CN101551745A - Method for greatly improving performance of workflow engine - Google Patents
Method for greatly improving performance of workflow engine Download PDFInfo
- Publication number
- CN101551745A CN101551745A CNA2009100155162A CN200910015516A CN101551745A CN 101551745 A CN101551745 A CN 101551745A CN A2009100155162 A CNA2009100155162 A CN A2009100155162A CN 200910015516 A CN200910015516 A CN 200910015516A CN 101551745 A CN101551745 A CN 101551745A
- Authority
- CN
- China
- Prior art keywords
- workflow engine
- cache
- data
- engine service
- cache object
- 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.)
- Pending
Links
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Information Transfer Between Computers (AREA)
Abstract
The invention relates to a method for greatly improving the performance of a workflow engine, which pertains to the filed of middleware technical framework, in particular to the field of the technical framework of the workflow engine. The method improves the performance of the workflow engine from the technical framework of the workflow engine and comprises the steps of server-side caching, client-side caching and the like; the method adopts aching technologies, object pool technologies and an asynchronous call way to reduce system expense, greatly improve the performance of the workflow engine, reduce the fund input of the whole business flow system, support the business flow system with the same data volume and highly concurrent access, achieve the same performance effect and reduce cluster nodes of certain number or lower the hardware configuration of cluster nodes when clustering is executed.
Description
Technical field
The present invention relates to a kind of method that increases substantially performance of workflow engine, this method belongs to middleware Technology framework field, particularly in the Technical Architecture field of workflow engine.
Background technology
In the last few years, along with the continuous development of informatization, the application of workflow was more and more universal, and each business process system is managed more clearly and be flexible; Simultaneously, business process system is also more and more higher to the requirement of workflow, basic application can't satisfy the demand of business process system, and business process system is more and more stronger to senior application demands such as the real-time monitoring analysis of business activity, flow data statistical study.
The real-time monitoring analysis requirement of business activity, professional real-time monitoring and optimization; According to fewer data volume, carry out real-time analysis and handle, be mainly used in bookkeeping.
Flow data statistical study requirement, the statistical study of historical data; According to many data, carry out batch quantity analysis and handle, be mainly used in planning.
Present most of working flow products still rests on the basic application that only satisfies business process system; Working flow products on this aspect when relating to the flow process of artificial participation, in order to tackle the accident burst accidents such as machine of delaying, and provides functions such as service data and the inquiry of historical data, and flow data need be persisted in the database; Relate to data base persistenceization, under big visit capacity, high concurrent situation, run into the problem of database I/O bottleneck through regular meeting; Working flow products has determined the supporting capacity of this working flow products to business process system to the degree that database I/O bottleneck solves.
Some working flow products with its product orientation on the senior application, set about satisfying functional requirements such as the real-time monitoring analysis of business activity, flow data statistical study; User's statistical study flow data must be persisted to all flow data in the database for convenience; Monitoring analysis, big data quantity statistical study will bring higher challenge to its working flow products in real time.
Working flow products makes things convenient for the business process system management all to the platform product development at present, like this workflow platform is supported that the requirement of remote access capability is more and more higher; This just brings the problem of another one reality, network I/O bottleneck; How to solve network I/O bottleneck, also determine the supporting capacity of workflow platform business process system.
More than analyzed some practical problemss that working flow products runs into from local I/O and network I/O aspect, work at present miscarriage product promote the supporting capacity to business process system generally from setting up aspects solution deficiencies separately such as suitable data storehouse index, upgrading hardware, optimization logic.
Also having the most direct method is exactly cluster, application server cluster and data-base cluster; Certainly this is unusual effective method, also is must do towards big visit capacity and high concurrent system.How further to improve the performance of workflow engine on this basis, just this invents described emphasis.
Work at present miscarriage product, it no matter is the function on the basic application level, for example: Pending tasks inquiry, the task of handling are inquired about, can be recovered task inquiry, specified requirements inquiry etc., or the function on the senior application, for example: the real-time monitoring analysis of business activity, flow data statistical study etc., all need the frequent visit workflow engine, engine internal need be carried out the lot of data inquiry continually simultaneously, and the function of these each aspects will inevitably be brought huge pressure to workflow engine; Refer more particularly to the flow process of artificial participation, Processing tasks must check out the business that needs processing earlier, and then carry out corresponding service processing, business process system is very frequent to calling with workflow engine intrinsic call of workflow engine like this, directly causes whole workflow engine very limited to the business process system supporting capacity; If workflow engine is handled the statistical study of some monitoring in real time and big data quantity again, will influence workflow engine more to the business process system supporting capacity.If still only pace up and down setting up suitable data storehouse index, upgrading hardware, optimization logic and Clustering, will inevitably increase the fund input of whole service flow system greatly.
Summary of the invention
Purpose of the present invention is exactly at above-mentioned deficiency, from the workflow technology framework, promotes performance of workflow engine, and a kind of method that increases substantially performance of workflow engine is provided.
The method that increases substantially performance of workflow engine provided by the invention: comprise the steps:
The 1st step, the server end buffer memory, promptly finish the frequency that reduces on the same function basis the computer hardware read and write, business process system is optimized the frequent data query of work engine, at the peripheral packing of workflow engine one deck cache layer, data are directly obtained in each visit from buffer memory, reduce database I/O is operated.
When workflow engine starts, will need the data cached initialization of carrying out; Flow data with producing in real time is cached in the memory object, avoids the new flow data that produces in the Query Database, and memory object is managed.
The 2nd step, client-cache is promptly finished on the same function basis request number of times that reduces server, carries out buffer memory from both direction: flow data buffer memory and represent a layer page cache, minimizing promptly reduces network layer I/O to the call number of workflow engine service.
The metadata cache object focuses on those often operation and big data of data volume, and for example the flow definition data are promptly often operated, and data volume is big again; Server end and client Work stream data cache object are synchronous, and the server end data change notifies all client-side data cache objects to upgrade; Different clients Work stream data cache object is synchronous, during the data cached renewal of active client, at first notifies the workflow engine server, and server end reinforms other all client-side data cache objects and upgrades then; Represent a layer member, carry out buffer memory at those pages that seldom change.
In the 3rd step, object pool utilizes the pond technology, avoids the system overhead that brings is created, encapsulates, destroyed to cache object continually.
In the 4th step, asynchronous call at the specific demand of certain time period, promotes the batch processing ability.
The 5th step, the cluster support, under big visit capacity, the high concurrent situation, the workflow engine service of cluster is an effective support.
Above-mentioned the 3rd step, the 4th step, the 5th step are common prior aries.
Above-mentioned server end cache object management process is such:
When 1-1) the workflow engine service starts, the needs data in buffer is carried out initialization add in the internal memory;
When 1-2) workflow engine was served newly-generated flow data, new cache object was created and is increased in the internal memory;
When 1-3) the existing flow data of workflow engine service was upgraded, the corresponding cache object was updated in the internal memory;
When 1-4) workflow engine service deletion has flow data, deletion corresponding cache object in the internal memory.
Above-mentioned client-cache Object Management group process is such:
When 2-1) business process system starts, the flow definition in the workflow engine service is loaded in the internal memory cache object initialization;
2-2) business process system imports the new technological process definition in the workflow engine service, and workflow engine Service Notification client-cache increases the new flow definition that imports, and newly imports the flow definition cache object and is created and is increased in the internal memory;
2-3) business process system will have flow definition and be updated in the workflow engine service, the corresponding flow definition of workflow engine Service Notification client buffer update, and flow definition cache object corresponding in the internal memory is updated;
2-4) business process system will have flow definition when deleting from the workflow engine service, and workflow engine Service Notification client-cache is deleted corresponding flow definition, the corresponding flow definition cache object of deletion in the internal memory.
In server end buffer memory and client-cache implementation process, should be noted that problem:
1. which type of object of buffer memory? the caching server end is more invoked flow data often, for example: the tabulation of nominator's Pending tasks; Cache client frequent access server end, the data that data volume is big, for example: the flow definition object of client (relies on the flow definition masterplate during process flow operation, need to load definition information, especially in the process of remote service, flow definition information self is just bigger, transmit the flow definition information of this big data quantity, will inevitably cause network is taken in large quantities, especially the flow definition data need frequent visit, will inevitably cause the obstruction of network, reduce the access efficiency of total system greatly, along with the increasing performance meeting straight line decline of concurrency); Represent a layer member, carry out buffer memory at those pages that seldom change.
Is 2. how cache object managed? flow data with producing in real time is cached in the memory object, inquires about the flow data of these new generations when avoiding inquiring about this partial data more again from database, and establishment, renewal, the deletion of cache object managed.
3. cache object initialization when? when workflow engine starts, will need the data cached initialization of carrying out.
Is 4. how the server end Data Update synchronized to client-cache? the server end data change is notified all clients to carry out the metadata cache object and is upgraded.
5. how synchronous is the data cached renewal between different clients? during the data cached renewal of active client, at first notify the workflow engine server, the workflow engine service end reinforms other all clients and carries out the metadata cache object and upgrade then.
Like this, the present invention reduces system overhead by caching technology, object pool technology, asynchronous call mode, promotes performance of workflow engine significantly, reduces the fund input of whole service flow system; Support same big data quantity, the business process system of high concurrent visit, reach the identical impact of performance, will reduce the clustered node of some during cluster, perhaps reduce the hardware configuration of clustered node.
Description of drawings
Fig. 1 is a server end cache object management flow chart in the embodiment of the invention;
Fig. 2 is a client-cache Object Management group process flow diagram in the embodiment of the invention.
Embodiment
A kind of method that increases substantially performance of workflow engine mainly comprises the client and server end, comprises the steps:
The 1st step, the server end buffer memory, promptly finish the frequency that reduces on the same function basis the computer hardware read and write, business process system is optimized the frequent data query of work engine, at the peripheral packing of workflow engine one deck cache layer, data are directly obtained in each visit from buffer memory, reduce database I/O is operated.
When workflow engine starts, will need the data cached initialization of carrying out; Flow data with producing in real time is cached in the memory object, avoids the new flow data that produces in the Query Database, and memory object is managed.
The 2nd step, client-cache is promptly finished on the same function basis request number of times that reduces server, carries out buffer memory from both direction: flow data buffer memory and represent a layer page cache, minimizing promptly reduces network layer I/O to the call number of workflow engine service.
The metadata cache object focuses on those often operation and big data of data volume, and for example the flow definition data are promptly often operated, and data volume is big again; Server end and client Work stream data cache object are synchronous, and the server end data change notifies all client-side data cache objects to upgrade; Different clients Work stream data cache object is synchronous, during the data cached renewal of active client, at first notifies the workflow engine server, and server end reinforms other all client-side data cache objects and upgrades then; Represent a layer member, carry out buffer memory at those pages that seldom change.
In the 3rd step, object pool utilizes the pond technology, avoids the system overhead that brings is created, encapsulates, destroyed to cache object continually.
In the 4th step, asynchronous call at the specific demand of certain time period, promotes the batch processing ability.
The 5th step, the cluster support, under big visit capacity, the high concurrent situation, the workflow engine service of cluster is an effective support.
Show that as Fig. 1 above-mentioned server end cache object management process is such:
When 1-1) the workflow engine service starts, the needs data in buffer is carried out initialization add in the internal memory;
When 1-2) workflow engine was served newly-generated flow data, new cache object was created and is increased in the internal memory;
When 1-3) the existing flow data of workflow engine service was upgraded, the corresponding cache object was updated in the internal memory;
When 1-4) workflow engine service deletion has flow data, deletion corresponding cache object in the internal memory.
Show that as Fig. 2 above-mentioned client-cache Object Management group process is such:
When 2-1) business process system starts, the flow definition in the workflow engine service is loaded in the internal memory cache object initialization;
2-2) business process system imports the new technological process definition in the workflow engine service, and workflow engine Service Notification client-cache increases the new flow definition that imports, and newly imports the flow definition cache object and is created and is increased in the internal memory;
2-3) business process system will have flow definition and be updated in the workflow engine service, the corresponding flow definition of workflow engine Service Notification client buffer update, and flow definition cache object corresponding in the internal memory is updated;
2-4) business process system will have flow definition when deleting from the workflow engine service, and workflow engine Service Notification client-cache is deleted corresponding flow definition, the corresponding flow definition cache object of deletion in the internal memory.
Claims (3)
1. a method that increases substantially performance of workflow engine is characterized in that: comprise the steps:
The 1st step, the server end buffer memory, promptly finish the frequency that reduces on the same function basis the computer hardware read and write, business process system is optimized the frequent data query of work engine, at the peripheral packing of workflow engine one deck cache layer, data are directly obtained in each visit from buffer memory, reduce database I/O is operated;
The 2nd step, client-cache is promptly finished on the same function basis request number of times that reduces server, carries out buffer memory from both direction: flow data buffer memory and represent a layer page cache, minimizing promptly reduces network layer I/O to the call number of workflow engine service;
In the 3rd step, object pool utilizes the pond technology, avoids the system overhead that brings is created, encapsulates, destroyed to cache object continually;
In the 4th step, asynchronous call at the specific demand of certain time period, promotes the batch processing ability;
The 5th step, the cluster support, under big visit capacity, the high concurrent situation, the workflow engine service of cluster is an effective support.
2. the method that increases substantially performance of workflow engine according to claim 1 is characterized in that: described server end cache object management process is such:
When 1-1) the workflow engine service starts, the needs data in buffer is carried out initialization add in the internal memory;
When 1-2) workflow engine was served newly-generated flow data, new cache object was created and is increased in the internal memory;
When 1-3) the existing flow data of workflow engine service was upgraded, the corresponding cache object was updated in the internal memory;
When 1-4) workflow engine service deletion has flow data, deletion corresponding cache object in the internal memory.
3. the method that increases substantially performance of workflow engine according to claim 1 is characterized in that: described client-cache Object Management group process is such:
When 2-1) business process system starts, the flow definition in the workflow engine service is loaded in the internal memory cache object initialization;
2-2) business process system imports the new technological process definition in the workflow engine service, and workflow engine Service Notification client-cache increases the new flow definition that imports, and newly imports the flow definition cache object and is created and is increased in the internal memory;
2-3) business process system will have flow definition and be updated in the workflow engine service, the corresponding flow definition of workflow engine Service Notification client buffer update, and flow definition cache object corresponding in the internal memory is updated;
2-4) business process system will have flow definition when deleting from the workflow engine service, and workflow engine Service Notification client-cache is deleted corresponding flow definition, the corresponding flow definition cache object of deletion in the internal memory.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2009100155162A CN101551745A (en) | 2009-05-13 | 2009-05-13 | Method for greatly improving performance of workflow engine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2009100155162A CN101551745A (en) | 2009-05-13 | 2009-05-13 | Method for greatly improving performance of workflow engine |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101551745A true CN101551745A (en) | 2009-10-07 |
Family
ID=41155998
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2009100155162A Pending CN101551745A (en) | 2009-05-13 | 2009-05-13 | Method for greatly improving performance of workflow engine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101551745A (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102236679A (en) * | 2010-04-27 | 2011-11-09 | 杭州德昌隆信息技术有限公司 | Method and device for outputting workflow based on browser page |
| CN102394807A (en) * | 2011-08-23 | 2012-03-28 | 北京京北方信息技术有限公司 | System and method for decentralized scheduling of autonomous flow engine load balancing clusters |
| CN103064964A (en) * | 2012-12-29 | 2013-04-24 | 天津南大通用数据技术有限公司 | Connection method of data base supporting distributed type affairs |
| CN104471572A (en) * | 2012-07-12 | 2015-03-25 | 微软公司 | Progressive query computation using streaming architectures |
| CN104699411A (en) * | 2013-12-06 | 2015-06-10 | 北京慧正通软科技有限公司 | Technical method for improving performance of cache in workflow engine process instance |
| CN104735152A (en) * | 2015-03-30 | 2015-06-24 | 四川神琥科技有限公司 | Mail reading method based on network |
| CN104751359A (en) * | 2013-12-30 | 2015-07-01 | 中国银联股份有限公司 | System and method for payment and settlement |
| CN106210022A (en) * | 2016-06-29 | 2016-12-07 | 天涯社区网络科技股份有限公司 | A kind of system and method for processing forum's height concurrent data requests |
| CN106529917A (en) * | 2016-12-15 | 2017-03-22 | 平安科技（深圳）有限公司 | Workflow processing method and device |
| CN107657419A (en) * | 2016-07-25 | 2018-02-02 | 武汉票据交易中心有限公司 | The processing method and relevant apparatus and server of a kind of operation flow |
| US10552435B2 (en) | 2017-03-08 | 2020-02-04 | Microsoft Technology Licensing, Llc | Fast approximate results and slow precise results |
| US10740328B2 (en) | 2016-06-24 | 2020-08-11 | Microsoft Technology Licensing, Llc | Aggregate-query database system and processing |
| CN112926206A (en) * | 2021-02-25 | 2021-06-08 | 北京工业大学 | Workflow engine cache elimination method based on industrial process background |
| CN115857907A (en) * | 2023-02-06 | 2023-03-28 | 卓望数码技术（深圳）有限公司 | Business flow dynamic assembly system and method |
-
2009
- 2009-05-13 CN CNA2009100155162A patent/CN101551745A/en active Pending
Cited By (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102236679A (en) * | 2010-04-27 | 2011-11-09 | 杭州德昌隆信息技术有限公司 | Method and device for outputting workflow based on browser page |
| CN102394807A (en) * | 2011-08-23 | 2012-03-28 | 北京京北方信息技术有限公司 | System and method for decentralized scheduling of autonomous flow engine load balancing clusters |
| CN102394807B (en) * | 2011-08-23 | 2015-03-04 | 京北方信息技术股份有限公司 | System and method for decentralized scheduling of autonomous flow engine load balancing clusters |
| CN104471572B (en) * | 2012-07-12 | 2018-11-16 | 微软技术许可有限责任公司 | It is calculated using the gradual inquiry of streaming framework |
| CN104471572A (en) * | 2012-07-12 | 2015-03-25 | 微软公司 | Progressive query computation using streaming architectures |
| US10140358B2 (en) | 2012-07-12 | 2018-11-27 | Microsoft Technology Licensing, Llc | Progressive query computation using streaming architectures |
| CN103064964A (en) * | 2012-12-29 | 2013-04-24 | 天津南大通用数据技术有限公司 | Connection method of data base supporting distributed type affairs |
| CN103064964B (en) * | 2012-12-29 | 2016-04-20 | 天津南大通用数据技术股份有限公司 | A kind of method of attachment supporting the database of distributed transaction |
| CN104699411A (en) * | 2013-12-06 | 2015-06-10 | 北京慧正通软科技有限公司 | Technical method for improving performance of cache in workflow engine process instance |
| CN104751359B (en) * | 2013-12-30 | 2020-08-21 | 中国银联股份有限公司 | System and method for payment clearing |
| CN104751359A (en) * | 2013-12-30 | 2015-07-01 | 中国银联股份有限公司 | System and method for payment and settlement |
| CN104735152A (en) * | 2015-03-30 | 2015-06-24 | 四川神琥科技有限公司 | Mail reading method based on network |
| US10740328B2 (en) | 2016-06-24 | 2020-08-11 | Microsoft Technology Licensing, Llc | Aggregate-query database system and processing |
| CN106210022A (en) * | 2016-06-29 | 2016-12-07 | 天涯社区网络科技股份有限公司 | A kind of system and method for processing forum's height concurrent data requests |
| CN107657419A (en) * | 2016-07-25 | 2018-02-02 | 武汉票据交易中心有限公司 | The processing method and relevant apparatus and server of a kind of operation flow |
| CN106529917A (en) * | 2016-12-15 | 2017-03-22 | 平安科技（深圳）有限公司 | Workflow processing method and device |
| CN106529917B (en) * | 2016-12-15 | 2020-07-03 | 平安科技（深圳）有限公司 | Workflow processing method and device |
| US10552435B2 (en) | 2017-03-08 | 2020-02-04 | Microsoft Technology Licensing, Llc | Fast approximate results and slow precise results |
| CN112926206A (en) * | 2021-02-25 | 2021-06-08 | 北京工业大学 | Workflow engine cache elimination method based on industrial process background |
| CN112926206B (en) * | 2021-02-25 | 2024-04-26 | 北京工业大学 | Workflow engine cache elimination method based on industrial process background |
| CN115857907A (en) * | 2023-02-06 | 2023-03-28 | 卓望数码技术（深圳）有限公司 | Business flow dynamic assembly system and method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101551745A (en) | Method for greatly improving performance of workflow engine | |
| US20210042282A1 (en) | Providing snapshots of journal tables | |
| CN104933112B (en) | Distributed interconnection Transaction Information storage processing method | |
| US8364751B2 (en) | Automated client/server operation partitioning | |
| CN103116627B (en) | A kind of method and system of high concurrent SOA technology access database | |
| CN110019267A (en) | A kind of metadata updates method, apparatus, system, electronic equipment and storage medium | |
| US8380663B2 (en) | Data integrity in a database environment through background synchronization | |
| US9836516B2 (en) | Parallel scanners for log based replication | |
| CN102855239A (en) | Distributed geographical file system | |
| US20070100826A1 (en) | Method for improving the performance of database loggers using agent coordination | |
| CN110309233A (en) | Method, apparatus, server and the storage medium of data storage | |
| CN112163001A (en) | High-concurrency query method, intelligent terminal and storage medium | |
| EP4254187A1 (en) | Cross-organization & cross-cloud automated data pipelines | |
| CN102937964A (en) | Intelligent data service method based on distributed system | |
| WO2023033721A2 (en) | Storage engine for hybrid data processing | |
| EP4118536A1 (en) | Extensible streams on data sources | |
| CN113568938A (en) | Data stream processing method and device, electronic equipment and storage medium | |
| CN103365987A (en) | Clustered database system and data processing method based on shared-disk framework | |
| CN108280123B (en) | HBase column polymerization method | |
| CN102024010A (en) | Data processing system and data processing method | |
| CN102325098A (en) | Group information acquisition method and system | |
| CN106776810A (en) | The data handling system and method for a kind of big data | |
| CN100486177C (en) | Method of synchronously operating network element by network management and its system | |
| CN114237858A (en) | Task scheduling method and system based on multi-cluster network | |
| CN115080609A (en) | Method and system for realizing high-performance and high-reliability business process engine |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Open date: 20091007 |