CN111752779A - Hard disk performance test optimization method and device - Google Patents
Hard disk performance test optimization method and device Download PDFInfo
- Publication number
- CN111752779A CN111752779A CN202010534655.2A CN202010534655A CN111752779A CN 111752779 A CN111752779 A CN 111752779A CN 202010534655 A CN202010534655 A CN 202010534655A CN 111752779 A CN111752779 A CN 111752779A
- Authority
- CN
- China
- Prior art keywords
- hard disk
- performance test
- test
- performance
- cpu
- 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.)
- Withdrawn
Links
- 238000011056 performance test Methods 0.000 title claims abstract description 186
- 238000000034 method Methods 0.000 title claims abstract description 75
- 238000005457 optimization Methods 0.000 title claims abstract description 8
- 238000012360 testing method Methods 0.000 claims abstract description 80
- 230000008569 process Effects 0.000 claims abstract description 55
- 238000004458 analytical method Methods 0.000 claims description 16
- 230000005540 biological transmission Effects 0.000 abstract description 8
- 239000007787 solid Substances 0.000 description 16
- 238000004891 communication Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 6
- 230000007246 mechanism Effects 0.000 description 5
- 230000015654 memory Effects 0.000 description 4
- 230000007704 transition Effects 0.000 description 4
- 238000011156 evaluation Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000003860 storage Methods 0.000 description 3
- 238000007405 data analysis Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000012938 design process Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/22—Detection or location of defective computer hardware by testing during standby operation or during idle time, e.g. start-up testing
- G06F11/2205—Detection or location of defective computer hardware by testing during standby operation or during idle time, e.g. start-up testing using arrangements specific to the hardware being tested
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/22—Detection or location of defective computer hardware by testing during standby operation or during idle time, e.g. start-up testing
- G06F11/26—Functional testing
- G06F11/263—Generation of test inputs, e.g. test vectors, patterns or sequences ; with adaptation of the tested hardware for testability with external testers
Landscapes
- Engineering & Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Quality & Reliability (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Debugging And Monitoring (AREA)
Abstract
The invention provides a hard disk performance test optimization method, which comprises the following steps: configuring a hard disk performance test environment; performing write preheating test on the hard disk for many times until the hard disk is in a stable state; binding CPUs corresponding to different hard disk performance test processes; setting performance test parameters, performing performance test by using a test tool, and continuously polling the hard disk IO read-write result by a CPU in the performance test process; the invention also provides a hard disk performance test optimizing device, which effectively solves the problem of performance fluctuation caused by the fact that a system is allocated to different physical cores when a test process is carried out and the problem of inaccurate test result in the hard disk performance test caused by the configuration of a transmission link of hard disk and CPU signals and the configuration of a BIOS and a CPU, and effectively improves the reliability and the accuracy of the hard disk performance test.
Description
Technical Field
The invention relates to the field of hard disk testing, in particular to a method and a device for optimizing hard disk performance testing.
Background
Big data analysis gradually becomes a hot topic in the IT industry, and the traditional storage system needs very low time delay when being applied to big data or data mining application of mass data storage. In order to ensure that the big data analysis service can normally operate, the corresponding storage system needs to be fast enough to fully exert the due fast performance.
Therefore, in the design and production process of the server, the test of the performance of the hard disk also becomes a very important ring, and in the actual test process of the performance of the hard disk, the factors influencing the performance of the hard disk are more, generally, the hard disk performance test process is carried out by a system to be allocated to different physical cores, and the problems of transmission links of signals between the hard disk and a CPU, the configuration of bios and the CPU and the like are not considered.
On one hand, the performance of the hard disk can be allocated to different physical cores by a system when a performance test process is carried out, and the problem of performance fluctuation can be caused; on the other hand, problems in aspects such as transmission links of signals between the hard disk and the CPU, configuration of the BIOS and the CPU and the like are not considered, so that the problem that a test result in a hard disk performance test is not accurate enough is caused, and the improvement of accuracy and smoothness of the hard disk performance test is not facilitated.
Disclosure of Invention
The invention aims to solve the problems in the prior art, innovatively provides a method and a device for optimizing the performance test of a hard disk, effectively solves the problem of performance fluctuation caused by the fact that a system is allocated to different physical cores when a test process is carried out and the problem of inaccurate test results in the performance test of the hard disk caused by the configuration of a transmission link of signals between the hard disk and a CPU and the configuration of a BIOS and the CPU, and effectively improves the reliability and the accuracy of the performance test of the hard disk.
The invention provides a method for optimizing a hard disk performance test, which comprises the following steps:
configuring a hard disk performance test environment;
performing write preheating test on the hard disk for many times until the hard disk is in a stable state;
binding CPUs corresponding to different hard disk performance test processes;
setting performance test parameters, performing performance test by using a test tool, and continuously polling the hard disk IO read-write result by a CPU in the performance test process;
and acquiring multiple performance test results of the hard disk, and analyzing the test results.
Optionally, the configuring the hard disk performance test environment includes:
the hard disk is directly connected with the CPU through a PCle interface;
closing the CPU hyper-thread, forbidding the C state and all power saving modes in the BIOS, and closing the Turbo mode;
selecting the latest stable version system kernel under the OS;
configuring a CPU as a performance mode;
IRQ interrupt service is disabled.
Optionally, the binding of the CPUs corresponding to the different hard disk performance test processes is specifically to bind the different hard disk performance test processes with a plurality of physical cores in the corresponding CPUs.
Optionally, obtaining multiple performance test results of the hard disk, and analyzing the test results specifically includes:
obtaining multiple performance test results of the hard disk, judging whether the difference value between the maximum value and the minimum value of the same performance test parameter in the test process in the single performance test result is larger than a first threshold value, if so, failing the stability of the performance test, and if not, passing the stability of the performance test.
Further, the first threshold is 20% of an average value in the process of testing the same performance testing parameter in a single performance test of the hard disk.
Optionally, obtaining multiple performance test results of the hard disk, and analyzing the test results specifically includes:
obtaining a multiple performance test result of the hard disk, and establishing a steady state value formula:
wherein x is the x-th performance test,
is the average value of the performance test times, y is the test result of the x performance test,
the average value of the test results of multiple performance tests is slope (y) which is the x-th steady state value, whether the difference value of the maximum value and the minimum value of the slope (y) is greater than a second threshold value is judged, if so, the stability of the multiple performance tests fails, and if not, the multiple performance tests failThe test passed stability.
Further, the second threshold is 10% of an average value of steady-state values slope (y) in multiple performance tests of the hard disk.
The second aspect of the present invention provides a hard disk performance test optimization apparatus, including:
the configuration module is used for configuring a hard disk performance test environment;
the write preheating module is used for performing write preheating test on the hard disk for multiple times until the hard disk is in a stable state;
the binding module binds CPUs corresponding to different hard disk performance test processes;
the test module is used for setting performance test parameters, performing performance test by using a test tool, and continuously polling the hard disk IO read-write result by the CPU in the performance test process;
and the acquisition analysis module is used for acquiring a performance test result of the hard disk for multiple times and analyzing the test result.
Optionally, the acquisition analysis module includes:
the first obtaining and analyzing submodule obtains a performance test result of the hard disk for multiple times, judges whether the difference value of the maximum value and the minimum value of the same performance test parameter in the test process in the single performance test result is larger than a first threshold value, if so, the stability of the performance test fails, and if not, the stability of the performance test passes.
Further, the acquisition analysis module further comprises:
and the second acquisition and analysis submodule acquires a multiple performance test result of the hard disk and establishes a steady state value formula:
wherein x is the x-th performance test,
is the average value of the performance test times, y is the test result of the x performance test,
and (4) determining whether the difference value between the maximum value and the minimum value of slope (y) is greater than a second threshold value, if so, failing to test the stability of the hard disk for multiple times, and if not, passing the stability of the hard disk for multiple times of performance tests.
The technical scheme adopted by the invention comprises the following technical effects:
1. the invention effectively solves the problem of performance fluctuation caused by the fact that a system is allocated to different physical cores when a test process is carried out and the problem of inaccurate test result in the hard disk performance test caused by the configuration of the transmission link of signals of the hard disk and the CPU and the configuration of the BIOS and the CPU, and effectively improves the reliability and the accuracy of the hard disk performance test.
2. The invention carries out write preheating test on the hard disk for many times until the hard disk is in a stable state, thereby improving the accuracy of the hard disk performance test.
3. According to the invention, different hard disk performance test processes are bound with the plurality of physical cores in the corresponding CPU, so that the process is prevented from being distributed across CPUs when the system distributes resources, the communication delay is reduced, and the hard disk performance test efficiency is improved.
4. The invention continuously polls the hard disk IO read-write result by the CPU in the hard disk performance test process, thereby avoiding the time delay problem in the interrupt communication mechanism and improving the hard disk performance test efficiency.
5. The invention not only carries out stable evaluation on the single performance test of the hard disk, but also carries out stable analysis on the multiple performance tests of the hard disk, thereby further improving the stability of the performance test of the hard disk.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention, as claimed.
Drawings
In order to more clearly illustrate the embodiments or technical solutions in the prior art of the present invention, the drawings used in the description of the embodiments or prior art will be briefly described below, and it is obvious for those skilled in the art that other drawings can be obtained based on these drawings without any creative effort.
FIG. 1 is a schematic flow diagram of a process according to an embodiment of the present invention;
FIG. 2 is a schematic flow chart illustrating step S1 of a method according to an embodiment of the present invention;
fig. 3 is a schematic diagram of a connection structure between a CPU and a hard disk according to a first embodiment of the present invention;
FIG. 4 is a prior art QPI communication connection diagram between the CPU0 and the CPU 1;
fig. 5 is a schematic flow chart illustrating step S5 in a method according to an embodiment of the present invention;
fig. 6 is another schematic flow chart of step S5 in a method according to an embodiment of the present invention;
FIG. 7 is a schematic structural diagram of an apparatus according to a second embodiment of the present invention;
FIG. 8 is a schematic structural diagram of an acquisition analysis module in an apparatus according to a second embodiment of the present invention;
fig. 9 is another schematic structural diagram of an acquisition analysis module in the second apparatus according to the embodiment of the present invention.
Detailed Description
In order to clearly explain the technical features of the present invention, the following detailed description of the present invention is provided with reference to the accompanying drawings. The following disclosure provides many different embodiments, or examples, for implementing different features of the invention. To simplify the disclosure of the present invention, the components and arrangements of specific examples are described below. Furthermore, the present invention may repeat reference numerals and/or letters in the various examples. This repetition is for the purpose of simplicity and clarity and does not in itself dictate a relationship between the various embodiments and/or configurations discussed. It should be noted that the components illustrated in the figures are not necessarily drawn to scale. Descriptions of well-known components and processing techniques and procedures are omitted so as to not unnecessarily limit the invention.
Example one
As shown in fig. 1, the present invention provides a method for optimizing a hard disk performance test, including:
s1, configuring a hard disk performance test environment;
s2, performing write preheating test on the hard disk for many times until the hard disk is in a stable state;
s3, binding CPUs corresponding to different hard disk performance test processes;
s4, setting performance test parameters, using a test tool to perform performance test, and continuously polling the hard disk IO read-write result by the CPU in the performance test process;
and S5, acquiring multiple performance test results of the hard disk, and analyzing the test results.
In step S1, as shown in fig. 2, configuring the hard disk performance test environment includes:
s11, the hard disk is directly connected with the CPU through the PCle interface;
s12, closing the CPU hyper-thread, forbidding the C state and all power saving modes in the BIOS, and closing the Turbo mode;
s13, selecting the latest stable version system kernel under the OS;
s14, configuring the CPU as a performance mode;
s15, the IRQ interrupt service is disabled.
The step S1 is to configure the optimal test environment in terms of hardware resource configuration and BIOS, and does not cause problems in terms of hardware and the like.
In step S11, to improve the performance test result of the hard disk, a high performance CPU with a dominant frequency of 3.0GHz or higher is preferably used, and a CPU with a model of ES1QU98, a dominant frequency of 2.7G, a 24 physical core, and a power consumption of 220W are preferably used. In addition, in order to ensure the capacity of the memory, in the performance test of the matched 24 solid state disks, the adopted 16 32Gx3200MHz and 32Gx3200MHz matches can also be adjusted according to the number of disks tested at the same time, and the adjustment rule is as follows: the ratio of the number of the performance tests of the solid state disk to the number of the memories is greater than or equal to 0.5, and the ratio can be adjusted according to actual conditions, and the invention is not limited herein.
As shown in fig. 3, the solid state disk is directly connected to the CPU through the PCle interface during the performance test. If a solid state disk is connected through a PCH (Platform Controller Hub), a delay is generated on a DMI (Direct media interface) bus to affect performance, and the solid state disk is directly connected to a CPU through a pce interface, so that delay of DMI transmission can be reduced, and efficiency of the hard disk can be effectively optimized.
Step S12 is to close the CPU hyper-thread for BIOS configuration, disable the C state (power saving mode state) and any power saving mode in BIOS, and close the Turbo mode, but the Turbo mode (high performance mode) can improve the delay, but has an impact on the service quality.
In step S13, under an OS (operating system), the latest kernel (operating system kernel) of the official website is selected, and the new kernel provides the latest IO driver, which can effectively improve performance.
In step S14, the CPU is configured to be in the performance mode, and all physical cores of the CPU are configured to be in the performance mode, so that the CPU can be guaranteed to run at the highest frequency, and the end-to-end delay can be reduced.
In step S15, an IRQ (Interrupt ReQuest) Interrupt service is disabled, and an IRQ balancer is responsible for balancing interrupts between cores, specifically implemented by commanding a systemctl stop irqbalance, and disabling this service can better control Interrupt management across CPU physical cores.
In step S2, the Performance of the hard disk is based on a PTS (Performance Test Specification) proposed by the SNIA (Storage Networking Industry Association), and the hard disk undergoes three stages from the start of the Test to the end, where the three states of FOB (Fresh Out of Box), Transition (Transition), and Steady State are the most stable and accurate values when the Performance Test is performed under the Steady State.
The hard disk works in the FOB state, the number of idle areas of the hard disk is large, and the operation such as garbage collection is less, so the writing speed is relatively in the peak value, and the period of time is about 0-50 min.
However, the running time is in transition as long as possible, the data arrangement is more random, the frequency of operation triggering such as garbage collection and block (data block) erasure is high, the overall performance is in a slow-sliding state, and the time is about in the period of 50-250 min.
And finally, the whole Steady State tends to be stable, and the performance is usually measured by the indexes of daily performance, stability and the like of the solid State disk, so that the solid State disk needs to be in a stable State by write preheating.
When the fio script is used for carrying out the hard disk performance test, the preprocessing of the script is to write a data block until the capacity of a written disk is full, and the time is only about 2 hours, namely about 120min, so that the disk is still in a transition state; the measured data will not be accurate.
Therefore, loop circulation is added into the fio to complete multiple preheating operations, and the solid state disk is ensured to be in a stable state. Before the next write hot operation starts, the full disk deleting operation is carried out on the data written with the hot data.
The invention carries out write preheating test on the hard disk for many times until the hard disk is in a stable state, thereby improving the accuracy of the hard disk performance test.
In step S3, the CPU corresponding to the different hard disk performance test processes is bound specifically to the different hard disk performance test processes and the plurality of physical cores in the corresponding CPU. The solid state disk test process mounted to the CPU0 is bound to a physical core of the CPU0, and the solid state disk test process mounted to the CPU1 is bound to the CPU1, so that the solid state disk test process can be bound to the corresponding CPU, the process is prevented from being distributed across CPUs when a system allocates resources, and data is prevented from being transmitted through QPI (Quick path interconnect) (as shown in fig. 4), which causes delay and performance reduction. For example, the CPU0 and the CPU1 respectively include 24 physical cores, and include 48 physical cores in total, that is, two CPU logical cores are 96, and taking 24 hard disks as an example, each hard disk process is correspondingly bound with 4 logical cores to implement multi-core corresponding binding, and the specific implementation manner may be implemented by a taskset-c command.
According to the invention, different hard disk performance test processes are bound with the plurality of physical cores in the corresponding CPU, so that the process is prevented from being distributed across CPUs when the system distributes resources, the communication delay is reduced, and the hard disk performance test efficiency is improved.
In step S4, the test tool in the current mainstream hard disk performance test is implemented by FIO, and FIO2.18 or higher version can be used because it supports synchronous I/O read/write engine, thereby improving I/O performance when the thread is 1. When the FIO script is used for testing, the configuration of two parameters, namely nunjobs (thread number) and iododepth (queue depth), can bring a great influence on the performance of the hard disk, and the spec given by a hard disk manufacturer can generally give the optimal configuration of the two parameters according to the test. Numjobs is the number of threads, the default is 1, the concurrency can be improved by setting the number greater than 1, the bandwidth can be improved by increasing the depth within a certain range, but higher delay is caused by overhigh depth, and the same process is executed by using the configured number of threads; iodepth is the queue depth, defaulted to 128K.
When a polling mechanism is started for communication between the CPU and the hard disk to replace interruption, writing and reading of data are completed more quickly, and therefore the bandwidth performance of the hard disk is improved.
For reading and writing data, the default typical I/O interrupt access cycle process is: when data is requested for processing, the operating system kernel will check the data in main memory. If the main memory does not have the required data, starting physical I/O to the specified solid state disk device through the device driver, processing the command by the specified solid state disk and sending an interrupt when the command is completed, confirming the interrupt command by the device driver, and then finishing the cycle.
In the performance test process, the CPU continuously polls the hard disk IO read-write result and starts I/O polling to further improve the performance (an I/O polling mode is introduced by Linux Kernel 4.4, so the used system Kernel needs to be updated to be more than 4.4). In the polling mode, the CPU continuously checks whether the hard disk I/O command is completed, skips the traditional interrupt sending mechanism, and continuously polls whether the last write-in and read action is completed, so that the data can be more quickly and timely read and written, and the I/O performance of the solid state disk is improved.
The invention continuously polls the hard disk IO read-write result by the CPU in the hard disk performance test process, thereby avoiding the time delay problem in the interrupt communication mechanism and improving the hard disk performance test efficiency.
In step S5, as shown in fig. 5, the method specifically includes:
s51, obtaining multiple performance test results of the hard disk, judging whether the difference value between the maximum value and the minimum value of the same performance test parameter in the test process in the single performance test result is larger than a first threshold value, if so, executing a step S52, and if not, executing a step S53;
s52, the stability of the performance test fails;
s53, the stability of the performance test is passed.
Specifically, the same performance test parameter may be the read-write bandwidth of the solid state disk, or may be other performance parameters, the first threshold is 20% of an average value in the test process of the same performance test parameter in a single performance test of the hard disk, and may also be adjusted, which is not limited herein.
Further, as shown in fig. 6, step S5 further includes:
s54, obtaining the multiple performance test results of the hard disk, and establishing a steady state value formula:
wherein x is the x-th performance test,
is the average value of the performance test times, y is the test result of the x performance test,
the slope (y) is the x-th steady state value which is the average value of the test results of multiple performance tests;
s55, judging whether the difference value between the maximum value and the minimum value of slope (y) is larger than a second threshold value, if so, executing a step S56, and if not, executing a step S57;
s56, failing to test the stability of the performance for many times;
and S57, passing the stability of multiple performance tests.
Specifically, the same performance test parameter may be the read-write bandwidth of the solid state disk, or may be other performance parameters, the second threshold is 10% of the average value of the steady-state values slope (y) in the multiple performance tests of the hard disk, and may also be adjusted, which is not limited herein.
The invention effectively solves the problem of performance fluctuation caused by the fact that a system is allocated to different physical cores when a test process is carried out and the problem of inaccurate test result in the hard disk performance test caused by the configuration of the transmission link of signals of the hard disk and the CPU and the configuration of the BIOS and the CPU, and effectively improves the reliability and the accuracy of the hard disk performance test.
The invention carries out write preheating test on the hard disk for many times until the hard disk is in a stable state, thereby improving the accuracy of the hard disk performance test.
The invention not only carries out stable evaluation on the single performance test of the hard disk, but also carries out stable analysis on the multiple performance tests of the hard disk, thereby further improving the stability of the performance test of the hard disk.
Example two
As shown in fig. 7, the technical solution of the present invention further provides an optimization apparatus for testing hard disk performance, including:
the configuration module 101 is used for configuring a hard disk performance test environment;
the write preheating module 102 is used for performing write preheating tests on the hard disk for multiple times until the hard disk is in a stable state;
the binding module 103 binds CPUs corresponding to different hard disk performance test processes;
the test module 104 sets performance test parameters, performs performance test by using a test tool, and continuously polls the hard disk IO read-write result by the CPU in the performance test process;
and the acquisition and analysis module 105 acquires the multiple performance test results of the hard disk and analyzes the test results.
Further, as shown in fig. 8, the acquisition analysis module 105 includes:
the first obtaining and analyzing submodule 1051 obtains a performance test result of the hard disk for multiple times, judges whether the difference value between the maximum value and the minimum value of the same performance test parameter in the test process in the single performance test result is larger than a first threshold value, if so, the stability of the performance test fails, and if not, the stability of the performance test passes.
Further, as shown in fig. 9, the acquisition analysis module 105 further includes:
the second obtaining and analyzing sub-module 1052 obtains the multiple performance test results of the hard disk, and establishes a steady-state value formula:
wherein x is the x-th performance test,
is the average value of the performance test times, y is the test result of the x performance test,
and (4) determining whether the difference value between the maximum value and the minimum value of slope (y) is greater than a second threshold value, if so, failing to test the stability of the hard disk for multiple times, and if not, passing the stability of the hard disk for multiple times of performance tests.
The invention effectively solves the problem of performance fluctuation caused by the fact that a system is allocated to different physical cores when a test process is carried out and the problem of inaccurate test result in the hard disk performance test caused by the configuration of the transmission link of signals of the hard disk and the CPU and the configuration of the BIOS and the CPU, and effectively improves the reliability and the accuracy of the hard disk performance test.
The invention carries out write preheating test on the hard disk for many times until the hard disk is in a stable state, thereby improving the accuracy of the hard disk performance test.
According to the invention, different hard disk performance test processes are bound with the plurality of physical cores in the corresponding CPU, so that the process is prevented from being distributed across CPUs when the system distributes resources, the communication delay is reduced, and the hard disk performance test efficiency is improved.
The invention continuously polls the hard disk IO read-write result by the CPU in the hard disk performance test process, thereby avoiding the time delay problem in the interrupt communication mechanism and improving the hard disk performance test efficiency.
The invention not only carries out stable evaluation on the single performance test of the hard disk, but also carries out stable analysis on the multiple performance tests of the hard disk, thereby further improving the stability of the performance test of the hard disk.
Although the embodiments of the present invention have been described with reference to the accompanying drawings, it is not intended to limit the scope of the present invention, and it should be understood by those skilled in the art that various modifications and variations can be made without inventive efforts by those skilled in the art based on the technical solution of the present invention.
Claims (10)
1. A hard disk performance test optimization method is characterized by comprising the following steps:
configuring a hard disk performance test environment;
performing write preheating test on the hard disk for many times until the hard disk is in a stable state;
binding CPUs corresponding to different hard disk performance test processes;
setting performance test parameters, performing performance test by using a test tool, and continuously polling the hard disk IO read-write result by a CPU in the performance test process;
and acquiring multiple performance test results of the hard disk, and analyzing the test results.
2. The method of claim 1, wherein the configuring the hard disk performance test environment comprises:
the hard disk is directly connected with the CPU through a PCle interface;
closing the CPU hyper-thread, forbidding the C state and all power saving modes in the BIOS, and closing the Turbo mode;
selecting the latest stable version system kernel under the OS;
configuring a CPU as a performance mode;
IRQ interrupt service is disabled.
3. The method according to claim 1, wherein the binding of the CPUs corresponding to the different hard disk performance testing processes is specifically a binding of the different hard disk performance testing processes with a plurality of physical cores in the corresponding CPUs.
4. The method of claim 1, wherein the obtaining of the results of multiple performance tests of the hard disk and the analyzing of the results specifically comprise:
obtaining multiple performance test results of the hard disk, judging whether the difference value between the maximum value and the minimum value of the same performance test parameter in the test process in the single performance test result is larger than a first threshold value, if so, failing the stability of the performance test, and if not, passing the stability of the performance test.
5. The method according to claim 4, wherein the first threshold is 20% of an average value of the hard disk in a single performance test in the same performance test parameter test process.
6. The method of claim 4, wherein the obtaining of the results of multiple performance tests of the hard disk and the analyzing of the results specifically comprise:
obtaining a multiple performance test result of the hard disk, and establishing a steady state value formula:
wherein x is the x-th performance test,
is the average value of the performance test times, y is the test result of the x performance test,
slope being the mean of the test results of multiple performance testsAnd (y) judging whether the difference value between the maximum value and the minimum value of slope (y) is greater than a second threshold value, if so, failing to test the stability of multiple performance tests, and if not, passing the stability of the multiple performance tests.
7. The method according to claim 6, wherein the second threshold is 10% of an average value of steady-state values slope (y) in multiple performance tests of the hard disk.
8. The utility model provides a hard disk performance test optimizing apparatus which characterized by includes:
the configuration module is used for configuring a hard disk performance test environment;
the write preheating module is used for performing write preheating test on the hard disk for multiple times until the hard disk is in a stable state;
the binding module binds CPUs corresponding to different hard disk performance test processes;
the test module is used for setting performance test parameters, performing performance test by using a test tool, and continuously polling the hard disk IO read-write result by the CPU in the performance test process;
and the acquisition analysis module is used for acquiring a performance test result of the hard disk for multiple times and analyzing the test result.
9. The hard disk performance test optimization device of claim 8, wherein the acquisition analysis module comprises:
the first obtaining and analyzing submodule obtains a performance test result of the hard disk for multiple times, judges whether the difference value of the maximum value and the minimum value of the same performance test parameter in the test process in the single performance test result is larger than a first threshold value, if so, the stability of the performance test fails, and if not, the stability of the performance test passes.
10. The hard disk performance test optimization device of claim 9, wherein the acquisition analysis module further comprises:
and the second acquisition and analysis submodule acquires a multiple performance test result of the hard disk and establishes a steady state value formula:
wherein x is the x-th performance test,
is the average value of the performance test times, y is the test result of the x performance test,
and (4) determining whether the difference value between the maximum value and the minimum value of slope (y) is greater than a second threshold value, if so, failing to test the stability of the hard disk for multiple times, and if not, passing the stability of the hard disk for multiple times of performance tests.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010534655.2A CN111752779A (en) | 2020-06-12 | 2020-06-12 | Hard disk performance test optimization method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010534655.2A CN111752779A (en) | 2020-06-12 | 2020-06-12 | Hard disk performance test optimization method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111752779A true CN111752779A (en) | 2020-10-09 |
Family
ID=72675214
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010534655.2A Withdrawn CN111752779A (en) | 2020-06-12 | 2020-06-12 | Hard disk performance test optimization method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111752779A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113110960A (en) * | 2021-04-15 | 2021-07-13 | 山东英信计算机技术有限公司 | Automatic tuning test method and system based on hard disk performance |
| CN113868052A (en) * | 2021-09-18 | 2021-12-31 | 苏州浪潮智能科技有限公司 | IOPS test method, device and storage medium based on PID algorithm |
| WO2023087941A1 (en) * | 2021-11-18 | 2023-05-25 | 苏州浪潮智能科技有限公司 | Method and apparatus for accessing debugging space of solid state disk |
-
2020
- 2020-06-12 CN CN202010534655.2A patent/CN111752779A/en not_active Withdrawn
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113110960A (en) * | 2021-04-15 | 2021-07-13 | 山东英信计算机技术有限公司 | Automatic tuning test method and system based on hard disk performance |
| CN113868052A (en) * | 2021-09-18 | 2021-12-31 | 苏州浪潮智能科技有限公司 | IOPS test method, device and storage medium based on PID algorithm |
| CN113868052B (en) * | 2021-09-18 | 2023-07-18 | 苏州浪潮智能科技有限公司 | IOPS (input/output) testing method and device based on PID (proportion integration differentiation) algorithm and storage medium |
| WO2023087941A1 (en) * | 2021-11-18 | 2023-05-25 | 苏州浪潮智能科技有限公司 | Method and apparatus for accessing debugging space of solid state disk |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111752779A (en) | Hard disk performance test optimization method and device | |
| US11474712B2 (en) | Method, apparatus, device and storage medium for managing access request | |
| WO2023066304A1 (en) | Job running parameter optimization method applied to super-computing cluster scheduling | |
| CN109828874B (en) | Automatic testing method for PSU current sharing performance of server | |
| CN113176493A (en) | Chip test mainboard, test system and test method | |
| CN113110960A (en) | Automatic tuning test method and system based on hard disk performance | |
| US8516482B2 (en) | Virtual machine assigning method and storage medium thereof, information processing device having virtual machine environment | |
| US20170115915A1 (en) | Memory module monitoring memory operation and power management method thereof | |
| CN101441497B (en) | Information processing apparatus having memory clock setting function and memory clock setting method | |
| US20240127870A1 (en) | Configuring a host interface of a memory device based on mode of operation | |
| CN103543982B (en) | A kind of clock frequency management method and mobile device | |
| CN110781046B (en) | USB mobile storage equipment quality detection system, method and related components | |
| CN106776153B (en) | Job control method and server | |
| CN112463503B (en) | ARM architecture complete machine server heat dissipation capacity testing method and system | |
| CN115576759A (en) | Pressure testing method and device, electronic equipment and readable storage medium | |
| CN115576876A (en) | DMA (direct memory Access) allocation and management method applied to NVMe (network video disk) SSD (solid State disk) | |
| CN111752778B (en) | Method and device for testing stability of solid state disk firmware under linux | |
| WO2021147382A1 (en) | Test task execution method and device | |
| LU101773B1 (en) | Method for improving sequential read performance of solid-state drive | |
| US10324752B2 (en) | Response times based on application states | |
| CN108874546B (en) | Linux system and method for optimizing resource use | |
| CN110609768A (en) | Method and device for measuring xGMI2 bandwidth between two paths of CPUs | |
| CN109857715B (en) | Log recording method and system for strengthening error checking support | |
| CN115061558B (en) | Hot reset method and device of PCIE equipment, storage medium and PCIE equipment | |
| CN112860561B (en) | Automatic performance testing method and terminal device |
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 | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20201009 |
|
| WW01 | Invention patent application withdrawn after publication |