CN107423183A - A kind of GTX series video card calculates the applied voltage test method of performance - Google Patents
A kind of GTX series video card calculates the applied voltage test method of performance Download PDFInfo
- Publication number
- CN107423183A CN107423183A CN201710277696.6A CN201710277696A CN107423183A CN 107423183 A CN107423183 A CN 107423183A CN 201710277696 A CN201710277696 A CN 201710277696A CN 107423183 A CN107423183 A CN 107423183A
- Authority
- CN
- China
- Prior art keywords
- applied voltage
- voltage test
- video card
- cuda
- performance
- 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
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
- G06F11/26—Functional testing
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)
- Test And Diagnosis Of Digital Computers (AREA)
Abstract
The present invention discloses a kind of applied voltage test method that GTX series video card calculates performance, is related to computer big data field, and the applied voltage test of performance is calculated for carrying out GTX series video card, and operating system is linux system on test machine, utilizes NVIDIA CUDA computing platforms；Its specific implementation step includes：Video driver and CUDA programs are installed, calculate applied voltage test, and PCIE bandwidth tests.The method of the invention is applied to test video card capabilities, compatibility and radiating；Realize that GTX series video card calculates the applied voltage test of performance by the inventive method, while the rotation speed of the fan in video card pressure process, temperature and power consumption can be monitored in real time.
Description
Technical field
The present invention relates to computer big data field, the pressurization that specifically a kind of GTX series video card calculates performance is surveyed
Method for testing.
Background technology
GTX series video card is NV desktops level game video card, instantly as video card GPU architecture constantly upgrades, new Pascal
GTX1080 the and NVIDIA TITANX video cards of framework can provide performance, efficiency and the game experiencing of brilliance.Even, NVIDIA
TITANX performances are expressed as：" NVIDIA TITANX are a shocking video cards, and no matter what you will do, and this is started
Property TITANX video cards can all assign that you complete over you think the ability of complete impossible thing ".
GTX1080 NVIDA CUDA cores number is that 2560, Base clock are 1607MHz, and video memory is 8GB GDDR5X.
The NVIDIA CUDA cores number of TITANX video cards is that 3584, Base clock are 1417MHz, and video memory is 12G GDDR5X.From master
Want in parameter as can be seen that although GTX video cards are defined as game card, can equally to realize the calculating performance of superelevation；Using
Pascal frameworks, compared to the M series GPU of present enterprise market, architecturally with primacy；Simultaneously compared to performance
GPU card is cheap.At present, this GTX series video card increasingly becomes the hot topic in deep learning market, can be with very cheap
Price help client's fast construction deep learning business platform, big customer can be by the video card in some non-key class business
Upper reduction TCO.
Under traditional approach, video card applied voltage test software is Furmark in graphics use.But instantly, in high-performance and mutually
When networked client has more demand, GTX video cards are applied under non-graphic purposes deep learning application scenarios, it is impossible to above-mentioned test
The method of video card carries out pressure test, while NVIDIA is without the pressurizing tool for specially corresponding to test GTX video cards.
The content of the invention
Demand and weak point of the present invention for the development of current technology, there is provided a kind of GTX series video card calculates performance
Applied voltage test method.
A kind of GTX series video card of the present invention calculates the applied voltage test method of performance, solves above-mentioned technical problem and uses
Technical scheme it is as follows：The GTX series video card calculates the applied voltage test method of performance, is calculated for carrying out GTX series video card
The applied voltage test of performance, operating system is linux system on test machine, utilizes NVIDIA CUDA computing platforms；It is specific real
Applying step includes：Video driver and CUDA programs are installed, calculate applied voltage test, and PCIE bandwidth tests.
Preferably, the GTX series video card is GTX1080 video cards；Operating system linux system on test machine, select
redhat6.6。
Preferably, the video driver version is elected as：367.27 CUDA program versions are elected as：cuda_8.0.
A kind of GTX series video card of the present invention calculates the applied voltage test method and method of performance, compared with prior art
Have an advantageous effect in that：The present invention is applied to test video card capabilities, compatibility and radiating, for GTX series video cards non-
Application on graphics use, realize GTX series video card and calculate the applied voltage test of performance, while video card pressurization can be monitored in real time
During rotation speed of the fan, temperature and power consumption.
Embodiment
For the object, technical solutions and advantages of the present invention are more clearly understood, below in conjunction with specific embodiment, to this hair
The applied voltage test method that a kind of bright GTX series video card calculates performance is further described.
The present invention proposes a kind of applied voltage test method that GTX series video card calculates performance, for carrying out GTX series video cards
The applied voltage test of performance is calculated, operating system is linux system on test machine, utilizes NVIDIA CUDA computing platforms；It has
Body implementation steps include：Video driver and CUDA programs are installed, calculate applied voltage test, and three portions of PCIE bandwidth tests
Point.Overcome under non-graphic purposes deep learning application scenarios, it is impossible to use test software
Furmark carries out the problem of GTX series video card applied voltage tests, realizes that GTX series video card calculates the pressurization survey of performance
Examination.
The CUDA is a kind of universal parallel computing architecture released by NVIDIA, and the framework makes GPU can solve the problem that complexity
Computational problem.
Embodiment：
GTX series video card described in the present embodiment calculates the applied voltage test method of performance, by taking GTX1080 video cards as an example, carries out
GTX series video card calculates the applied voltage test of performance；Operating system linux system on test machine, by taking redhat6.6 as an example；Download
GTX1080 video drivers and CUDA programs.
In the present embodiment, video driver version is：367.27 CUDA program versions are：cuda_8.0.
GTX series video card described in the present embodiment calculates the applied voltage test method of performance, and its specific implementation step includes：Installation
Video driver and CUDA programs, calculate applied voltage test, and three parts of PCIE bandwidth tests.
The installation video driver and CUDA programs, detailed process are as follows：
1) GTX1080 video cards are correctly installed, run video driver；
#lspci is run afterwards | grep nVdia, it can be seen that video card information and driving version；
2) CUDA_8.0.44_Linux.run is installed, after installation, into/root/NVIDIA_CUDA-8.0_
Samples/5_Simulations/nbody catalogues；
The calculating applied voltage test, detailed process are as follows：
1) make orders generation executable file is performed；
2) ./nbody-benchmark-device=0-numbodies=10000000 orders are run
Here-device=0 represents the 0th piece of GPU, it is necessary to one order of each piece of card start-up when more cards pressurize；
3) using nvidia-smi orders observation GPU temperature and power consumption numerical value；
Because GPU_boost functions, GPU has a target temperature, after reaching the temperature, GPU locking masters
Frequency is unable to reach power consumption limit value in base frequencies, performance actual power loss.
The PCIE bandwidth tests, detailed process are as follows：
1) entrance/root/NVIDIA_CUDA-8.0_Samples/1_Utilities/bandwidthTest；
2) ./bandwidthTest is run；Test is not interrupted, smoothly terminated, and Result=PASS on screen.
Above-mentioned embodiment is only the specific case of the present invention, and scope of patent protection of the invention includes but is not limited to
Above-mentioned embodiment, any person of an ordinary skill in the technical field that meet claims of the present invention and any
The appropriate change or replacement done to it, it should all fall into the scope of patent protection of the present invention.
Claims (6)
1. a kind of GTX series video card calculates the applied voltage test method of performance, it is characterised in that based on carrying out GTX series video cards
The applied voltage test of performance is calculated, operating system is linux system on test machine, utilizes NVIDIA CUDA computing platforms；Its is specific
Implementation steps include：Video driver and CUDA programs are installed, calculate applied voltage test, and PCIE bandwidth tests.
2. a kind of GTX series video card calculates the applied voltage test method of performance according to claim 1, it is characterised in that described
GTX series video card is GTX1080 video cards；Operating system linux system on test machine, from redhat6.6.
3. a kind of GTX series video card calculates the applied voltage test method of performance according to claim 2, it is characterised in that described
Video driver version is elected as：367.27 CUDA program versions are elected as：cuda_8.0.
4. a kind of GTX series video card calculates the applied voltage test method of performance according to claim 3, it is characterised in that described
Installation video driver and CUDA programs, detailed process include：
1) GTX1080 video cards are installed, run video driver；
#lspci is run afterwards | grep nVdia, it can be seen that video card information and driving version；
2) CUDA_8.0.44_Linux.run is installed, after installation, into/root/NVIDIA_CUDA-8.0_Samples/
5_Simulations/nbody catalogues.
5. a kind of GTX series video card calculates the applied voltage test method of performance according to claim 4, it is characterised in that described
Applied voltage test is calculated, detailed process includes：
1) make orders generation executable file is performed；
2) ./nbody-benchmark-device=0-numbodies=10000000 orders are run；
- device=0 represents the 0th piece of GPU, it is necessary to one order of each piece of card start-up when more cards pressurize；
3) using nvidia-smi orders observation GPU temperature and power consumption numerical value.
6. a kind of GTX series video card calculates the applied voltage test method of performance according to claim 5, it is characterised in that described
PCIE bandwidth tests, detailed process include：
1) entrance/root/NVIDIA_CUDA-8.0_Samples/1_Utilities/bandwidthTest；
2) ./bandwidthTest is run；Test is not interrupted, smoothly terminated, and Result=PASS on screen.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710277696.6A CN107423183A (en) | 2017-04-25 | 2017-04-25 | A kind of GTX series video card calculates the applied voltage test method of performance |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710277696.6A CN107423183A (en) | 2017-04-25 | 2017-04-25 | A kind of GTX series video card calculates the applied voltage test method of performance |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107423183A true CN107423183A (en) | 2017-12-01 |
Family
ID=60424288
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710277696.6A Pending CN107423183A (en) | 2017-04-25 | 2017-04-25 | A kind of GTX series video card calculates the applied voltage test method of performance |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107423183A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109002155A (en) * | 2018-10-10 | 2018-12-14 | 郑州云海信息技术有限公司 | A kind of GPU server radiating regulation method and system |
| CN109491871A (en) * | 2018-11-20 | 2019-03-19 | 浪潮电子信息产业股份有限公司 | Method and device for acquiring equipment information of GPU |
| CN109684144A (en) * | 2018-12-26 | 2019-04-26 | 郑州云海信息技术有限公司 | A kind of method and device of GPU-BOX system testing |
| CN110175096A (en) * | 2019-05-21 | 2019-08-27 | 苏州浪潮智能科技有限公司 | A kind of GPU applied voltage test method, system, terminal and storage medium |
| CN114138579A (en) * | 2021-11-29 | 2022-03-04 | 苏州浪潮智能科技有限公司 | Prometous-based GPU (graphics processing Unit) interactive test method, device, equipment and readable medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102541679A (en) * | 2011-12-30 | 2012-07-04 | 曙光信息产业股份有限公司 | Method and system for testing GPU (graphic processing unit) cards |
| CN102567166A (en) * | 2011-12-30 | 2012-07-11 | 曙光信息产业股份有限公司 | Testing method and testing system of graphics card |
| CN104268046A (en) * | 2014-10-17 | 2015-01-07 | 浪潮电子信息产业股份有限公司 | Linux manual interaction NVIDIA GPU automatic testing method |
-
2017
- 2017-04-25 CN CN201710277696.6A patent/CN107423183A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102541679A (en) * | 2011-12-30 | 2012-07-04 | 曙光信息产业股份有限公司 | Method and system for testing GPU (graphic processing unit) cards |
| CN102567166A (en) * | 2011-12-30 | 2012-07-11 | 曙光信息产业股份有限公司 | Testing method and testing system of graphics card |
| CN104268046A (en) * | 2014-10-17 | 2015-01-07 | 浪潮电子信息产业股份有限公司 | Linux manual interaction NVIDIA GPU automatic testing method |
Non-Patent Citations (2)
| Title |
|---|
| 52NLP: "深度学习主机环境配置：Ubuntu16.04+Nvidia GTX 1080+CUDA8.0", 《HTTP://WWW.52NLP.CN/%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0%E4%B8%BB%E6%9C%BA%E7%8E%AF%E5%A2%83%E9%85%8D%E7%BD%AE-UBUNTU-16-04-NVIDIA-GTX-1080-CUDA-8》 * |
| PHANTOMVK: "深度学习环境配置：Ubuntu16.04&CUDA8.0", 《HTTPS://WWW.JIANSHU.COM/P/B043737C375C》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109002155A (en) * | 2018-10-10 | 2018-12-14 | 郑州云海信息技术有限公司 | A kind of GPU server radiating regulation method and system |
| CN109491871A (en) * | 2018-11-20 | 2019-03-19 | 浪潮电子信息产业股份有限公司 | Method and device for acquiring equipment information of GPU |
| CN109684144A (en) * | 2018-12-26 | 2019-04-26 | 郑州云海信息技术有限公司 | A kind of method and device of GPU-BOX system testing |
| CN109684144B (en) * | 2018-12-26 | 2021-11-02 | 郑州云海信息技术有限公司 | Method and device for testing GPU-BOX system |
| CN110175096A (en) * | 2019-05-21 | 2019-08-27 | 苏州浪潮智能科技有限公司 | A kind of GPU applied voltage test method, system, terminal and storage medium |
| CN114138579A (en) * | 2021-11-29 | 2022-03-04 | 苏州浪潮智能科技有限公司 | Prometous-based GPU (graphics processing Unit) interactive test method, device, equipment and readable medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107423183A (en) | A kind of GTX series video card calculates the applied voltage test method of performance | |
| US11314499B2 (en) | Simulating end-to-end upgrade process in production environment | |
| Graziano | A performance analysis of Xen and KVM hypervisors for hosting the Xen Worlds Project | |
| US20190087307A1 (en) | Method and apparatus for developing unmanned vehicle application | |
| CN105814541B (en) | The method of computer equipment and computer equipment internal storage starting | |
| US20130152074A1 (en) | Method for automatic consecutive installing operating systems | |
| CN102023889A (en) | Method for realizing operation of X86 VBIOS (video-bios) on MIPS (million instruction per second) framework | |
| US20180046486A1 (en) | Cloud based platform simulation for management controller development | |
| CN103488513B (en) | Equipment plug and play general drive method | |
| CN107608674A (en) | One kind applies skin change method, device, computer and storage medium | |
| CN106648827A (en) | Method for online adding virtual machine resources | |
| CN110297771A (en) | Distributed test method and device for the warehouse receipt page | |
| US10318343B2 (en) | Migration methods and apparatuses for migrating virtual machine including locally stored and shared data | |
| CN112199273A (en) | Virtual machine pressure/performance testing method and system | |
| CN111123737A (en) | Digital hardware virtual simulation system | |
| CN105980957B (en) | Power monitoring system for virtual platform simulation | |
| US10268466B2 (en) | Software installer with built-in hypervisor | |
| CN106919407B (en) | The method that computer embeds product information synchronized update | |
| CN110414220A (en) | Method and device for extracting operation files in dynamic execution process of program in sandbox | |
| CN105045694A (en) | Method for automatically testing Nitrox accelerator card | |
| CN107016148B (en) | Enterprise protocol management | |
| US9239893B2 (en) | Method and process computer for computation of the state variables of a hybrid differential-algebraic process model | |
| CN109308196B (en) | Multi-model equipment multiplexing method and device and storage medium | |
| US9552267B2 (en) | SATA receiver equalization margin determination/setting method and apparatus | |
| Kakkar et al. | Risk analysis in mobile application development |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20171201 |
|
| RJ01 | Rejection of invention patent application after publication |