device has n-byte query register, query limits location systems, each of which consists of counting timer and OR element, OR element, AND element, keys cascade.

EFFECT: higher reliability and speed of operation.

1 dwg

The invention relates to computing and is used in automatic control systems of technological processes

The invention relates to computing, and in particular to a device priority, and can be used to control access to multiple subscribers to the shared resource

The invention relates to the field of computer technology, in particular to a device priority service requests

The invention relates to systems and methods for displaying information on remote computers

FIELD: computer science.

SUBSTANCE: device has n-byte query register, query limits location systems, each of which consists of counting timer and OR element, OR element, AND element, keys cascade.

EFFECT: higher reliability and speed of operation.

1 dwg

FIELD: method and device for processing data for preserving recurrent status in data processing device.

SUBSTANCE: device has data processing block, having multiple functioning modes, for each of which special memory stack is present. Method describes operation of this device. Data carrier contains program, configuring data processing device for performing stages of method.

EFFECT: decreased size of code and decreased interruption processing delay.

3 cl, 16 dwg

FIELD: engineering of information processing systems.

SUBSTANCE: system contains master-system for processing information, interface, central communication device, client system for processing information, object model. In accordance to method each master system sends to central communication device elements of its data array determined in appropriate master-representation, while in master-representation of connected master system elements of data array are contained, for which system has data priority.

EFFECT: simplified specification and development of interfaces between technical applications.

2 cl, 6 dwg

FIELD: engineering of interrupt processing mechanisms in computer systems.

SUBSTANCE: system contains processor with multiple contexts for execution of commands stored in memory. In response to common interrupt logical processors of processor with multiple contexts compete for receiving access to jointly utilized register. First logical processor gaining access to aforementioned jointly utilized register processes common interrupt. Remaining logical processors return from interrupt.

EFFECT: increased productiveness of system.

4 cl, 5 dwg

FIELD: computer engineering, possible use in data exchange systems and local computing networks.

SUBSTANCE: device contains N≥2 client blocks, clock impulse generator, N client time controllers, OR element, AND-NOT element, selector-multiplexer, two N-input AND-NOT elements, two priority encoders, main wait time controller.

EFFECT: increased probability of timely servicing of clients under conditions of real functioning process of data exchange systems, with continuous dynamics of change of modes of different priority requests from clients.

4 cl, 7 dwg

FIELD: engineering of computers for controlling memory, in particular, external memory controllers.

SUBSTANCE: memory control device for operation in memory controller network contains memory controller being an owner unit, capable of controlling the blocking of certain data area during execution of input-output outputs, and component for exchanging messages, providing for transmission of at least one message with blocking request, permission of blocking, blocking removal request and blocking removal signal, and also input-output component, while any image of aforementioned data area, received by instant copying thereof, is maintained as coherent relatively to data area itself, and input-output component may position previous direct confirmation, that this data area remains coherent to any such image, to cash-memory, and may perform input-output operations on basis of aforementioned previous direct confirmation. Method describes operation of aforementioned device. Software product for computer is realized on machine-readable carrier and contains a program recorded thereon, realizing operations of aforementioned method.

EFFECT: expanded functional capabilities.

3 cl, 3 dwg

FIELD: engineering of means for pausing execution of a stream until certain memory access occurs.

SUBSTANCE: in one variant of realization, processor contains a set of executive devices, capable of executing a set of streams. First stream includes a command, which determines the address being tracked. Logical pausing means pause execution of first stream, and monitor causes renewal of first flow as reaction to access of given address being tracked.

EFFECT: increased processor productiveness.

5 cl, 14 dwg

FIELD: methods for automatic execution of a program, connected to data file, when data file and program being executed are positioned on different computer units.

SUBSTANCE: in methods, program being executed is accessed through graphic image of data file type, realized in the network, which includes client system and a set of server systems. Client system receives the scheme, which determines connection between the set of programs being executed and corresponding set of data file types. Graphic image of data files is displayed, information about selection of graphic image of data file is received from server system, on basis of it program to be executed is selected and executed.

EFFECT: increased productivity of system due to distributed execution of programs.

9 cl, 19 dwg, 3 tbl

FIELD: method and system for providing user interface information to client.

SUBSTANCE: in accordance to the invention, access system contains registration mechanism. Client environment for automatic processing of user interface receives registration information from the client and transmits user interface information after receipt. Server for automatic processing of user interface receives registration information from client environment for automatic processing of user interface and notifies processor of user interface about registration, and also receives user interface information from user interface processor. The server contains filtration device for filtering out information of no interest to client, and notification device for notifying the client about information which is of interest to the client.

EFFECT: ensured capacity for filtration and coordination of excessive and disorienting notifications.

2 cl, 11 dwg

FIELD: telecommunications.

SUBSTANCE: device contains a set of central processor units, which are assigned a common external up address in telecommunication network which allows packet data. IP messages, addressed to a network element, are received, and received IP messages which contain first messages are identified. First value is identified in first message and first message is transmitted to central processor unit on basis of identified first value, if identified first value is not equal to zero.

EFFECT: ensured load balancing for central processor when using several types of traffic.

3 cl, 3 dwg

Up!