CN104486127A - Redundancy trusted server management method based on trusted management unit - Google Patents

Redundancy trusted server management method based on trusted management unit Download PDF

Info

Publication number
CN104486127A
CN104486127A CN201410801295.2A CN201410801295A CN104486127A CN 104486127 A CN104486127 A CN 104486127A CN 201410801295 A CN201410801295 A CN 201410801295A CN 104486127 A CN104486127 A CN 104486127A
Authority
CN
China
Prior art keywords
unit
tcm
redundancy
computing unit
administrative unit
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
Application number
CN201410801295.2A
Other languages
Chinese (zh)
Inventor
沈忱
于治楼
姜凯
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.)
Inspur Group Co Ltd
Original Assignee
Inspur Group 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 Inspur Group Co Ltd filed Critical Inspur Group Co Ltd
Priority to CN201410801295.2A priority Critical patent/CN104486127A/en
Publication of CN104486127A publication Critical patent/CN104486127A/en
Pending legal-status Critical Current

Links

Abstract

The invention discloses a redundancy trusted server management method based on a trusted management unit and belongs to the field of server management. The method comprises steps as follows: TCMs (trusted cryptography modules) are integrated on the management unit and calculation units of a redundancy trusted server; a system is started, the TCM of the management unit supplies power to the calculation units through a control power supply after successfully verifying the integrity of a BMC program, the calculation units are started, and otherwise, the reliable BMC program is restored through a backup; after the calculation units are started, the TCMs of the calculation units verify the integrity of BIOSs of the calculation units; the TCM of the management unit monitors the operation of the TCMs of the calculation units through an I2C bus to ensure the operation of the redundancy trusted server. According to the method, the TCMs and BMC management chips of the management unit as well as the TCMs of other calculation units are added, so that the safety and the reliability of the whole redundancy trusted server are improved fundamentally.

Description

A kind of redundancy trusted servers management method based on credible administrative unit
Technical field
The present invention relates to a kind of redundancy trusted servers management method, belong to server admin field, specifically a kind of redundancy trusted servers management method based on credible administrative unit.
Background technology
For modern company, what the network operation was the most valuable is data.Storage is server, and data are souls of network.The core system that the data of all processes are completed a business transaction by its computer center provided, any artificial mistake, the loss of data that the damage, computer virus, natural disaster etc. of hard disk cause, and all will cause immeasurable losses.There is redundant server just can reduce above-mentioned risk.The fast development of current domestic information technology, the demand of autonomous domestic redundant server is also increasing.Existing autonomous domestic redundant server is that several independently computing unit individuality adds crosspoint, administrative unit, memory cell composition mostly.In order to increase the security reliability of server, often through setting up TCM chip at computing unit to improve the reliability of whole computing unit, but have ignored administrative unit in the use of server and be only the key controlling authority at the highest level, user remote access administrative unit can control certain computing unit switching on and shutting down, therefore the security reliability of administrative unit should come into one's own more, is protected.For at present not to the effective safeguard measure of the security reliability of administrative unit, and its security reliability is most important, the invention provides a kind of redundancy trusted servers management method based on credible administrative unit, by setting up TCM chip in administrative unit, and the TCM of the TCM of administrative unit and each computing unit is passed through I2C bus bar, while guaranteeing plate level Administrative Security reliability, increase administrative unit TCM to the management of the TCM of each computing unit, further improve the security reliability of whole credible redundant server, be particularly useful for domestic redundant server, outstanding contribution has been made in exploitation for domestic credible redundant server, there is very wide application market.
Summary of the invention
The present invention is directed to deficiency and the problem of prior art existence, for better solving the problem of the security reliability of credible redundant server, a kind of redundancy trusted servers management method based on credible administrative unit being provided, being particularly useful in domestic redundant server.
The concrete scheme proposed is:
Based on a redundancy trusted servers management method for credible administrative unit, concrete steps are:
TCM chip is carried in the administrative unit of redundancy trusted servers and computing unit plate;
Start up system, the TCM of administrative unit verifies that the integrality of BMC program is just powered to computing unit by controlling power supply after successful, start-up simulation unit; Unsuccessful then will by the reliable BMC program of backup and reduction;
After computing unit starts, the integrality of the TCM verification computation unit B IOS of computing unit, is verified rear normal boot system cloud gray model;
The TCM that the TCM of administrative unit continues through I2C monitoring bus computing unit runs, and the TCM of computing unit breaks down, and administrative unit, by switching or the computing unit of closing fault, ensures that redundancy trusted servers is run.
Described start up system is started by the mode of starting key or is started by network remote access.
Described start up system is started by the mode of starting key, then user needs to use IC-card identity verification; Start up system is started by network remote access, then user needs input password identity verification.
The TCM of administrative unit is by the TCM of I2C monitoring bus computing unit, and the TCM of each computing unit is by the TCM interconnection on independently GPIO and administrative unit.
A kind of redundancy trusted servers, described a kind of redundancy trusted servers management method based on credible administrative unit is used to manage, comprise computing unit, network exchange unit, memory cell, administrative unit, computing unit and administrative unit all plate carry TCM chip, and each unit is connected by I2C bus.
Each described computing unit plate carries TCM chip, by the TCM interconnection on independently GPIO and administrative unit.
Usefulness of the present invention is: the present invention is by carrying TCM chip in the administrative unit of redundancy trusted servers and computing unit plate, when system is started, the TCM of administrative unit verifies that the integrality of BMC program is just powered to computing unit by controlling power supply after successful, after computing unit starts, the integrality of the TCM verification computation unit B IOS of computing unit, be verified rear ability normal boot system cloud gray model, the security reliability of whole redundancy trusted servers is ensure that like this from basis, and administrative unit can by the startup of tele-control system, TCM due to administrative unit passes through the TCM of I2C monitoring bus computing unit, the TCM of computing unit breaks down, administrative unit will switch or the computing unit of closing fault, guarantee the normal operation of redundancy trusted servers further.
Accompanying drawing explanation
Fig. 1 trusted servers overall architecture schematic diagram.
Embodiment
The present invention will be further described by reference to the accompanying drawings.
For the redundancy trusted servers of domestic Godson 3A platform.
As Fig. 1, this is four unit redundancy trusted servers, comprises four Godson 3A computing units, the server unit 1 namely in figure, server unit 2, server unit 3, server unit 4;
Two network exchange unit, the crosspoint 1 namely in figure, crosspoint 2;
An administrative unit, the management controlled area namely in figure;
Memory cell, of comprising in i.e. figure stores control board, two memory planes, and two memory planes are cells of memory arrays.
Also have two power subsystem compositions.
In figure server unit every 2 be divided into two groups be arranged on management both sides, controlled area, 2 crosspoints are separately positioned on management both sides, controlled area between management controlled area and server unit, store control board, two memory planes and power subsystem and are successively set on management side, controlled area near server unit.
By finding out in figure that each server unit is provided with TCM chip, and pass through independently GPIO and the TCM chip interconnects on management controlled area, the TCM chip of management controlled area is connected with Long-distance Control network interface.
Above redundancy trusted servers, use a kind of redundancy trusted servers management method based on credible administrative unit, concrete steps are:
The administrative unit of redundancy trusted servers and computing unit plate carry TCM chip;
Administrative unit, by treating dynamo-electric operation, can be run as long as plug voltage.By the mode start up system of starting key, user needs to use IC-card identity verification, and the TCM starting administrative unit verifies that the integrality of BMC program is just powered to computing unit by controlling power supply after successful, start-up simulation unit; Unsuccessful then will by the reliable BMC program of backup and reduction;
After computing unit starts, the integrality of the TCM verification computation unit B IOS of computing unit, is verified rear normal boot system cloud gray model; The TCM that the TCM of administrative unit continues through I2C monitoring bus computing unit runs, and the TCM of computing unit breaks down, and administrative unit, by switching or the computing unit of closing fault, ensures that redundancy trusted servers is run.
Also can be accessed by network remote, user inputs password identity verification, is verified rear computing unit and just starts, if identity verification is unsuccessful, and will by the reliable BMC program of backup and reduction; When computing unit normally runs, the integrality of the TCM verification computation unit B IOS of computing unit, be verified rear normal boot system cloud gray model, administrative unit TCM is by I2C bus real-time monitoring calculation unit TCM state, if the TCM of computing unit breaks down, administrative unit will switch or the computing unit of closing fault.

Claims (6)

1., based on a redundancy trusted servers management method for credible administrative unit, it is characterized in that concrete steps are:
TCM chip is carried in the administrative unit of redundancy trusted servers and computing unit plate;
Start up system, the TCM of administrative unit verifies that the integrality of BMC program is just powered to computing unit by controlling power supply after successful, start-up simulation unit; Unsuccessful then will by the reliable BMC program of backup and reduction;
After computing unit starts, the integrality of the TCM verification computation unit B IOS of computing unit, is verified rear normal boot system cloud gray model;
The TCM that the TCM of administrative unit continues through I2C monitoring bus computing unit runs, and the TCM of computing unit breaks down, and administrative unit, by switching or the computing unit of closing fault, ensures that redundancy trusted servers is run.
2. a kind of redundancy trusted servers management method based on credible administrative unit according to claim 1, is characterized in that described start up system is started by the mode of starting key or started by network remote access.
3. a kind of redundancy trusted servers management method based on credible administrative unit according to claim 2, is characterized in that described start up system is started by the mode of starting key, then user needs to use IC-card identity verification; Start up system is started by network remote access, then user needs input password identity verification.
4. a kind of redundancy trusted servers management method based on credible administrative unit according to any one of claim 1-3, it is characterized in that the TCM of the TCM of administrative unit by I2C monitoring bus computing unit, the TCM of each computing unit is by the TCM interconnection on independently GPIO and administrative unit.
5. a redundancy trusted servers, a kind of redundancy trusted servers management method based on credible administrative unit described in any one of claim 1-4 is used to manage, it is characterized in that comprising computing unit, network exchange unit, memory cell, administrative unit, computing unit and administrative unit all plate carry TCM chip, and each unit is connected by I2C bus.
6. a kind of redundancy trusted servers according to claim 5, is characterized in that each described computing unit plate carries TCM chip, by the TCM interconnection on independently GPIO and administrative unit.
CN201410801295.2A 2014-12-22 2014-12-22 Redundancy trusted server management method based on trusted management unit Pending CN104486127A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201410801295.2A CN104486127A (en) 2014-12-22 2014-12-22 Redundancy trusted server management method based on trusted management unit

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201410801295.2A CN104486127A (en) 2014-12-22 2014-12-22 Redundancy trusted server management method based on trusted management unit

Publications (1)

Publication Number Publication Date
CN104486127A true CN104486127A (en) 2015-04-01

Family

ID=52760635

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201410801295.2A Pending CN104486127A (en) 2014-12-22 2014-12-22 Redundancy trusted server management method based on trusted management unit

Country Status (1)

Country Link
CN (1) CN104486127A (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106599677A (en) * 2016-12-23 2017-04-26 郑州云海信息技术有限公司 Password control system and control method used for baseboard management controller
CN107360167A (en) * 2017-07-17 2017-11-17 浪潮(北京)电子信息产业有限公司 A kind of authentication method and device
CN110197070B (en) * 2018-02-27 2023-07-21 其阳科技股份有限公司 Computer system and method with power-on trusted verification and fault-tolerant transfer

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1723425A (en) * 2002-10-09 2006-01-18 英特尔公司 Encapsulation of a TCPA trusted platform module functionality within a server management coprocessor subsystem
CN101877040A (en) * 2009-12-07 2010-11-03 中国航天科工集团第二研究院七○六所 High-reliability computing platform
CN102650944A (en) * 2011-02-28 2012-08-29 国民技术股份有限公司 Operation system security bootstrap device and bootstrap device
CN104200148A (en) * 2014-08-25 2014-12-10 浪潮集团有限公司 Smart card redundancy switching method based on independently domestic redundancy server

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1723425A (en) * 2002-10-09 2006-01-18 英特尔公司 Encapsulation of a TCPA trusted platform module functionality within a server management coprocessor subsystem
CN101877040A (en) * 2009-12-07 2010-11-03 中国航天科工集团第二研究院七○六所 High-reliability computing platform
CN102650944A (en) * 2011-02-28 2012-08-29 国民技术股份有限公司 Operation system security bootstrap device and bootstrap device
CN104200148A (en) * 2014-08-25 2014-12-10 浪潮集团有限公司 Smart card redundancy switching method based on independently domestic redundancy server

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106599677A (en) * 2016-12-23 2017-04-26 郑州云海信息技术有限公司 Password control system and control method used for baseboard management controller
CN107360167A (en) * 2017-07-17 2017-11-17 浪潮(北京)电子信息产业有限公司 A kind of authentication method and device
CN110197070B (en) * 2018-02-27 2023-07-21 其阳科技股份有限公司 Computer system and method with power-on trusted verification and fault-tolerant transfer

Similar Documents

Publication Publication Date Title
CN102255888B (en) Method and apparatus data storage device being carried out security sweep from remote server
CN208210006U (en) A kind of high safety trusted servers based on domestic TPM
CN107403098A (en) The active safety means of defence and credible industrial control computer of credible industrial control computer startup stage
CN107506663A (en) Server security based on credible BMC starts method
TWI740158B (en) A server system, a centralized flash memory module, and a method of updating flash firmware image
CN103530548B (en) Startup method that built-in terminal based on mobile trustable computation module is credible
CN106557143B (en) Apparatus and method for data storage device
CN103473152A (en) Main management module and standby management module backuping and updating method for blade server
CN205139890U (en) Two memory system that live of two accuses based on X86 framework
CN108228374A (en) A kind of fault handling method of equipment, apparatus and system
CN106326061A (en) High-speed cache data processing method and equipment
CN105700924B (en) A kind of method and device for multiple module burning programs
CN104486127A (en) Redundancy trusted server management method based on trusted management unit
US20170132079A1 (en) Rebuilding and verifying an encoded data slice utilizing slice verification information
CN109358490A (en) A kind of redundance unit and its test method, system and storage medium
CN107678997A (en) Hot-plug method, system, device and the readable storage medium storing program for executing of PCIE plug-in cards
CN105051692A (en) Automated failure handling through isolation
CN111125707A (en) BMC (baseboard management controller) safe starting method, system and equipment based on trusted password module
US11010086B2 (en) Data synchronization method and out-of-band management device
CN114090174A (en) City rail edge cloud integration equipment based on super integration
CN104200148A (en) Smart card redundancy switching method based on independently domestic redundancy server
TWI791913B (en) System and method to recover fpga firmware over a sideband interface
CN102520880B (en) System raid management module design method
CN105843336B (en) Cabinet and its firmware updating method with more cabinet management modules
CN106528324A (en) Fault recovery method and apparatus

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication

Application publication date: 20150401

RJ01 Rejection of invention patent application after publication