CN1588316A - Property optimizing method for applying server - Google Patents
Property optimizing method for applying server Download PDFInfo
- Publication number
- CN1588316A CN1588316A CN 200410050035 CN200410050035A CN1588316A CN 1588316 A CN1588316 A CN 1588316A CN 200410050035 CN200410050035 CN 200410050035 CN 200410050035 A CN200410050035 A CN 200410050035A CN 1588316 A CN1588316 A CN 1588316A
- Authority
- CN
- China
- Prior art keywords
- task
- thread
- service object
- service
- queue
- 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.)
- Granted
Links
Images
Landscapes
- Multi Processors (AREA)
- Computer And Data Communications (AREA)
Abstract
The present invention relates to performance optimizing method for application server. After receiving the network service request, the network engine creates corresponding task associated with the service target before handing over the task to the task despatcher. If the number m of the work threads in current thread pool is smaller than the upper limit T_M of the thread pool, the despatcher will create one new thread in the pool for processing the task. If m=T_M, the despatcher will queue the task into the task queue. After completing the current task, the work threads in current thread pool will request new task from the task queue. Managing the application threads in the mode of finite thread pool plus infinite task queue can control the number of concurrent threads to reduce thread dispatching and competition overhead, raise utilization of the processor, increase the effective work time of the processor and increase the throughput of application server.
Description
Technical field
The present invention relates to a kind of performance optimization method of application server, belong to the Computer Applied Technology field.
Background technology
Application server is the high-rise running environment of application program under the network environment, service functions such as member operational management, communication, name, affairs, safety are provided for application system, make the application developer energy can be concentrated on the service logic of goal systems, simplified the exploitation and the maintenance of application system.Application server itself is a kind of server program, and it must possess good performance to satisfy the demand of application program.The situation that performance optimization arrives simultaneously at a large amount of concurrent client requests usually, target are to increase handling capacity and reduction response time as far as possible, make application server be operated in a kind of optimum condition.
Present application server adopts the multiple line distance management dispatching technique more, and the multiple line distance management dispatching technique mainly is the mode by the concurrent execution of a plurality of threads, reduces the standby time of processor, increases the handling capacity of processor.If but the use of multithreading is not added restriction, and then may run counter to desire, reduce handling capacity on the contrary.This is because after thread increases, create, destroy thread on the one hand and need consume processor resource, frequent switching will take a large amount of processor times between a plurality of on the other hand threads, make processor be in " making fruitless efforts " state, the CPU effective rate of utilization is low, and seldom cause the processing time of individual task to increase effective time that really is used to the request of handling, total throughout descends, and demands urgently further being improved.
Summary of the invention
Technical matters to be solved by this invention is to provide a kind of performance optimization method of application server, improves the utilization factor of processor, increases handling capacity, can improve the performance of application server system simply, effectively, reliably.
Technical matters to be solved by this invention is achieved by the following technical solution:
A kind of performance optimization method of application server, it may further comprise the steps:
Worker thread in step 4, the thread pool will be to task queue application new task after finishing current task.If there is task to exist in the task queue, then worker thread obtains the new task execution; If there is not task pending in the task queue, then worker thread changes sleep over to, and the wait task formation after new task has been arranged wakes it up.
The mode that the present invention adopts limited thread pool to add unlimited task queue manages using thread, and limited thread pool is the concurrent number of threads in the control system effectively, thereby reduces thread scheduling and competition expense, the effective rate of utilization of raising processor.
The present invention has also adopted local communication optimization technology, its objective is by in-process network call is converted into this locality to call, and to reduce network overhead, what is more important cooperates the thread pool technology, finishes a call chain with a task in a thread.
Described step 2 also comprise all service objects at first all the local service object registration center in local process register, the local service handle is registered to this local service object registration center.Network engine when related with the service object, judges whether this service object requires to call other service object service is provided with task, if do not require that calling other service object provides service, network engine is given task dispatcher with task; Provide service if require to call other service object, whether at first will arriving local service object registration center, to search the required service object who calls be native object; If the required service object who calls is not a native object, network engine is given task dispatcher with task; If the required service object who calls is native object, then carry out this locality between this service object and its required service object who calls and communicate by letter, in this thread, finish service object's call request with same task.
The present invention is directed to the strong characteristics of server application concurrency, use the multiple line distance management dispatching technique to improve the utilization factor of processor as far as possible on the one hand, increase handling capacity; Optimize local communication on the other hand,, thereby reduce concurrency, reduce the competition expense, avoid unnecessary local network communication with the local call chain of same thread execution.Characteristics of the present invention are the multiple line distance management dispatching technique is combined with local communication optimization technology, work in coordination on the whole, realize the target of performance optimization jointly.Above-mentioned two technology are used and can obtain best effects.
Description of drawings
Fig. 1 is a workflow diagram of the present invention;
Fig. 2 finishes the work synoptic diagram of a services request chain for a task of the present invention.
Embodiment
Below in conjunction with the drawings and specific embodiments technical scheme of the present invention is further specified:
As shown in Figure 1, a kind of performance optimization method of application server, it may further comprise the steps:
The mode that the present invention adopts limited thread pool to add unlimited task queue manages using thread, and limited thread pool is the concurrent number of threads in the control system effectively, thereby reduces thread scheduling and competition expense, the effective rate of utilization of raising processor.
The present invention has also adopted local communication optimization technology, its objective is by in-process network call is converted into this locality to call, and to reduce network overhead, what is more important cooperates the thread pool technology, finishes a call chain with a task in a thread.
As shown in Figure 2, described step 2 also comprise all service objects 3 at first all the local service object registration center 8 in local process register, the local service handle is registered to this local service object registration center 8.Network engine 2 is with task and service object 3 when related, judge whether this service object 31 requires to call other service object 32 or service object 33 provides service, if do not require and call other service object 32 or service object 33 provides service, network engine 2 is given task dispatcher 5 with task 4; If require to call other service object 32 or service object 33 provides service, to arrive at first that the required service object who calls 32 searches in local service object registration center 8 or whether service object 33 is native object; If required service object who calls 32 or service object 33 are not native object, network engine 2 is given task dispatcher 5 with task 4; If the required service object who calls 32 or service object 33 are native object, then carry out this locality between this service object 31 and its required service object who calls 32 or the service object 33 and communicate by letter, in this thread, finish service object's call request with same task 4.
The present invention is directed to the strong characteristics of server application concurrency, use the multiple line distance management dispatching technique to improve the utilization factor of processor as far as possible on the one hand, increase handling capacity; Optimize local communication on the other hand,, thereby reduce concurrency, reduce the competition expense, avoid unnecessary local network communication with the local call chain of same thread execution.Characteristics of the present invention are the multiple line distance management dispatching technique is combined with local communication optimization technology, work in coordination on the whole, realize the target of performance optimization jointly.Above-mentioned two technology are used and can obtain best effects.
Below, the effect of the present invention to performance optimization has been described for the test result of the present invention in the reality test.Test environment:
Application server and database server:
Hardware: De11PE2400 CPU-is two to strong 2.8G; Internal memory-1G
Operating system: Windows 2000 Server, ServicePack 3
Database: Oracle 8i
Client:
Hardware: PC CPU-AMD XP 1600+; Internal memory-256M
Operating system: Windows 2000 Professional, ServicePack 3
Test result as shown in Figure 3.
The test case that this test is used is the process that simulation electronic commercial affairs sent and handled order, and the actual client of client simulation sends request.Wherein, on behalf of client, speed send the concurrent speed of request, and it is many more that interior client of big more representation unit time sends request; The order transaction rate, the expression server generates the speed of order, and the speed of order process rate representation server process order, handling capacity then are that comprehensive top two speed provide the comprehensive measurement value, and numerical value is big more, and the expression server handling ability is strong more.When 25x speed, handling capacity can reach 2500, and when response valve can reach requirement.And under the same test environment, not using PKUAS of the present invention then can't be by test, and under 1x speed, the response time can't meet the demands; Handling capacity then is issued to extreme value about 250 in 3x speed.
It should be noted last that, above embodiment is only unrestricted in order to technical scheme of the present invention to be described, although the present invention is had been described in detail with reference to preferred embodiment, those of ordinary skill in the art is to be understood that, can make amendment or be equal to replacement technical scheme of the present invention, and not breaking away from the spirit and scope of technical solution of the present invention, it all should be encompassed in the middle of the claim scope of the present invention.
Claims (4)
1, a kind of performance optimization method of application server is characterized in that it may further comprise the steps:
Step 1, network engine receive network service request, create corresponding task;
Step 2, network engine are related with the service object with task, give task dispatcher with task then;
Step 3, task dispatcher check at first whether to exist in the current thread pond and are in dormant thread that if having, task dispatcher wakes this task of sleep thread process up; Otherwise judge the number m of worker thread in the current thread pond, if m less than the maximum upper limit T_M of thread pool, task dispatcher is created a new thread and is handled this task in thread pool; If m equals T_M, thread pool is saturated, and T_M thread work arranged in the system, and task dispatcher is put into task in the task queue and ranks;
Worker thread in step 4, the thread pool will be to task queue application new task after finishing current task; If there is task to exist in the task queue, then worker thread obtains the new task execution; If there is not task pending in the task queue, then worker thread changes sleep over to, and the wait task formation after new task has been arranged wakes it up.
2, the performance optimization method of application server according to claim 1 is characterized in that: described task queue capacity is for infinitely great.
3, the performance optimization method of application server according to claim 1, it is characterized in that: described step 2 also comprise all service objects at first all the local service object registration center in local process register, the local service handle is registered to this local service object registration center.
4, the performance optimization method of application server according to claim 3, it is characterized in that: described step 2 also comprises when the network engine task is related with the service object, judge whether this service object requires to call other service object service is provided, if do not require that calling other service object provides service, network engine is given task dispatcher with task; Provide service if require to call other service object, whether at first will arriving local service object registration center, to search the required service object who calls be native object; If the required service object who calls is not a native object, network engine is given task dispatcher with task; If the required service object who calls is native object, then carry out this locality between this service object and its required service object who calls and communicate by letter, in this thread, finish service object's call request with same task.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2004100500352A CN100347675C (en) | 2004-06-29 | 2004-06-29 | Property optimizing method for applying server |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2004100500352A CN100347675C (en) | 2004-06-29 | 2004-06-29 | Property optimizing method for applying server |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1588316A true CN1588316A (en) | 2005-03-02 |
| CN100347675C CN100347675C (en) | 2007-11-07 |
Family
ID=34602163
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2004100500352A Expired - Fee Related CN100347675C (en) | 2004-06-29 | 2004-06-29 | Property optimizing method for applying server |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100347675C (en) |
Cited By (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100430898C (en) * | 2006-12-20 | 2008-11-05 | 金魁 | Application system for high grade multiple line distance management |
| CN101136766B (en) * | 2006-11-10 | 2010-04-21 | 中兴通讯股份有限公司 | Task scheduling method for telecommunication network management system |
| CN101248405B (en) * | 2005-04-29 | 2010-09-15 | 微软公司 | Multithreading with concurrency domains |
| CN101916331A (en) * | 2010-08-06 | 2010-12-15 | 深圳市蓝韵实业有限公司 | Data preprocessing and scheduling method for three-dimensional server |
| CN102033784A (en) * | 2010-11-18 | 2011-04-27 | 山东中创软件工程股份有限公司 | Method and device for optimizing application server |
| CN101005486B (en) * | 2006-12-28 | 2011-07-27 | 金蝶软件（中国）有限公司 | Resource access control method and system |
| CN102262564A (en) * | 2011-08-16 | 2011-11-30 | 天津市天祥世联网络科技有限公司 | Thread pool structure of video monitoring platform system and realizing method |
| CN102402459A (en) * | 2010-09-10 | 2012-04-04 | 中兴通讯股份有限公司 | Method and device for summarizing performance data of network management system |
| CN101354661B (en) * | 2007-06-28 | 2012-06-27 | 英特尔公司 | System and method to optimize OS scheduling decisions for power savings based on temporal characteristics of the scheduled entity and system workload |
| CN101739293B (en) * | 2009-12-24 | 2012-09-26 | 航天恒星科技有限公司 | Method for scheduling satellite data product production tasks in parallel based on multithread |
| CN101782864B (en) * | 2009-12-01 | 2013-05-08 | 深圳市蓝韵网络有限公司 | Method for improving communication service stability of Web server |
| CN101753608B (en) * | 2008-12-09 | 2013-06-05 | 中国移动通信集团公司 | Dispatching method and system of distributed system |
| CN103218264A (en) * | 2013-03-26 | 2013-07-24 | 广东威创视讯科技股份有限公司 | Multi-thread finite state machine switching method and multi-thread finite state machine switching device based on thread pool |
| US8910189B2 (en) | 2007-02-28 | 2014-12-09 | International Business Machines Corporation | Methods and systems for automatically determining configuration parameters |
| CN104281489A (en) * | 2013-07-12 | 2015-01-14 | 携程计算机技术（上海）有限公司 | Multithreading request method and system under SOA (service oriented architecture) |
| CN104850496A (en) * | 2015-05-15 | 2015-08-19 | 浪潮电子信息产业股份有限公司 | RHEL (red hat enterprise linux) server adjusting and optimizing method facing low delay application |
| CN105389208A (en) * | 2015-11-10 | 2016-03-09 | 中国建设银行股份有限公司 | Job processing method and apparatus |
| CN103024388B (en) * | 2012-12-17 | 2016-05-25 | 广东威创视讯科技股份有限公司 | Video real-time decoding method and the system of many pictures |
| CN106648646A (en) * | 2016-12-13 | 2017-05-10 | 深圳市元征软件开发有限公司 | Method and device for managing threads |
| CN106940658A (en) * | 2017-02-21 | 2017-07-11 | 腾讯科技（深圳）有限公司 | Task processing method and device based on thread pool |
| CN109086194A (en) * | 2018-06-11 | 2018-12-25 | 山东中创软件商用中间件股份有限公司 | A kind of thread pool capacity adjustment method, device, equipment and readable storage medium storing program for executing |
| CN109544789A (en) * | 2019-01-23 | 2019-03-29 | 小雨科技(上海)有限公司 | A kind of automatic concurrent shipment system |
| CN109658600A (en) * | 2018-12-24 | 2019-04-19 | 小雨科技(上海)有限公司 | A kind of automatic concurrent shipment system and method |
| CN111381955A (en) * | 2018-12-28 | 2020-07-07 | 无锡智道安盈科技有限公司 | Method for optimizing decision engine throughput rate and response time |
| CN111858056A (en) * | 2020-07-24 | 2020-10-30 | 南方电网数字电网研究院有限公司 | Task scheduling method and device, computer equipment and storage medium |
| CN111858046A (en) * | 2020-07-13 | 2020-10-30 | 海尔优家智能科技（北京）有限公司 | Service request processing method and device, storage medium and electronic device |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106936688B (en) * | 2015-12-30 | 2020-11-24 | 北京国双科技有限公司 | Notification sending method and device |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB2334116A (en) * | 1998-02-04 | 1999-08-11 | Ibm | Scheduling and dispatching queued client requests within a server computer |
| US6886041B2 (en) * | 2001-10-05 | 2005-04-26 | Bea Systems, Inc. | System for application server messaging with multiple dispatch pools |
-
2004
- 2004-06-29 CN CNB2004100500352A patent/CN100347675C/en not_active Expired - Fee Related
Cited By (33)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101248405B (en) * | 2005-04-29 | 2010-09-15 | 微软公司 | Multithreading with concurrency domains |
| CN101136766B (en) * | 2006-11-10 | 2010-04-21 | 中兴通讯股份有限公司 | Task scheduling method for telecommunication network management system |
| CN100430898C (en) * | 2006-12-20 | 2008-11-05 | 金魁 | Application system for high grade multiple line distance management |
| CN101005486B (en) * | 2006-12-28 | 2011-07-27 | 金蝶软件（中国）有限公司 | Resource access control method and system |
| US8910189B2 (en) | 2007-02-28 | 2014-12-09 | International Business Machines Corporation | Methods and systems for automatically determining configuration parameters |
| CN101354661B (en) * | 2007-06-28 | 2012-06-27 | 英特尔公司 | System and method to optimize OS scheduling decisions for power savings based on temporal characteristics of the scheduled entity and system workload |
| CN101753608B (en) * | 2008-12-09 | 2013-06-05 | 中国移动通信集团公司 | Dispatching method and system of distributed system |
| CN101782864B (en) * | 2009-12-01 | 2013-05-08 | 深圳市蓝韵网络有限公司 | Method for improving communication service stability of Web server |
| CN101739293B (en) * | 2009-12-24 | 2012-09-26 | 航天恒星科技有限公司 | Method for scheduling satellite data product production tasks in parallel based on multithread |
| CN101916331B (en) * | 2010-08-06 | 2013-05-08 | 深圳市蓝韵网络有限公司 | Data preprocessing and scheduling method for three-dimensional server |
| CN101916331A (en) * | 2010-08-06 | 2010-12-15 | 深圳市蓝韵实业有限公司 | Data preprocessing and scheduling method for three-dimensional server |
| CN102402459A (en) * | 2010-09-10 | 2012-04-04 | 中兴通讯股份有限公司 | Method and device for summarizing performance data of network management system |
| CN102402459B (en) * | 2010-09-10 | 2013-09-11 | 中兴通讯股份有限公司 | Method and device for summarizing performance data of network management system |
| CN102033784A (en) * | 2010-11-18 | 2011-04-27 | 山东中创软件工程股份有限公司 | Method and device for optimizing application server |
| CN102033784B (en) * | 2010-11-18 | 2013-08-07 | 山东中创软件工程股份有限公司 | Method and device for optimizing application server |
| CN102262564A (en) * | 2011-08-16 | 2011-11-30 | 天津市天祥世联网络科技有限公司 | Thread pool structure of video monitoring platform system and realizing method |
| CN103024388B (en) * | 2012-12-17 | 2016-05-25 | 广东威创视讯科技股份有限公司 | Video real-time decoding method and the system of many pictures |
| CN103218264A (en) * | 2013-03-26 | 2013-07-24 | 广东威创视讯科技股份有限公司 | Multi-thread finite state machine switching method and multi-thread finite state machine switching device based on thread pool |
| CN104281489B (en) * | 2013-07-12 | 2017-11-21 | 上海携程商务有限公司 | Multithreading requesting method and system under SOA framework |
| CN104281489A (en) * | 2013-07-12 | 2015-01-14 | 携程计算机技术（上海）有限公司 | Multithreading request method and system under SOA (service oriented architecture) |
| CN104850496A (en) * | 2015-05-15 | 2015-08-19 | 浪潮电子信息产业股份有限公司 | RHEL (red hat enterprise linux) server adjusting and optimizing method facing low delay application |
| WO2016188102A1 (en) * | 2015-11-10 | 2016-12-01 | 中国建设银行股份有限公司 | Task processing method and device |
| CN105389208A (en) * | 2015-11-10 | 2016-03-09 | 中国建设银行股份有限公司 | Job processing method and apparatus |
| CN106648646A (en) * | 2016-12-13 | 2017-05-10 | 深圳市元征软件开发有限公司 | Method and device for managing threads |
| CN106940658B (en) * | 2017-02-21 | 2022-09-09 | 腾讯科技（深圳）有限公司 | Task processing method and device based on thread pool |
| CN106940658A (en) * | 2017-02-21 | 2017-07-11 | 腾讯科技（深圳）有限公司 | Task processing method and device based on thread pool |
| CN109086194A (en) * | 2018-06-11 | 2018-12-25 | 山东中创软件商用中间件股份有限公司 | A kind of thread pool capacity adjustment method, device, equipment and readable storage medium storing program for executing |
| CN109658600A (en) * | 2018-12-24 | 2019-04-19 | 小雨科技(上海)有限公司 | A kind of automatic concurrent shipment system and method |
| CN111381955A (en) * | 2018-12-28 | 2020-07-07 | 无锡智道安盈科技有限公司 | Method for optimizing decision engine throughput rate and response time |
| CN109544789A (en) * | 2019-01-23 | 2019-03-29 | 小雨科技(上海)有限公司 | A kind of automatic concurrent shipment system |
| CN111858046A (en) * | 2020-07-13 | 2020-10-30 | 海尔优家智能科技（北京）有限公司 | Service request processing method and device, storage medium and electronic device |
| CN111858056A (en) * | 2020-07-24 | 2020-10-30 | 南方电网数字电网研究院有限公司 | Task scheduling method and device, computer equipment and storage medium |
| CN111858056B (en) * | 2020-07-24 | 2021-12-17 | 南方电网数字电网研究院有限公司 | Task scheduling method and device, computer equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN100347675C (en) | 2007-11-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN100347675C (en) | Property optimizing method for applying server | |
| US9524296B2 (en) | Managing events in a computing environment | |
| CN106020927B (en) | Task scheduling and the universal method of resource distribution in a kind of cloud computing system | |
| US20080306950A1 (en) | Arrival rate throttles for workload management | |
| US8392404B2 (en) | Dynamic query and step routing between systems tuned for different objectives | |
| CN109205415A (en) | A kind of Elevator group control method, device, computer equipment and storage medium | |
| CN1155340A (en) | Computer system having client-server architecture | |
| CN110654946B (en) | Community elevator dispatching method and system based on artificial intelligence | |
| US20080288946A1 (en) | States matrix for workload management simplification | |
| US20090049024A1 (en) | Dynamic query optimization between systems based on system conditions | |
| CN112559159A (en) | Task scheduling method based on distributed deployment | |
| US20090132602A1 (en) | Virtual data maintenance | |
| CN105912383A (en) | High-reliability dependent task scheduling and resource configuration method | |
| CN103677959B (en) | A kind of virtual machine cluster migration method and system based on multicast | |
| El Ghor et al. | Energy efficient scheduler of aperiodic jobs for real-time embedded systems | |
| CN109693978B (en) | Intelligent elevator dispatching method and device, computer equipment and storage medium | |
| CN100383743C (en) | Real-time task scheduling method in Java operating system | |
| Davis et al. | Quantifying the sub-optimality of uniprocessor fixed priority non-pre-emptive scheduling | |
| US20110055168A1 (en) | System, method, and computer-readable medium to facilitate application of arrival rate qualifications to missed throughput server level goals | |
| CN113127176A (en) | Multi-role task allocation method and system for working platform | |
| Naik et al. | Scheduling tasks on most suitable fault tolerant resource for execution in computational grid | |
| US20120059817A1 (en) | Method for increasing the efficiency of synchronized scans through intelligent query dispatching | |
| CN113298374B (en) | Resource management system based on big data and intelligent medical treatment | |
| CN1302412C (en) | Computer group system and its operation managing method | |
| CN112291320A (en) | Distributed two-layer scheduling method and system for quantum computer cluster |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20071107 Termination date: 20130629 |