CN109739745A - Based on flow indicator to database hardware resource analysis method - Google Patents
Based on flow indicator to database hardware resource analysis method Download PDFInfo
- Publication number
- CN109739745A CN109739745A CN201811506069.6A CN201811506069A CN109739745A CN 109739745 A CN109739745 A CN 109739745A CN 201811506069 A CN201811506069 A CN 201811506069A CN 109739745 A CN109739745 A CN 109739745A
- Authority
- CN
- China
- Prior art keywords
- server
- tpmc
- test
- business
- value
- 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
Landscapes
- Debugging And Monitoring (AREA)
Abstract
It is a kind of based on flow indicator to database hardware resource analysis method, according to the complexity factor of special control system, it acquires, the tpmC value of the DATABASE HARDWARE needed to configure be equal to specially control system complexity factor with it is desirable that tpmC value product, a standard calculations of special control system are obtained, the hardware resource matching of specially control system is carried out according to standard calculations, therefore guarantees the reliability of operation energy of specially control system.
Description
One, technical field
The present invention relates to a kind of pair of database hardware resource analysis method, it is especially a kind of based on flow indicator to data
Library hardware resource analysis method.
Two, background technique
For specially controlling system, in order to guarantee that business operates normally, need to guarantee the configuration of the hardware resource in special control system,
To guarantee the reliability of operation energy of specially control system.
Based on applicant's technical problem existing in the Disclosure of invention on July 15th, 2018 and background technique, technology
Feature and technical effect make application technical solution of the invention.
Three, summary of the invention
Object of the invention be it is a kind of based on flow indicator to database hardware resource analysis method.
In order to overcome the technical drawbacks described above, the object of the present invention is to provide a kind of based on flow indicator to DATABASE HARDWARE
Resource analysis method, therefore guarantee the reliability of operation energy of specially control system.
In order to achieve the above objectives, the technical solution adopted by the present invention is that: the steps include: according to specially control system complexity system
Number, is acquired, the tpmC value of the DATABASE HARDWARE needed to configure be equal to the special complexity factor of control system with it is desirable that tpmC
It is worth product.
Due to devising the complexity factor of specially control system, a standard calculations of special control system are obtained, according to mark
Quasi- calculated value carries out the hardware resource matching of specially control system, therefore guarantees the reliability of operation energy of specially control system.
The present invention devises, and the complexity factor according to special control system is radix matching database hardware resource.
The present invention devises, and the steps include: to establish TPC-C benchmark business model in special control system, passes through pressure test
Contrast standard business obtains maximum tpmC value by TPC-C benchmark business model, by practical business, by TPC-C benchmark business
Model obtains the tpmC value of practical business, and the complexity factor for specially controlling system is equal to the tpmC of maximum tpmC value and practical business
The ratio of value.
The present invention devises, and TPC-C benchmark business model trade transactions to be treated are predominantly following several:
One, new order (New-Order): client inputs a new order transaction,
Two, delivery operation (Payment): update clients' accounts remaining sum to reflect its payment status,
Three, it delivers (Delivery): delivery (simulation batch processing transaction),
Four, order status inquiries (Order-Status): the state of inquiry client's last sale,
Five, inventory status query (Stock-Level): inquiry warehouse inventory situation, so as to replenish in time.
The present invention devises, and the Warehouse module of TPC-C benchmark business model is respectively set to and District module
With Stock module relation, Item module is set as being set as and Customer mould with Stock module relation, District module
Block association, Customer module be respectively set to History module and Order module relation, module is set as Order respectively
With New-Order module and Order-Line module relation, Stock module pass is set as and Order-Line module relation.
The present invention devises, and is filled according to core business scene-inspector's log of the sale monopoly operation system of practical business
Inquiry inspection plan, fill in inspection log (head row table), return to inspection log list, four sql extract, and form one group
Data bank service.
The present invention devises, test method:
One, TPCC is tested
TPC-C test content: db transaction processing test measures server and database software processing online query is handed over
The performance of easy to handle (OLTP).Regular TPC-C test result publication must provide tpmC value, i.e., complete how many pen per minute
TPC-C database transaction (TPC-C Transaction Per Minute).Two, testing tool
Test process is tested using benchmarksql-4.1.1 tool, using the combinations of transactions and weight of standard
Value is as follows:
Three, environment is tested
Two kinds of situations of test procedure test, a kind of situation are that pressure surveys single database maximum tpmC value, another feelings
Condition is that pressure surveys server in multiple databases, and the tpmC value maximum upper limit of server is tested.
The case where test is all using 100 warehouses of initialization,
| ./runLoader.sh props.db2 numWarehouses 100 |
Configuration file:
The present invention devises, and chooses three server As, server B and server C in special control system, by pressure test,
The maximum tpmC value for obtaining server A is 553617.49, the maximum tpmC value of server B is 382983.53, server C most
Big tpmC value is 779081,
The inquiry inspection meter filled according to core business scene-inspector's log of the sale monopoly operation system of practical business
It drawing, fills in inspection log (head row table), returns to inspection log list, four sql are extracted, one group of data bank service is formed,
(how to be converted into new order (New-Order) in TPC-C benchmark business model, delivery operation (Payment), delivery
(Delivery), order status inquiries (Order-Status), inventory status query (Stock-Level)), by TPC-C benchmark
Business model, the tpmC value for obtaining the practical business of server A is 134209.8, the tpmC value of the practical business of server B is
92092.7, the tpmC value of the practical business of server C is 200847.8, is obtained by calculation, and the server A of system is specially controlled
Complexity factor is 4.12, and the complexity factor for specially controlling the server B of system is 4.15, specially controls the complexity of the server C of system
Spending coefficient is 3.88, is obtained by carrying out complexity factor mean value calculation, and the complexity factor for specially controlling system is 4, works as expectation
When the tpmC needed is 10000, the tpmC value of the DATABASE HARDWARE needed to configure is 40000.
The present invention devises, and first pressure test is done with benchmark model on testing service device, under cpu full load situation
(because the scalability of cpu is comparatively relatively low for server, and the scalability of disk and memory is higher, works as magnetic
Disk and memory are contemplated that extension disk and memory when first going out bottleneck), to obtain benchmark tpmC value, it is assumed that m；Then again by mian part
Core business scenario sql extracts to form data bank service in the system of administration, then carries out pressure test with same server, from
And obtain business tpmC value, it is assumed that be n.Calculate the coefficient of m/n, i.e. complexity of the practical business relative to benchmark business, it is assumed that
For x.According to the sometime point maximum concurrency of actual business requirement, it is assumed that be y, the only tpmC when the server bought
When value meets xy, this server is just able to satisfy the demand of our deployment.Therefore, before purchasing server, supply can be allowed commercial
Our benchmark model to server carry out pressure test, the benchmark tpmC value obtained just or be slightly larger than xy when, i.e., it is believed that
It is the server for being suitble to us to purchase.
In the technical scheme, hardware resource refers to cpu memory and disk.
In the technical scheme, it is important technical characteristic that specially the complexity factor of control system, which is radix, is referred to based on flow
Target has novelty, creativeness and practicability, in this technology side in the technical field of database hardware resource analysis method
Term in case is all that can be explained and be understood with patent document in the art.
Four, Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Relationship and big logotype of the Fig. 1 for the trade transactions of TPC-C benchmark business model of the invention,
Fig. 2 is the database cpu monitor datagram of one of one embodiment of the invention,
Fig. 3 is the memory service condition figure of one of one embodiment of the invention,
Fig. 4 is the Internet Use figure of one of one embodiment of the invention,
Fig. 5 is the disk monitoring data figure of one of one embodiment of the invention,
Fig. 6 is process queue situation map during the pressure of one of one embodiment of the invention is surveyed,
Fig. 7 is the two database cpu monitor datagram of one embodiment of the invention,
Fig. 8 is the two memory service condition figure of one embodiment of the invention,
Fig. 9 is the two Internet Use figure of one embodiment of the invention,
Figure 10 is the two disk monitoring data figure of one embodiment of the invention,
Figure 11 is process queue situation map during two pressure of one embodiment of the invention is surveyed,
Figure 12 is the three database cpu monitor datagram of one embodiment of the invention,
Figure 13 is the three memory service condition figure of one embodiment of the invention,
Figure 14 is the three Internet Use figure of one embodiment of the invention,
Figure 15 is the three disk monitoring data figure of one embodiment of the invention,
Figure 16 is process queue situation map during three pressure of one embodiment of the invention is surveyed.
Five, specific embodiment
According to guidelines for examination, such as " having ", "comprising" and " comprising " term used in the present invention should understand that
For the presence or addition for not allotting one or more of the other element or combinations thereof.
In the description of the present invention, it should be noted that term " center ", "upper", "lower", "left", "right", "vertical",
The orientation or positional relationship of the instructions such as "horizontal", "inner", "outside" be based on the orientation or positional relationship shown in the drawings, merely to
Convenient for description the present invention and simplify description, rather than the device or element of indication or suggestion meaning must have a particular orientation,
It is constructed and operated in a specific orientation, therefore is not considered as limiting the invention.In addition, term " first ", " second ",
" third " is used for descriptive purposes only and cannot be understood as indicating or suggesting relative importance.
In the description of the present invention, it should be noted that unless otherwise clearly defined and limited, term " installation ", " phase
Even ", " connection " shall be understood in a broad sense, for example, it may be being fixedly connected, may be a detachable connection, or be integrally connected；It can
To be mechanical connection, it is also possible to be electrically connected；It can be directly connected, can also can be indirectly connected through an intermediary
Connection inside two elements.For the ordinary skill in the art, above-mentioned term can be understood at this with concrete condition
Concrete meaning in invention.
As long as in addition, the non-structure each other of technical characteristic involved in invention described below different embodiments
It can be combined with each other at conflict.
Below with reference to embodiment, the present invention is further described, following embodiment is intended to illustrate invention rather than to this
Invention further limits.One of one embodiment of the invention the steps include:
TPC-C benchmark business model is established in special control system, by pressure test contrast standard business, by TPC-C base
Quasi- business model obtains maximum tpmC value, obtains the tpmC of practical business by TPC-C benchmark business model by practical business
Value specially controls ratio of the complexity factor equal to maximum tpmC value and the tpmC value of practical business of system, according to special control system
Complexity factor, the tpmC value of the DATABASE HARDWARE needed to configure be equal to the special complexity factor of control system with it is desirable that
TpmC value product.
In the present embodiment, TPC-C benchmark business model is a testing standard for Database server performance, this
A standard describes an operation system scene (hereinafter referred to as benchmark service), i.e., one large-scale goods wholesale sales company,
It possesses several merchandise warehouses for being distributed in different zones.When operation expanding, company will add new warehouse, each
Warehouse is responsible for 10 point of sale supplies of material, wherein each point of sale is 3000 offering customers services, what each client submitted is ordered
Dan Zhong, average each order have 10 products, in all orders about 1% product its directly belonging to warehouse in do not deposit
Goods, it is necessary to be supplied by the warehouse in other regions.Meanwhile each warehouse will service companies sale 100000 kinds of commodity library
Deposit record.
In the present embodiment, TPC-C benchmark business model trade transactions to be treated are predominantly following several:
One, new order (New-Order): client inputs a new order transaction,
Two, delivery operation (Payment): update clients' accounts remaining sum to reflect its payment status,
Three, it delivers (Delivery): delivery (simulation batch processing transaction),
Four, order status inquiries (Order-Status): the state of inquiry client's last sale,
Five, inventory status query (Stock-Level): inquiry warehouse inventory situation, so as to replenish in time,
In the relationship and size such as attached drawing 1 of the trade transactions of TPC-C benchmark business model, Warehouse module is set respectively
It is set to and is set as and Stock module relation, District module with District module and Stock module relation, Item module
Be set as with Customer module relation, Customer module be respectively set to History module and Order module relation,
Order respectively module be set as with New-Order module and Order-Line module relation, Stock module pass be set as with
Order-Line module relation,
Wherein, the table of the digital representation in bezel, cluster will store how many records, and the adjustment of warehouse number W in testing can
Embody the ability for the data scale that database can be supported；Of son between the set membership of tables of data registration evidence between table
Number, for example a Warehouse will be corresponding to 10 District etc., in addition, "+" number indicates that this corresponding relationship may be more
It is more.
In the present embodiment, it flow indicator (abbreviation tpmC): according to the definition of international standard, just depicts system and is holding
While this 4 kinds transaction of row delivery operation, order status inquiries, delivery and inventory status query, can handle how many per minute
A new order transaction.The response time of All Activity must satisfy the requirement of TPC-C test specification, and shared by various number of transaction
Ratio should also meet the requirement of TPC-C test specification.In this case, flow indicator value is bigger illustrates the online of system
Transaction capabilities is higher, that is, the pressure that database server can bear is bigger, for example, certain database server with
This benchmark service model is tested, and final tpmC=1000, illustrating this database server per minute can bear most
Big data handling capacity is 1000.
In the present embodiment, test method:
One, TPCC is tested
TPC-C test content: db transaction processing test measures server and database software processing online query is handed over
The performance of easy to handle (OLTP).Regular TPC-C test result publication must provide tpmC value, i.e., complete how many pen per minute
TPC-C database transaction (TPC-C Transaction Per Minute).Two, testing tool
Test process is tested using benchmarksql-4.1.1 tool, using the combinations of transactions and weight of standard
Value is as follows:
Three, environment is tested
Two kinds of situations of test procedure test, a kind of situation are that pressure surveys single database maximum tpmC value, another feelings
Condition is that pressure surveys server in multiple databases, and the tpmC value maximum upper limit of server is tested.
The case where test is all using 100 warehouses of initialization,
| ./runLoader.sh props.db2 numWarehouses 100 |
Configuration file:
Four, test result
The attainable maximum tpmC value of test database server institute, in order to reach maximum upper limit, in this test cases
6 pressure survey processes are opened, are pressed survey process continuous service 10 minutes, final result, total number of transactions 5541927, total tpmC value
553617.49, and it is clearly visible disk I/O bottleneck, specific data form is as follows:
Five, test case
Core business scene-inspector's log that interior guard system of having a monopoly in is chosen in operational trials is filled in, and each inspector fills in
Log need three steps operation: 1, inquiry inspection plan, 2, fill in inspection log (head row table), 3, return inspection log list, will
Three steps operate total four sql and extract, and form one group of data bank service, concurrently execute four by pressure test tool
Sql, test case certification and major functional steps are described as follows:
(1), inquiry inspector checks plan
(2), it increases inspection log (being related to two tables of table to the end and row table) newly and saves.
(3), inspection log list is returned after the completion of saving.
Test process
The detailed test that this test has carried out 450 and given records affairs under concurrent pressure during the test in detail
TPS and the response time.
During pressure is surveyed, set time point is set, and 450 threads as far as possible operate simultaneously, when being during which not provided with pause
Between.
Test result
This performance test mainly tests 450 and gives the processing capacity of database, record affairs response time and
TPS.Test discovery, system 450 and it is following concurrent when, system processing capacity is relatively stable, and cpu and disk utilization rate reach
80% or so, it is basic it is considered that reaching the peak value of system processing capacity.
Affairs TPS, response time:
Test result verifying:
Check database journal, comparison number of transactions discovery, affairs percent of pass is 100%.
The two of one embodiment of the invention, test method:
One, TPCC is tested
TPC-C test content: db transaction processing test measures server and database software processing online query is handed over
The performance of easy to handle (OLTP).Regular TPC-C test result publication must provide tpmC value, i.e., complete how many pen per minute
TPC-C database transaction (TPC-C Transaction Per Minute).
Two, testing tool
Test process is tested using benchmarksql-4.1.1 tool, using the combinations of transactions and weight of standard
Value is as follows:
Three, environment is tested
Two kinds of situations of test procedure test, a kind of situation are that pressure surveys single database maximum tpmC value, another feelings
Condition is that pressure surveys server in multiple databases, and the tpmC value maximum upper limit of server is tested.
The case where test is all using 100 warehouses of initialization,
| ./runLoader.sh props.db2 numWarehouses 100 |
Configuration file:
Four, test result
Test two opens 3 using test same test data setting in order to reach maximum upper limit in this test cases
Survey process is pressed, is pressed survey process continuous service 10 minutes, final result, total number of transactions 1512266, total tpmC value 151050.26, and
It is clearly visible disk I/O bottleneck, specific data form is as follows:
Five, test case
Core business scene-inspector's log that interior guard system of having a monopoly in is chosen in operational trials is filled in, and each inspector fills in
Log need three steps operation: 1, inquiry inspection plan, 2, fill in inspection log (head row table), 3, return inspection log list, will
Three steps operate total four sql and extract, and form one group of data bank service, concurrently execute four by pressure test tool
Sql, test case certification and major functional steps are described as follows:
(1), inquiry inspector checks plan
(2), it increases inspection log (being related to two tables of table to the end and row table) newly and saves.
(3), inspection log list is returned after the completion of saving.
Test process
The detailed test that this test has carried out 450 and given records affairs under concurrent pressure during the test in detail
TPS and the response time.
During pressure is surveyed, set time point is set, and 450 threads as far as possible operate simultaneously, when being during which not provided with pause
Between.
Test result
This performance test mainly tests 450 and gives the processing capacity of database, record affairs response time and
TPS.Test discovery, system 450 and it is following concurrent when, system processing speed is higher, and cpu utilization rate is lower, but disk makes
Reach 90%-100% or so with peak value is reached, it is basic it is considered that reaching the peak value of system processing capacity.
Affairs TPS, response time:
| Thread | Total number of transactions | Time started | End time | Total duration (S) | tpmC |
| Thread 1 | 300000 | 15:44:01 | 15:53:46 | 585.7 | 30732.5 |
| Thread 2 | 300000 | 15:44:01 | 15:53:47 | 586.7 | 30680.1 |
| Thread 3 | 300000 | 15:44:01 | 15:53:47 | 586.7 | 30680.1 |
| It is total | 900000 | 92092.7 |
Test result verifying:
Check database journal, comparison number of transactions discovery, affairs percent of pass is 100%.
The three of one embodiment of the invention, test method:
One, TPCC is tested
TPC-C test content: db transaction processing test measures server and database software processing online query is handed over
The performance of easy to handle (OLTP).Regular TPC-C test result publication must provide tpmC value, i.e., complete how many pen per minute
TPC-C database transaction (TPC-C Transaction Per Minute).
Two, testing tool
Test process is tested using benchmarksql-4.1.1 tool, using the combinations of transactions and weight of standard
Value is as follows:
Three, environment is tested
Two kinds of situations of test procedure test, a kind of situation are that pressure surveys single database maximum tpmC value, another feelings
Condition is that pressure surveys server in multiple databases, and the tpmC value maximum upper limit of server is tested.
The case where test is all using 100 warehouses of initialization,
| ./runLoader.sh props.db2 numWarehouses 100 |
Configuration file:
Four, test result
It is arranged using same test data, in order to reach maximum upper limit, 3 pressure survey processes is opened in this test cases,
Pressure survey process continuous service 10 minutes, final result, total number of transactions 7793892, total tpmC value 779081, and obvious resource uses
Upper bottleneck, specific data form are as follows:
| Process | Total number of transactions | Time started | End time | tpmC |
| tpcc | 2598153 | 2016-07-27 22:56:27 | 2016-07-27 23:06:31 | 259805.98 |
| tpcc1 | 2532319 | 2016-07-27 22:56:20 | 2016-07-27 23:06:23 | 252943.38 |
| tpcc2 | 2663420 | 2016-07-27 22:56:15 | 2016-07-27 23:06:15 | 266331.64 |
| It is total | 7793892 | 779081 |
Five, test case
Core business scene-inspector's log that interior guard system of having a monopoly in is chosen in operational trials is filled in, and each inspector fills in
Log need three steps operation: 1, inquiry inspection plan, 2, fill in inspection log (head row table), 3, return inspection log list, will
Three steps operate total four sql and extract, and form one group of data bank service, concurrently execute four by pressure test tool
Sql, test case certification and major functional steps are described as follows:
(1), inquiry inspector checks plan
(2), it increases inspection log (being related to two tables of table to the end and row table) newly and saves.
(3), inspection log list is returned after the completion of saving.
Test process
The detailed test that this test has carried out 450 and given records affairs under concurrent pressure during the test in detail
TPS and the response time.
During pressure is surveyed, set time point is set, and 450 threads as far as possible operate simultaneously, when being during which not provided with pause
Between.
Test result
This performance test mainly tests 450 and gives the processing capacity of database, record affairs response time and
TPS.Test discovery, system 450 and it is following concurrent when, system processing speed is higher, and cpu utilization rate is lower, but disk makes
Reach 90%-100% or so with peak value is reached, it is basic it is considered that reaching the peak value of system processing capacity.
Affairs TPS, response time:
| Thread | Total number of transactions | Time started | End time | Total duration (S) | tpmC |
| Thread 1 | 300000 | 23:11:35 | 23:16:03 | 268.8 | 66960.5 |
| Thread 2 | 300000 | 23:11:36 | 23:16:03 | 268.6 | 67001.9 |
| Thread 3 | 300000 | 23:11:35 | 23:16:03 | 269.1 | 66885.4 |
| It is total | 900000 | 200847.8 |
Test result verifying:
Check database journal, comparison number of transactions discovery, affairs percent of pass is 100%.
Second embodiment of the invention is chosen three server As, server B and server C in special control system, is passed through
Pressure test, the maximum tpmC value for obtaining server A is 553617.49, the maximum tpmC value of server B is 382983.53, clothes
The maximum tpmC value of business device C is 779081,
The inquiry inspection meter filled according to core business scene-inspector's log of the sale monopoly operation system of practical business
It drawing, fills in inspection log (head row table), returns to inspection log list, four sql are extracted, one group of data bank service is formed,
(how to be converted into new order (New-Order) in TPC-C benchmark business model, delivery operation (Payment), delivery
(Delivery), order status inquiries (Order-Status), inventory status query (Stock-Level)), by TPC-C benchmark
Business model, the tpmC value for obtaining the practical business of server A is 134209.8, the tpmC value of the practical business of server B is
92092.7, the tpmC value of the practical business of server C is 200847.8, is obtained by calculation, and the server A of system is specially controlled
Complexity factor is 4.12, and the complexity factor for specially controlling the server B of system is 4.15, specially controls the complexity of the server C of system
Spending coefficient is 3.88, is obtained by carrying out complexity factor mean value calculation, and the complexity factor for specially controlling system is 4, works as expectation
When the tpmC needed is 10000, the tpmC value of the DATABASE HARDWARE needed to configure is 40000.
In the present embodiment, pressure test first is done with benchmark model on testing service device, under cpu full load situation
(because the scalability of cpu is comparatively relatively low for server, and the scalability of disk and memory is higher, works as magnetic
Disk and memory are contemplated that extension disk and memory when first going out bottleneck), to obtain benchmark tpmC value, it is assumed that m；Then again by mian part
Core business scenario sql extracts to form data bank service in the system of administration, then carries out pressure test with same server, from
And obtain business tpmC value, it is assumed that be n.Calculate the coefficient of m/n, i.e. complexity of the practical business relative to benchmark business, it is assumed that
For x.According to the sometime point maximum concurrency of actual business requirement, it is assumed that be y, the only tpmC when the server bought
When value meets xy, this server is just able to satisfy the demand of our deployment.Therefore, before purchasing server, supply can be allowed commercial
Our benchmark model to server carry out pressure test, the benchmark tpmC value obtained just or be slightly larger than xy when, i.e., it is believed that
It is the server for being suitble to us to purchase.
The present invention has the following characteristics:
1, the complexity factor that system is specially controlled due to devising, has obtained a standard calculations of special control system, according to
Standard calculations carry out the hardware resource matching of specially control system, therefore guarantee the reliability of operation energy of specially control system.
2, the complexity factor that system is specially controlled due to devising ensure that the matching between the specially hardware resource of control system
Can, improve the arithmetic speed of specially control system.
3, the restriction of numberical range has been carried out to planform due to devising, has made the technical side that numberical range is of the invention
Technical characteristic in case is not to be calculated by formula or test the technical characteristic obtained by limited times, experiments have shown that the numerical value
The technical characteristic of range achieves good technical effect.
4, due to devising technical characteristic of the invention, in the effect of the independent and mutual set of technical characteristic,
By experiments have shown that, performance indexes of the invention be existing performance indexes be at least 1.7 times, pass through assessment tool
There is good market value.
It is the identical or similar technical characteristic of radix there are also other complexity factors with special control system is all this hair
One of bright embodiment, and each technical characteristic of embodiment described above can be combined arbitrarily, for meet Patent Law,
The requirement of patent regulation and guidelines for examination, no longer to all possible combination of each technical characteristic in above-described embodiment
Embodiment is all described.
Above-described embodiment be it is provided by the present invention based on flow indicator to database hardware resource analysis method
A kind of way of realization, other deformations of provided scheme, increase or reduce composition therein or step according to the present invention, or
The present invention is used for other technical fields close with the present invention by person, is all belonged to the scope of protection of the present invention.
Claims (9)
1. it is a kind of based on flow indicator to database hardware resource analysis method, it is characterized in that: according to the complexity of special control system
Spend coefficient, acquire, the tpmC value of the DATABASE HARDWARE needed to configure be equal to the special complexity factor of control system with it is desirable that
TpmC value product.
2. it is according to claim 1 based on flow indicator to database hardware resource analysis method, it is characterized in that: according to
Specially the complexity factor of control system is radix matching database hardware resource.
3. it is according to claim 1 based on flow indicator to database hardware resource analysis method, it is characterized in that: its walk
Suddenly it is: establishes TPC-C benchmark business model in special control system, by pressure test contrast standard business, by TPC-C benchmark industry
Business model, obtains maximum tpmC value, obtains the tpmC value of practical business by TPC-C benchmark business model by practical business,
Specially the complexity factor of control system is equal to the ratio of the tpmC value of maximum tpmC value and practical business.
4. it is according to claim 3 based on flow indicator to database hardware resource analysis method, it is characterized in that: TPC-
C benchmark business model trade transactions to be treated are predominantly following several:
One, new order (New-Order): client inputs a new order transaction,
Two, delivery operation (Payment): update clients' accounts remaining sum to reflect its payment status,
Three, it delivers (Delivery): delivery (simulation batch processing transaction),
Four, order status inquiries (Order-Status): the state of inquiry client's last sale,
Five, inventory status query (Stock-Level): inquiry warehouse inventory situation, so as to replenish in time.
5. it is according to claim 3 based on flow indicator to database hardware resource analysis method, it is characterized in that: TPC-
The Warehouse module of C benchmark business model is respectively set to and District module and Stock module relation, Item module
It is set as being set as setting respectively with Customer module relation, Customer module with Stock module relation, District module
Be set to History module and Order module relation, module is set as and New-Order module and Order- Order respectively
Line module relation, Stock module pass are set as and Order-Line module relation.
6. it is according to claim 3 based on flow indicator to database hardware resource analysis method, it is characterized in that: according to
Inspection day is filled in inquiry inspection plan that core business scene-inspector's log of the sale monopoly operation system of practical business is filled in
Will (head row table) returns to inspection log list, and four sql are extracted, and forms one group of data bank service.
7. it is according to claim 6 based on flow indicator to database hardware resource analysis method, it is characterized in that: test
Method:
One, TPCC is tested
TPC-C test content: db transaction processing test is measured at server and database software processing online query transaction
Manage the performance of (OLTP).Regular TPC-C test result publication must provide tpmC value, i.e., complete how many TPC- per minute
Trade (TPC-C Transaction Per Minute) in C data library.
Two, testing tool
Test process is tested using benchmarksql-4.1.1 tool, using the combinations of transactions and weighted value of standard,
It is as follows:
Three, environment is tested
Two kinds of situations of test procedure test, a kind of situation are that pressure surveys single database maximum tpmC value, another situation is that
Pressure surveys server in multiple databases, and the tpmC value maximum upper limit of server is tested.
The case where test is all using 100 warehouses of initialization,
Configuration file:
8. it is according to claim 3 based on flow indicator to database hardware resource analysis method, it is characterized in that: choosing
Three server As, server B and server C in special control system, by pressure test, the maximum tpmC value for obtaining server A is
553617.49, the maximum tpmC value of server B is 382983.53, the maximum tpmC value of server C is 779081,
The inquiry inspection plan filled according to core business scene-inspector's log of the sale monopoly operation system of practical business is filled out
It writing inspection log (head row table), returns to inspection log list, four sql are extracted, one group of data bank service is formed, (how
It is converted into new order (New-Order) in TPC-C benchmark business model, delivery operation (Payment), delivery
(Delivery), order status inquiries (Order-Status), inventory status query (Stock-Level)), by TPC-C benchmark
Business model, the tpmC value for obtaining the practical business of server A is 134209.8, the tpmC value of the practical business of server B is
92092.7, the tpmC value of the practical business of server C is 200847.8, is obtained by calculation, and the server A of system is specially controlled
Complexity factor is 4.12, and the complexity factor for specially controlling the server B of system is 4.15, specially controls the complexity of the server C of system
Spending coefficient is 3.88, is obtained by carrying out complexity factor mean value calculation, and the complexity factor for specially controlling system is 4, works as expectation
When the tpmC needed is 10000, the tpmC value of the DATABASE HARDWARE needed to configure is 40000.
9. it is according to claim 3 based on flow indicator to database hardware resource analysis method, it is characterized in that: first existing
Pressure test is done with benchmark model on testing service device, (because of the expansion of the cpu for server under cpu full load situation
Malleability is comparatively relatively low, and the scalability of disk and memory is higher, is contemplated that expansion when disk and memory first bottleneck out
Open up disk and memory), to obtain benchmark tpmC value, it is assumed that m；Then core business scenario sql in the system that will be disposed again
It extracts to form data bank service, then carries out pressure test with same server, to obtain business tpmC value, it is assumed that be n.
Calculate the coefficient of m/n, i.e. complexity of the practical business relative to benchmark business, it is assumed that be x.According to a certain of actual business requirement
Time point maximum concurrency, it is assumed that be y, only when the tpmC value for the server bought meets xy, this server could expire
The demand of our deployment of foot.Therefore, before purchasing server, our benchmark model of supply commercialization can be allowed to carry out server
Pressure test, the benchmark tpmC value obtained just or be slightly larger than xy when, that is, be regarded as the server for being suitble to us to purchase.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811506069.6A CN109739745A (en) | 2018-12-10 | 2018-12-10 | Based on flow indicator to database hardware resource analysis method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811506069.6A CN109739745A (en) | 2018-12-10 | 2018-12-10 | Based on flow indicator to database hardware resource analysis method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109739745A true CN109739745A (en) | 2019-05-10 |
Family
ID=66358815
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811506069.6A Pending CN109739745A (en) | 2018-12-10 | 2018-12-10 | Based on flow indicator to database hardware resource analysis method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109739745A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112732544A (en) * | 2021-01-14 | 2021-04-30 | 东莞理工学院 | Computer hardware adaptation intelligent analysis system |
| CN113312107A (en) * | 2021-06-16 | 2021-08-27 | 成都市以太节点科技有限公司 | Security defense resource configuration method and system based on tpMC |
| CN113886256A (en) * | 2021-10-14 | 2022-01-04 | 广州九尾信息科技有限公司 | Method for monitoring and training model of software quality |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7703079B1 (en) * | 2005-05-03 | 2010-04-20 | Oracle America, Inc. | System performance prediction |
| CN102253883A (en) * | 2011-07-29 | 2011-11-23 | 国网信息通信有限公司 | Method and system for evaluating performance of server |
| CN102411515A (en) * | 2011-07-29 | 2012-04-11 | 国网信息通信有限公司 | Method and system for estimating capacity of server |
| CN103399797A (en) * | 2013-07-19 | 2013-11-20 | 华为技术有限公司 | Server resource allocation method and device |
| CN105224536A (en) * | 2014-05-29 | 2016-01-06 | 国际商业机器公司 | The method and apparatus of partition database |
| CN108228462A (en) * | 2018-01-08 | 2018-06-29 | 马上消费金融股份有限公司 | parameter testing method and device for O L TP system |
| CN108228740A (en) * | 2017-12-15 | 2018-06-29 | 国网青海省电力公司信息通信公司 | Electric power full-service uniform data centre data analysis domain comparing tool |
| CN108345537A (en) * | 2018-01-26 | 2018-07-31 | 北京计算机技术及应用研究所 | The concurrently execution control method of TPC-C benchmark test tools |
-
2018
- 2018-12-10 CN CN201811506069.6A patent/CN109739745A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7703079B1 (en) * | 2005-05-03 | 2010-04-20 | Oracle America, Inc. | System performance prediction |
| CN102253883A (en) * | 2011-07-29 | 2011-11-23 | 国网信息通信有限公司 | Method and system for evaluating performance of server |
| CN102411515A (en) * | 2011-07-29 | 2012-04-11 | 国网信息通信有限公司 | Method and system for estimating capacity of server |
| CN103399797A (en) * | 2013-07-19 | 2013-11-20 | 华为技术有限公司 | Server resource allocation method and device |
| CN105224536A (en) * | 2014-05-29 | 2016-01-06 | 国际商业机器公司 | The method and apparatus of partition database |
| CN108228740A (en) * | 2017-12-15 | 2018-06-29 | 国网青海省电力公司信息通信公司 | Electric power full-service uniform data centre data analysis domain comparing tool |
| CN108228462A (en) * | 2018-01-08 | 2018-06-29 | 马上消费金融股份有限公司 | parameter testing method and device for O L TP system |
| CN108345537A (en) * | 2018-01-26 | 2018-07-31 | 北京计算机技术及应用研究所 | The concurrently execution control method of TPC-C benchmark test tools |
Non-Patent Citations (1)
| Title |
|---|
| VINCENT: "TPC-C基准测试简介", 《HTTP://BLOG.SINA.COM.CN/S/BLOG_4485748101019WSH.HTML》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112732544A (en) * | 2021-01-14 | 2021-04-30 | 东莞理工学院 | Computer hardware adaptation intelligent analysis system |
| CN112732544B (en) * | 2021-01-14 | 2022-08-02 | 东莞理工学院 | Computer hardware adaptation intelligent analysis system |
| CN113312107A (en) * | 2021-06-16 | 2021-08-27 | 成都市以太节点科技有限公司 | Security defense resource configuration method and system based on tpMC |
| CN113312107B (en) * | 2021-06-16 | 2022-05-13 | 成都市以太节点科技有限公司 | Security defense resource configuration method and system based on tpMC |
| CN113886256A (en) * | 2021-10-14 | 2022-01-04 | 广州九尾信息科技有限公司 | Method for monitoring and training model of software quality |
| CN113886256B (en) * | 2021-10-14 | 2023-10-10 | 广州九尾信息科技有限公司 | Method for training model by software quality monitoring |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Brown et al. | Measuring the sources of marketing channel power: A comparison of alternative approaches | |
| Daugherty et al. | Leveraging logistics/distribution capabilities: The Effect of logistics service on market share. | |
| US7813949B2 (en) | Method and system for flexible budgeting in a purchase order system | |
| US7742948B2 (en) | Method of and system for allocating an OTB-relevant purchasing contract | |
| Ståhl Elvander et al. | Framework for characterizing the design of VMI systems | |
| JP6231204B2 (en) | Generating similarity between items | |
| US20020147630A1 (en) | Assortment decisions | |
| CN109739745A (en) | Based on flow indicator to database hardware resource analysis method | |
| US20150052019A1 (en) | System and Method for Multiple Weighted Factor Routing Schemes in Heterogeneous Fulfillment Networks Serving Multiple Clients with Distinct Routing Policies | |
| JP7402794B2 (en) | Fast learning recommendation system for product baskets | |
| US11762819B2 (en) | Clustering model analysis for big data environments | |
| Lin et al. | An integrated method for finding key suppliers in SCM | |
| US7174304B1 (en) | System and method for estimating product distribution using a product specific universe | |
| US9916563B1 (en) | Version recall for computerized catalog management | |
| Razi et al. | An applied model for improving inventory management in ERP systems | |
| US9704120B2 (en) | Inventory balance common object | |
| US20140180853A1 (en) | Framework for Generating a Personalized Item List | |
| Hormby et al. | Marriott International increases revenue by implementing a group pricing optimizer | |
| Lin et al. | Combined pricing and supply chain operations under price-dependent stochastic demand | |
| Kharvi et al. | An optimal inventory policy for subsequent price reduction problem | |
| US11868957B1 (en) | System and method of anomaly detection using machine learning and a local outlier factor | |
| KR102222383B1 (en) | Providing method, apparatus and computer-readable medium of online ordering of food materials | |
| CN108228462A (en) | parameter testing method and device for O L TP system | |
| CN112330235A (en) | Data processing method and device for inventory management, electronic equipment and medium | |
| Singha et al. | Computational experiment of methods to determine periodic (R, Q) inventory policy parameters: a case study of information decentralised distribution network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |