CN104951024B - A kind of large data all-in-one machine based on electric power application - Google Patents

A kind of large data all-in-one machine based on electric power application Download PDF

Info

Publication number
CN104951024B
CN104951024B CN201510429747.3A CN201510429747A CN104951024B CN 104951024 B CN104951024 B CN 104951024B CN 201510429747 A CN201510429747 A CN 201510429747A CN 104951024 B CN104951024 B CN 104951024B
Authority
CN
China
Prior art keywords
storage server
rack
measuring point
converting equipment
power
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.)
Active
Application number
CN201510429747.3A
Other languages
Chinese (zh)
Other versions
CN104951024A (en
Inventor
丁书耕
邢宏伟
王刚
王震
陈亮
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shandong Luruan Digital Technology Co Ltd
Original Assignee
Shandong Luneng Software Technology Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shandong Luneng Software Technology Co Ltd filed Critical Shandong Luneng Software Technology Co Ltd
Priority to CN201510429747.3A priority Critical patent/CN104951024B/en
Publication of CN104951024A publication Critical patent/CN104951024A/en
Application granted granted Critical
Publication of CN104951024B publication Critical patent/CN104951024B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Landscapes

  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention provides a kind of large data all-in-one machine based on electric power application, comprise rack, external server and one or more converting equipment measuring point sensor, also comprise the one or more bottoms being arranged at rack bottom and calculate storage server, be arranged at the power module in the middle part of rack and gigabit supervising the network module, be set in turn in the 10000000000 fiber data networking modules on rack top from top to bottom, one or more top calculates storage server and one or more high-performance calculation storage server, be arranged at fan and the fan governor of rear of cabinet, integrated level is high, equipment availability is high, favorable expandability, stable system performance.

Description

A kind of large data all-in-one machine based on electric power application
Technical field
The present invention relates to power equipment monitoring analysis application, be specifically related to a kind of large data all-in-one machine based on electric power application.
Background technology
At present; in order to ensure power grid security, stable, Effec-tive Function; usually converting equipment status monitoring can be carried out; the equipment measuring point data monitored is stored into real-time data base; and periodically by data conversion storage in real-time data base to historical data base, for equipment fault early-warning and monitoring provide analysis data.In actual monitoring scene, because power equipment Monitoring Data amount is large, the sharply expansion of data scale, and along with the progress of science and technology and the demand for development of all-around function integration analyzing and processing, traditional database can not meet the requirement of storage and integrated analysis process.
Along with the develop rapidly of computer technology, the data of every profession and trade increase rapidly, data quantitative change increasing, type also gets more and more, data structure is also tending towards complicated, traditional database not only each equipment independent mode, and needs larger space for its deployment, there is shortcomings such as not easily disposing, cost is higher, the General Requirements of user can not be met.The appearance of large data all-in-one machine and database thereof and be developed into emerging development trend, but, the not special large data all-in-one machine for power industry design and use and data analysing method thereof at present.
Summary of the invention
The object of the invention is to overcome the deficiencies in the prior art, provide a kind of towards the large market demand scene of power industry, the customizable large data all-in-one machine of soft or hard one, modularization, power business.
The invention provides a kind of large data all-in-one machine based on electric power application, comprise rack, real-time data base and one or more converting equipment measuring point sensor, also comprise the one or more bottoms being arranged at rack bottom and calculate storage server, be arranged at the power module in the middle part of rack and gigabit supervising the network module, be set in turn in the 10000000000 fiber data networking modules on rack top from top to bottom, one or more top calculates storage server and one or more high-performance calculation storage server, be arranged at fan and the fan governor of rear of cabinet, fan governor is connected with power module and fan respectively,
Described gigabit supervising the network module comprises server centered administration module;
Wherein the output terminal of converting equipment measuring point sensor connects the input end of real-time data base input end and 10,000,000,000 fiber data networking modules respectively, real-time data base output terminal is also connected with the input end of 10,000,000,000 fiber data networking modules, described converting equipment measuring point sensor is used for real-time acquisition converting equipment measuring point information, after converting equipment measuring point sensor obtains converting equipment measuring point information, the converting equipment measuring point information of acquisition is sent to real-time data base and/or 10,000,000,000 fiber data networking modules, the converting equipment measuring point information received stores by real-time data base, when receiving transmission order, the converting equipment measuring point information of storage is sent to 10,000,000,000 fiber data networking modules, 10000000000 fiber data networking modules calculate storage server with bottom respectively, top calculates storage server and high-performance calculation storage server connects, for after the converting equipment measuring point information distribution that will receive from real-time data base and/or converting equipment measuring point sensor, bottom is sent to calculate storage server respectively, top calculates storage server and high-performance calculation storage server carries out computing, converting equipment measuring point information data after computing calculates storage server through gigabit supervising the network module transfer to bottom, top calculates in the storer in storage server and/or high-performance calculation storage server, its middle and lower part calculates storage server, top calculates storage server and high-performance calculation storage server comprises 2 road CPU processors respectively, described 2 road CPU processors are that intel is to strong E5-2620v2 type processor, high-performance calculation storage server also comprises 1 graphic process unit GPU, in power module, an independent power supply is powered to the GPU in high-performance calculation storage server, all the other power supplys are powered to other equipment in all-in-one, wherein power module is 5+5 redundant power supply power supply, power >=19200W,
Go back set temperature sensor in rack, for measuring the temperature in rack, fan governor carries out the output power of step control fan according to the temperature in rack, wherein meets:
1 grade: when 0≤P≤0.25, export 40% power;
2 grades: when 0.25 < P≤0.5, export 65% power;
3 grades: when 0.5 < P≤0.8, export 85% power;
4 grades: when 0.8 < P≤1, export 100% power;
Wherein S is the machine in-cabinet temperature that temperature sensor records, and E is the maximum allowable temperature of large data all-in-one machine when normally working, and P is temperature control parameter.
Preferably, described converting equipment measuring point information is one in oil dissolved gas information, shelf depreciation information, electric quantity information, winding deformation information, temperature information and online infrared imaging information or combination.
Preferably, rack is the 42U rack of standard, and wherein rack includes L-type guide rail, rail height ﹤ 2.2mm.
Preferably, power module adopts overall concentration supply power.
Preferably, adopt unitary back heat radiation, hot plug supported by fan.
Preferably, bottom calculates storage server, top calculates storage server and high-performance calculation storage server all adopts frame structure, and each calculates storage server and comprises: inside save as 16G*8, DDR3, ECC, frequency >=1333MHz; Also comprise enterprise-level 800G solid-state hard disk SSD; Also comprise 3.5 cun of enterprise-level hard disks: 2T*10, SATA, 7200rpm, 6G interface bandwidth, and support hot plug.
Preferably, the model of graphic process unit GPU is NVIDIATeslaK40.
Preferably, described rack is multiple.
The present invention compared with prior art, can realize:
1) the large data all-in-one machine for power industry design and use, for the large data analysis of power industry provides the foundation;
2) full distributed large data processing architecture, the IT infrastructure module such as computing unit (high-performance calculation unit), storage unit, communication unit, administrative unit of complexity is carried out integrated, the complete device of out-of-the-box (ReadtoRun) is provided, carry out overall performance tuning, the function avoiding causing because each device fabrication business operative norm is inconsistent, performance loss, lifting means availability.
3) calculate the calculation features such as (lifting counting yield), streaming calculating (tackling real-time computation requirement) according to large Data Analysis Platform internal memory and customization optimization is carried out to hardware system, internal memory-SSD-HDD three layers of storage system effectively alleviate memory pressure, the use of 10000000000 fiber optic networks, improve the traffic rate of each node, to tackle Large-scale parallel computing demand;
4) system extension linear increase, product provides puts from 1/4 configuration, 1/2 configuration, full configuration colony system initial configuration scales such as always extending to the configuration of multimachine cabinet, according to project actual demand flexible configuration product, do not need for later stage system upgrade forward purchasing equipment.When there is performance bottleneck in system, linear expansion system as required.
Accompanying drawing explanation
The large data all-in-one machine structural representation that Fig. 1 applies based on electric power
Fig. 2 rack rear view
Embodiment
The following detailed description of specific embodiment of the invention; what be necessary to herein means out is; below implement just to further illustrate for of the present invention; limiting the scope of the invention can not be interpreted as; some nonessential improvement and adjustment that this art skilled person makes the present invention according to the invention described above content, still belong to protection scope of the present invention.
Present invention also offers a kind of large data all-in-one machine based on electric power application, as shown in Figure 1, comprise rack 9, real-time data base 8 and one or more converting equipment measuring point sensor 7, also comprise the one or more bottoms being arranged at rack 9 bottom and calculate storage server 5, be arranged at the power module 4 in the middle part of rack and gigabit supervising the network module 3, be set in turn in the 10000000000 fiber data networking modules 1 on rack 9 top from top to bottom, one or more top calculates storage server 6 and one or more high-performance calculation storage server 2, as shown in Figure 2, also comprise the fan 10 and fan governor 11 that are arranged at rack 9 back side, fan governor 11 is connected with power module 4 and fan 10 respectively, wherein the output terminal of converting equipment measuring point sensor 7 connects the input end of real-time data base 8 input end and 10,000,000,000 fiber data networking modules 1 respectively, real-time data base 8 output terminal is also connected with the input end of 10,000,000,000 fiber data networking modules 1, described converting equipment measuring point sensor 7 is for real-time acquisition converting equipment measuring point information, such as oil dissolved gas information, shelf depreciation information, electric quantity information, winding deformation information, temperature information, online infrared imaging information etc., after converting equipment measuring point sensor 7 obtains converting equipment measuring point information, the converting equipment measuring point information of acquisition is sent to real-time data base 8 and/or 10,000,000,000 fiber data networking module 1, the converting equipment measuring point information received stores by real-time data base 8, when receiving transmission order, the converting equipment measuring point information of storage is sent to 10,000,000,000 fiber data networking modules 1, 10000000000 fiber data networking modules 1 calculate storage server 5 with bottom respectively, top calculates storage server 6 and high-performance calculation storage server 2 connects, for after the converting equipment measuring point information distribution that will receive from real-time data base 8 and/or converting equipment measuring point sensor 7, bottom is sent to calculate storage server 5 respectively, top calculates storage server 6 and high-performance calculation storage server 2 carries out computing, converting equipment measuring point information data after computing is transferred to bottom through gigabit supervising the network module 3 and calculates storage server 5, top calculates in the storer in storage server 6 and/or high-performance calculation storage server 2.
In a preferred embodiment, rack 1 can be the 42U rack of standard, and rack includes L-type guide rail, rail height ﹤ 2.2mm, adopt overall concentration supply power, power module 4 is 5+5 redundant power supply power supply, power >=19200VA, and unitary back is dispelled the heat, fan window 9 groups, often organize fan window 3 140*38mm fans 10, hot plug supported by fan 10, and gigabit supervising the network module 3 comprises server centered administration module.
Bottom calculates storage server 5, top calculates storage server 6 similar, all adopt frame structure, bottom calculates storage server 5, top calculates storage server 6 by the L-type guide rails assembling in rack in rack, and each calculates storage server and comprises 2 road intel to strong E5-2620v2 type processor, inside save as 16G*8, DDR3, ECC, frequency >=1333MHz, also comprise enterprise-level 800G solid-state hard disk SSD, 3.5 cun of enterprise-level hard disks are 2T*10, SATA, 7200rpm, 6G interface bandwidth, and support hot plug, network interface card is 10,000,000,000 mouthful * 2, and comprise independent independent management network port (RJ45) SAS card: LSISAS2308 master control, support raid1.0.
High-performance calculation storage server 2 and calculating storage server 5 similar, it not only has calculating storage server 5 structure, also comprises 1 e *8PCIE expansion slot, in addition compared with traditional server, high-performance calculation storage server 2 additionally also comprises 1 GPU, (preferred model is NVIDIATeslaK40).In addition, calculating storage server under normal circumstances all adopts same power supply to power, but owing to being provided with the high-performance calculation storage server comprising GPU, make such power supply mode often can cause the instability of voltage, all-in-one easily crashes, and easily burn GPU, through research, the power supply mode of power module 4 is improved by the present invention, by an independent power supply in power module 4, independent power supply is carried out to the GPU in high-performance calculation storage server 2, find to adopt such power supply mode after overtesting after, the performance of large data all-in-one machine obtains significantly to be improved, the time of continuous operation obviously increases, and the frequency that crashes significantly declines.
10000000000 fiber data networking modules 1 can be chosen as 10,000,000,000 optical fiber switchs, and (preferred model is StarNet RG-S6220,48 mouthfuls of optical fiber switchs, configuration optic module.
For the configuration of rack, can be configured according to the demand of reality, such as 1/4 configuration, 1/2 configuration, full configuration put colony system initial configuration scales such as always extending to the configuration of multimachine cabinet.
Although for illustrative purposes; describe illustrative embodiments of the present invention; but it should be appreciated by those skilled in the art that; when not departing from scope of invention disclosed in claims and spirit; the change of various amendment, interpolation and replacement etc. can be carried out in form and details; and all these change the protection domain that all should belong to claims of the present invention; and application claims protection each department of product and method in each step, can combine with the form of combination in any.Therefore, be not intended to limit the scope of the invention to the description of embodiment disclosed in the present invention, but for describing the present invention.Correspondingly, scope of the present invention not by the restriction of above embodiment, but is limited by claim or its equivalent.

Claims (8)

1. the large data all-in-one machine based on electric power application, it is characterized in that: comprise rack, real-time data base and one or more converting equipment measuring point sensor, also comprise the one or more bottoms being arranged at rack bottom and calculate storage server, be arranged at the power module in the middle part of rack and gigabit supervising the network module, be set in turn in the 10000000000 fiber data networking modules on rack top from top to bottom, one or more top calculates storage server and one or more high-performance calculation storage server, be arranged at fan and the fan governor of rear of cabinet, fan governor is connected with power module and fan respectively,
Described gigabit supervising the network module comprises server centered administration module;
Wherein the output terminal of converting equipment measuring point sensor connects the input end of real-time data base input end and 10,000,000,000 fiber data networking modules respectively, real-time data base output terminal is also connected with the input end of 10,000,000,000 fiber data networking modules, described converting equipment measuring point sensor is used for real-time acquisition converting equipment measuring point information, after converting equipment measuring point sensor obtains converting equipment measuring point information, the converting equipment measuring point information of acquisition is sent to real-time data base and/or 10,000,000,000 fiber data networking modules, the converting equipment measuring point information received stores by real-time data base, when receiving transmission order, the converting equipment measuring point information of storage is sent to 10,000,000,000 fiber data networking modules, 10000000000 fiber data networking modules calculate storage server with bottom respectively, top calculates storage server and high-performance calculation storage server connects, for after the converting equipment measuring point information distribution that will receive from real-time data base and/or converting equipment measuring point sensor, bottom is sent to calculate storage server respectively, top calculates storage server and high-performance calculation storage server carries out computing, converting equipment measuring point information data after computing calculates storage server through gigabit supervising the network module transfer to bottom, top calculates in the storer in storage server and/or high-performance calculation storage server, its middle and lower part calculates storage server, top calculates storage server and high-performance calculation storage server comprises 2 road CPU processors respectively, described 2 road CPU processors are that intel is to strong E5-2620v2 type processor, high-performance calculation storage server also comprises 1 graphic process unit GPU, in power module, an independent power supply is powered to the GPU in high-performance calculation storage server, all the other power supplys are powered to other equipment in all-in-one, wherein power module is 5+5 redundant power supply power supply, power >=19200W,
Go back set temperature sensor in rack, for measuring the temperature in rack, fan governor carries out the output power of step control fan according to the temperature in rack, wherein meets:
P = 0.9 S E
1 grade: when 0≤P≤0.25, export 40% power;
2 grades: when 0.25 < P≤0.5, export 65% power;
3 grades: when 0.5 < P≤0.8, export 85% power;
4 grades: when 0.8 < P≤1, export 100% power;
Wherein S is the machine in-cabinet temperature that temperature sensor records, and E is the maximum allowable temperature of large data all-in-one machine when normally working, and P is temperature control parameter.
2. as claimed in claim 1 based on the large data all-in-one machine of electric power application, it is characterized in that: described converting equipment measuring point information is one in oil dissolved gas information, shelf depreciation information, electric quantity information, winding deformation information, temperature information and online infrared imaging information or combination.
3., as claimed in claim 1 based on the large data all-in-one machine of electric power application, it is characterized in that: rack is the 42U rack of standard, and wherein rack includes L-type guide rail, rail height ﹤ 2.2mm.
4., as claimed in claim 1 based on the large data all-in-one machine of electric power application, it is characterized in that: power module adopts overall concentration supply power.
5. as claimed in claim 1 based on the large data all-in-one machine of electric power application, it is characterized in that: adopt unitary back heat radiation, hot plug supported by fan.
6. as claimed in claim 1 based on the large data all-in-one machine of electric power application, it is characterized in that: bottom calculates storage server, top calculates storage server and high-performance calculation storage server all adopts frame structure, each calculates storage server and comprises: inside save as 16G*8, DDR3, ECC, frequency >=1333MHz; Also comprise enterprise-level 800G solid-state hard disk SSD; Also comprise 3.5 cun of enterprise-level hard disks: 2T*10, SATA, 7200rpm, 6G interface bandwidth, and support hot plug.
7. the large data all-in-one machine based on electric power application as described in claim 1 or 6, is characterized in that: the model of graphic process unit GPU is NVIDIATeslaK40.
8., as claimed in claim 7 based on the large data all-in-one machine of electric power application, it is characterized in that: described rack is multiple.
CN201510429747.3A 2015-07-21 2015-07-21 A kind of large data all-in-one machine based on electric power application Active CN104951024B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201510429747.3A CN104951024B (en) 2015-07-21 2015-07-21 A kind of large data all-in-one machine based on electric power application

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201510429747.3A CN104951024B (en) 2015-07-21 2015-07-21 A kind of large data all-in-one machine based on electric power application

Publications (2)

Publication Number Publication Date
CN104951024A CN104951024A (en) 2015-09-30
CN104951024B true CN104951024B (en) 2016-03-09

Family

ID=54165735

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201510429747.3A Active CN104951024B (en) 2015-07-21 2015-07-21 A kind of large data all-in-one machine based on electric power application

Country Status (1)

Country Link
CN (1) CN104951024B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105937753A (en) * 2016-06-02 2016-09-14 国家电网公司 Machine cabinet provided with intelligent energy-saving environmental-protection lighting system in machine room
CN106774702A (en) * 2016-11-29 2017-05-31 中国石油集团川庆钻探工程有限公司地球物理勘探公司 A kind of in-situ processing device for physical prospecting big data
CN107390849A (en) * 2017-06-29 2017-11-24 郑州云海信息技术有限公司 A kind of system for realizing Rack GPU power supplies
CN107704204A (en) * 2017-09-30 2018-02-16 长沙准光里电子科技有限公司 A kind of big data all-in-one

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN201757871U (en) * 2010-07-26 2011-03-09 深圳市腾讯计算机***有限公司 Server cabinet
CN102116824A (en) * 2010-11-30 2011-07-06 国网电力科学研究院 Distributed local discharge on-line monitoring method and device for high voltage cable system
CN103186180A (en) * 2011-12-30 2013-07-03 鸿富锦精密工业(深圳)有限公司 Cabinet system and server thereof
CN103375420A (en) * 2012-04-28 2013-10-30 鸿富锦精密工业(深圳)有限公司 Equipment cabinet system and fan control system and control method thereof
CN204065991U (en) * 2014-07-17 2014-12-31 甘肃洁星通信科技有限公司 A kind of large data server

Also Published As

Publication number Publication date
CN104951024A (en) 2015-09-30

Similar Documents

Publication Publication Date Title
CN104991958A (en) Analysis system and method for power equipment monitoring data
CN104951024B (en) A kind of large data all-in-one machine based on electric power application
CN105045922A (en) Electrical equipment monitoring method and device
CN112860695B (en) Monitoring data query method, device, equipment, storage medium and program product
CN103235817B (en) A kind of extensive infection control data storage processing method
CN102571904A (en) Construction method of NAS cluster system based on modularization design
CN208013975U (en) The hardware device of on-line intelligence ability platform
CN204968320U (en) All -in -one heat dissipation rack and rack group based on power equipment monitoring
CN204965251U (en) All -in -one device based on power equipment monitoring
CN202135167U (en) Industrial design platform apparatus based on cloud calculating configuration
CN205015812U (en) Big data all -in -one and rack based on electric power is used
CN204557308U (en) A kind of high density Novel cutter flap-type server based on fusion architecture
CN108334550A (en) A kind of data processing system based on Human Resource Management System
CN106919533B (en) 4U high-density storage type server
CN105760284A (en) Website performance monitoring method and device
CN104981136B (en) A kind of large data all-in-one machine radiating cabinet based on electric power application
CN105184670A (en) Power application-based large data management analysis system and method thereof
CN204831369U (en) Power equipment real -time supervision device
CN204790778U (en) Big data all -in -one based on power equipment monitoring
CN204465589U (en) A kind of cloud computing equipment with high operational performance
CN105181012A (en) Power equipment monitoring device having intelligent heat-dissipation function and monitoring method thereof
CN102799708B (en) Graphic processing unit (GPU) high-performance calculation platform device applied to electromagnetic simulation
CN104011693B (en) The apparatus and method that processor frequencies layer based on I/O bandwidth is set
CN108898655A (en) A kind of wisdom computer room Modeling Platform based on technology of Internet of things
CN108763253A (en) A kind of the big data visualization system and method for investment combination

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
CP01 Change in the name or title of a patent holder
CP01 Change in the name or title of a patent holder

Address after: 250101 5th floor, block B, Yinhe building, 2008 Xinluo street, high tech Zone, Jinan City, Shandong Province

Patentee after: Shandong luruan Digital Technology Co.,Ltd.

Address before: 250101 5th floor, block B, Yinhe building, 2008 Xinluo street, high tech Zone, Jinan City, Shandong Province

Patentee before: SHANDONG LUNENG SOFTWARE TECHNOLOGY Co.,Ltd.