CN100383739C - Mirror starting optimization of built-in operation system - Google Patents

Mirror starting optimization of built-in operation system Download PDF

Info

Publication number
CN100383739C
CN100383739C CNB2006100498559A CN200610049855A CN100383739C CN 100383739 C CN100383739 C CN 100383739C CN B2006100498559 A CNB2006100498559 A CN B2006100498559A CN 200610049855 A CN200610049855 A CN 200610049855A CN 100383739 C CN100383739 C CN 100383739C
Authority
CN
China
Prior art keywords
operating
mirror image
memory
image
kernel
Prior art date
Application number
CNB2006100498559A
Other languages
Chinese (zh)
Other versions
CN1818869A (en
Inventor
陈天洲
黄江伟
梁晓
钱杰
吴心亮
郑臻伟
Original Assignee
浙江大学
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 浙江大学 filed Critical 浙江大学
Priority to CNB2006100498559A priority Critical patent/CN100383739C/en
Publication of CN1818869A publication Critical patent/CN1818869A/en
Application granted granted Critical
Publication of CN100383739C publication Critical patent/CN100383739C/en

Links

Abstract

The present invention discloses a starting optimization method for the mirror image starting of an embedded operating system. The present invention has the method that when the operation system is started, a guiding program directly reads the mirror image of the minimum task of the system which is stored in advance into an internal memory from a memory, various structural properties and the memorized information of a system register are set according to the read-in information of the system, the system is quickly restored to a working condition, and the method of quick start is achieved. The present invention provides the starting optimization method for the mirror image starting of the operating system, and improves the starting process of the original operating system. Therefore, the starting process of the system is fast, and the working condition of the system can be achieved quickly.

Description

The startup optimization method of embedded OS image starting

Technical field

The present invention relates to based on the starting of embedded operation system field, particularly relate to a kind of startup optimization method of embedded OS image starting.

Background technology

Embedded systems such as mobile phone are owing to the patience of considering the people, so can not spend the long time when starting.For the convenience on using, to shorten the guiding and the start-up time of mobile phone operating system as far as possible, though kernel is through reducing significantly, and various functional modules are all optimized as far as possible, but because the restriction of hardware condition, if special design is not done in guiding and startup, still can be caused start-up time long.

Two kinds of technology of main now employing shorten this time.One, solidify the various settings that can not expand hardware, but the interface of the standardization expansion equipment of trying one's best is saved the Equipment Inspection time.Two, time-delay loading technology, both LazyLoading, this technology does not load current nonessential module, and these modules can load in the stand-by time after startup, perhaps reloads after specifically using, broken up start-up time, thereby shorten the vital reaction time concerning the user.

Concerning first kind of technology, present and topmost in shorter a period of time in the future still is to solidify the setting that can not expand hardware, because but the expansion equipment of mobile phone also seldom at first now, secondly the hardware condition restriction is more, should be more satisfactory measure so some fixing settings are solidificated in operating system inside.And consider in the future development trend, mobile phone can connect increasing peripheral hardware, and these peripheral hardwares all must be plug and play, so can not solidify their setting.If but formulate a cover standard interface for all mobile phone peripheral hardwares, the common interface module that at most only needs when then starting to load a standard gets final product, and has saved start-up time greatly, and has saved storage space.

For second kind of technology, because operating system will be in a lot of local technology that adopt " modularization ", so very convenient Lazy Loading.And in order further to shorten the reaction time, to " dynamically arrange " essential module, according to statistics to user's conditions such as use habit, the module that some users can not use after start immediately is set to nonessential module, and the function setting that the user just uses after the start of being everlasting is necessary module, the reaction time has just obtained very big improvement concerning the user like this.

Summary of the invention

The object of the present invention is to provide a kind of startup optimization method of embedded OS image starting.

The technical scheme that the present invention solves its technical matters employing is as follows:

1) the operating system mirror image is prepared:

The boot of operating system is in the process that starts, at first will own copy memory headroom to from external memory, arrive the external memory location lookup operating system nucleus mirror image of appointment then, if the operating system nucleus mirror image exists, so the operating system nucleus mirror image is written into kernel from the external memory position of appointment, and decompress, in start-up course, operating system will judge whether to exist the operating system mirror image of quick startup, if there is no then normally start, if exist then startup fast, if boot is found only to have original operating system mirror image, operating system will start by normal procedure;

When os starting behind the most succinct groundwork state, the operating system sleeping process that will start the operating system, the groundwork of this process is to do various preliminary works for preserving the operating system mirror image;

2) preserve the operating system mirror image:

The preservation work of this part main complete operation system for field, at first will set up a virtual terminal, be used for simulating the communication of kernel and user's attitude process, begin to finish to remove the kernel state process then, zombie, all process outside the current process, after this call function discharges memory headroom as much as possible, to reduce the size of operating system mirror image, then some drivings is handled, to all peripheral hardwares of in power-supply management system, registering, send suspending command, insert the internal memory barrier then, begin to preserve the processor context, at last operating system is mirrored on the outside appointed equipment, at last with the operating system dormancy;

3) loading of operating system mirror image:

The boot of operating system is in the process that starts, at first will own copy memory headroom to from external memory, arrive the external memory location lookup operating system nucleus mirror image of appointment then, if the operating system nucleus mirror image exists, so the operating system nucleus mirror image is written into kernel from the external memory position of appointment, and decompresses, in start-up course, if boot is found to have the operating system mirror image that starts fast, operating system is written into internal memory with this mirror image, and decompresses;

Operating system is founded a virtual terminal earlier, be used for showing the system information of loading in the mirrored procedure, the correctness of checked operation system image at first, if it is correct, then the operating system mirror image is written into, the page of all preservations is obtained in circulation, and recovery memory content, value by the internal memory barrier restore processor register that inserts, discharge the memory headroom of applying in the rejuvenation at last, and recover notice by the equipment transmitting system in power management module, registered, finish the loading of whole operation system image.

The present invention compares with background technology, and the useful effect that has is:

1, there is not time-delay: top this method, can only cause a kind of false appearance for the user, system looks and can use, and gets up, and in fact must load to get module in usefulness.And the startup optimization method of embedded OS image starting, be real,, do not need time-delay to load directly by the applied environment of system is set for application provides a complete system, and in usefulness loading necessary module, saved time and energy;

2, agility: what the startup optimization method of embedded OS image starting adopted is the startup method that system environments directly is set, and this method is more quick;

3, versatility: the startup mirror image can be set automatically, need not carry out different settings at different application platforms, changing platform does not need to change mirror image, as long as restarting systems allows system according to system situation a mirror image is set automatically then, just can.

Description of drawings

Accompanying drawing is the process flow diagram of total system.

Embodiment

When implementing based on the starting of embedded operation system optimization method, the image starting technology is widely used.

The startup optimization method specific implementation flow process of embedded OS image starting is as follows.

One, system image is prepared

The system image set-up procedure mainly is divided into following a few part:

1, booting operating system

The boot of operating system is in the process that starts, at first will own copy memory headroom to from external memory, arrive the external memory location lookup system image of appointment then, if system image exists, so the operating system nucleus mirror image is written into kernel from the external memory of system, because the system kernel image ratio is bigger, is saved in external memory after all being compressed generally speaking, so will decompress;

With the ARM-Linux operating system on the embedded experiment porch of Sistang is example, his booting operating system process as seen below.

Before the kernel operation, need system bootstrap routine to finish and load kernel and some complementary work, jump to the start address and the execution of kernel code then.

At first, start back Sitsang plank and address 0x0,c00 0000 is mapped to 0 (can by the wire jumper setting), in fact start from 0x0,c00 0000, the bootloader that enters, but because flash speed is slow, so the bootloader front has a bit of program that bootloader is copied to 0x0AFE0100 among the SDRAM, again from 0x 0,800 0000 operation bootloader, being this section small routine is flashloader, and flashloader is initialization SDRAM at first.Flashloader to 0x0AFE0100, has so knock-oned bootloader load in the past, and 0x0AFE0100 is exactly the real bootloader that burns in flash 0x0C000100 in fact.

BootLoader calls in internal memory with operating system code, then control is given this section of the Setup.S program among the arch/arm/boot.This section of Setup.S program under real pattern, system carried out basic detection and be provided with after change protected mode over to and give head_armv.S control.A step is the decompression procedure of system kernel, this part code is at address 0x1000 (file/Boot/head.S), this section program initialization register, carry out decompress_kernel () then, this function comes from zBoot/inflate.c, zBoot/unzip.c and three files of zBoot/misc.c.Data are placed on address 0x100000 behind the decompress(ion).Head.S sets up the framework of memory management and interrupt management.

2, os starting

1), judges whether to exist quick startup mirror image

After kernel mirror image decompressed, system will judge whether to exist quick startup mirror image, normally started as crossing not exist then, if exist then startup fast.

2), the normal startup

In start-up course, if boot finds only to exist original operating system mirror image, system will start by normal procedure, at first the various parts of initialization are in order incited somebody to action by system, at first initialization architecture, initialization interrupt vector table, scheduler, internal memory or the like, the various services of start-up system at last then.

With ARM-Linux is example, and this start-up course mainly begins up to creating the init process from Start_kernel.After the Init process transfer lock_kernel, before calling prepare_namespace loading rootfs, call the do_basic_setup function, carry out the setting on some bases, as pci, sbus, ecard etc., do_basic_setup calls start_context_thread () and do_initcalls () then, load the driver of some equipment, afterwards, enter the function of this new interpolation of software_resume (), prepare to check exchange partition, carry out the recovery of system.

After some necessary conditions of carrying out the software_resume function detect, call the read_suspend_image function, the mirror image of preserving before reading, if not existing, mirror image do not read failure, the software_resume function can call kernel_thread (swsusp_mainloop, NULL, CLONE_FS|CLONE_FILES|CLONE_SIGHAND|SIGCHLD) create the process of a kernel state, swsusp_mainloop () function call daemonize () discharges quoting of user's space is made the finger daemon that oneself becomes a backstage, and rename oneself is kswsuspd, just be absorbed in unlimited circulation afterwards, and detected swsusp_state[0 every one second] value whether occur changing (if this value becomes 1 from 0, then can jump out circulation, call the function of dormancy).So far, the software_resume function returns, and continues do_basic_setup and init (), normally starts until system.

3), image starting

If boot finds to exist the mirror image of operating system, boot is written into internal memory with this mirror image, and decompresses; System foundes a virtual terminal earlier, is used for showing the system information of loading in the mirrored procedure, at first checks the correctness of mirror image, in order the memory pages in the mirror image is returned to internal memory then, and the context with processor recovers at last.Specifically illustrating of this part work will be put into the back.

3, the foundation of mirror image working environment

After the most succinct groundwork state was arrived in system start-up, operating system was crossed process with the start-up system dormancy, and the groundwork of this process is to do various preliminary works for the saved system mirror image;

With (SuSE) Linux OS can image starting be example, this part work mainly comprises:

1) add one section to the data segment the inside of kernel: _ nosave, so that some variablees are put into wherein.So, when using these variablees, just can not relate to storehouse.

2) the equipment static char resume_file[256 of definition preservation/recovery mirror image]="/dev/modify ", the major device number of this equipment is 241, secondary device number is 2, root_dev_names[that simultaneously need be in init/do_mounts.c] in add one: { " modify ", 0xf102}, after this by the name_to_kdev_t function, designated parameter "/dev/modify " just can obtain the device number 0xf102 of this equipment, just can obtain the major device number 241 of this equipment and slave unit numbers 2 respectively by grand MAJOR () and MINOR () again.

3) definition status amount: int swsusp_state[3] be used for managing some parameter informations that carry out in dormancy and the mirrored procedure.

Two, saved system mirror image

This part mainly finishes the preservation work of system for field, at first just sets up a virtual terminal, is used for analog communication; Begin finish to remove the kernel state process then, zombie, all process outside the current process, after this call function discharges memory headroom as much as possible, to reduce the size of memory mirror, then some drivings are handled, all peripheral hardwares of in power-supply management system, registering, send suspending command, insertion system barrier then, blocking system is carried out, and begins to preserve the processor context, at last on the equipment that memory mirror is formulated to the outside, at last with the operating system dormancy.

With ARM-Linux is example, and whole process is as follows:

1), system will establish mirror image and must call one and be called the software_suspend function.Software_suspend () has called the do_software_suspend function, has really begun the dormancy of system and the foundation of mirror image.

2), software_suspend () has called the do_software_suspend function, at first set up a virtual terminal, begin to call the freeze_processes function then and finish to remove the kernel state process, zombie, all process outside the current process, after this call the free_some_memory function and discharge memory headroom as much as possible, to reduce the size of memory mirror.Then some drivings are handled, called pm_send_all (PM_SUSPEND, (void*) 3) hangs up suspend to the equipment transmitting system of having registered in power management PM module notice.

3), last do_software_suspend function call do_suspend_lowlevel function.Do_suspend_lowlevel calls do_magic_suspend_1 () successively, and save_processor_context () and three function d o_magic_suspend_1 of do_magic_suspend_2 () () call mb () and barrier () comes to prepare for ensuing save_processor_context function.Void Barrier (void) notice compiler inserts an internal memory barrier, but to hardware incapability, the code after the compiling can deposit the numerical value that all modifications in the current C PU register is crossed in internal memory, reads from internal memory again when needing these data again.

4), the save_processor_context function is preserved the operation context of the CPU under the current state.A struct who is used for preserving these values has _ attribute of attribute_ ((packed)), and can be with its member variable compact arrangement.

5), enter the do_magic_suspend_2 function afterwards.This function is responsible for memory pages is saved in the swap subregion.This function at first calls read_swapfiles finds appointment in preamble from possible 32 switching equipment (comprising file and subregion) subregion :/dev/modify.Call save_suspend_image then and finish after whole replicate runs, call suspend_power_down, close or restart computing machine.

Three, the loading of operating system mirror image

The boot of operating system is written into kernel with the operating system nucleus mirror image from the external memory of system, and decompresses in the process that starts, in start-up course, if boot finds to exist os starting to optimize mirror image, system is written into internal memory with this mirror image, and decompresses;

System foundes a virtual terminal earlier, is used for showing the system information of loading in the mirrored procedure, at first checks the correctness of mirror image, in order the memory pages in the mirror image is returned to internal memory then, and the context with processor recovers at last;

With ARM-Linux is example, and whole process is as follows:

1), after the start, BootLoader loads linux kernel and decompresses, start_kernel is up to creating the init process then.After the Init process transfer lock_kernel, before calling prepare_namespace loading rootfs, call the do_basic_setup function, carry out the setting on some bases, as pci, sbus, ecard etc., do_basic_setup calls start_context_thread () and do_initcalls () then, load the driver of some equipment, afterwards, enter the software_resume () function of oneself, prepare to check exchange partition, carry out the recovery of system.

2), the software_resume function carries out after some necessary conditions detections, call the read_suspend_image function, the mirror image of preserving before reading, read_suspend_image call bottom more _ read_suspend_image and finish mirror image initiatively and read work.

3), _ read_suspend_image calls prepare_suspend_console and creates a virtual control desk, is used for showing some relevant informations in the recovery.Calling bdev_read_page afterwards to obtain one page from/dev/modify equipment the inside, obtain the header information of mirror image, and whether calling the sanity_check function, to detect the header information of mirror image correct, to determine that whether present system is the system before the dormancy, if not then starting according to normal system.

4), next obtain the page (order of the order that reads memory pages when preserving memory pages is just in time opposite) of the internal memory of pagedir and all preservation by recursive call bdev_read_page.And the answer memory content, the read_suspend_image function returns then.Next call do_suspend_lowlevel and come recovery system further.

5), do_suspend_lowlevel is by to the setting of CPU register Cr3, the physical address of swapper_pg_dir as the physical address that changes to page directory.And, created new stack pointer by setting to ecx.CR3 is used to preserve the initial physical address of page directory tables.Because catalogue is page alignment, thus only high 20 effectively, low 12 reservations are not used.When packing a new value in CR3, low 12 are necessary for 0; But during value, low 12 are left in the basket from CR3.Whenever with the value of MOV instruction replacement CR3 the time, can cause the content invalid of paging mechanism high-speed buffer, use the method, can before enabling paging mechanism, promptly, refresh the high-speed cache of paging mechanism in advance before the PG position 1.Even the CR3 register in the PG position of CR0 register or the PE position be also can pack in 0 o'clock, under in real pattern, also CR3 can be set, so that carry out the initialization of paging mechanism.When task was switched, CR3 will be changed, if but the value of CR3 is identical with the value of CR3 in the former task in the new task, and processor does not refresh the branch page cache so, and execution speed is faster arranged when also showing with convenient task sharing.

6), next do_suspend_lowlevel calls do_magic_resume_1 () restore_processor_context and do_magic_resume_2 () finishes all resuming work successively.

7), do_magic_resume_1 () has called mb (); Barrier (); Come to prepare for ensuing restore_processor_context function.Void Barrier (void) notice compiler inserts an internal memory barrier, but to hardware incapability, the code after the compiling can deposit the numerical value that all modifications in the current C PU register is crossed in internal memory, reads from internal memory again when needing these data again.After this, to begin with sizeof (char) be that unit carries out duplicating of memory pages to do_suspend_lowlevel.Why can not use the such function of copy_page here, be because when memory pages duplicates, can destroy storehouse (stack), before the value of recovering the CPU register fully, can not re-use storehouse and local variable, but can use be positioned at the kernel code section _ variable in nosave district.Afterwards, restore_processor_context () function begins to recover the value of CPU register.Enter do_magic_resume_2 () at last.

8), do_magic_resume_2 () mainly handles the on-the-spot work of some cleanings.It discharges the pagedir that distributes when dormancy, to guarantee that internal memory returns to state before the dormancy fully then by calling pm_send_all (PM_RESUME, (void*) 0) recovers resume to the equipment transmitting system of having registered in power management PM module notice.So far, all work have all been finished.System has returned to the state before the dormancy.

The process flow diagram of total system as shown in drawings.

Claims (1)

1. the startup optimization method of an embedded OS image starting is characterized in that:
1) the operating system mirror image is prepared:
The boot of operating system is in the process that starts, at first will own copy memory headroom to from external memory, arrive the external memory location lookup operating system nucleus mirror image of appointment then, if the operating system nucleus mirror image exists, so the operating system nucleus mirror image is written into kernel from the external memory position of appointment, and decompress, in start-up course, operating system will judge whether to exist the operating system mirror image of quick startup, if there is no then normally start, if exist then startup fast, if boot is found only to have original operating system mirror image, operating system will start by normal procedure;
When os starting behind the most succinct groundwork state, the operating system sleeping process that will start the operating system, the groundwork of this process is to do various preliminary works for preserving the operating system mirror image;
2) preserve the operating system mirror image:
The preservation work of this part main complete operation system for field, at first will set up a virtual terminal, be used for simulating the communication of kernel and user's attitude process, begin to finish to remove the kernel state process then, zombie, all process outside the current process, after this call function discharges memory headroom as much as possible, to reduce the size of operating system mirror image, then some drivings is handled, to all peripheral hardwares of in power-supply management system, registering, send suspending command, insert the internal memory barrier then, begin to preserve the processor context, at last operating system is mirrored on the outside appointed equipment, at last with the operating system dormancy;
3) loading of operating system mirror image:
The boot of operating system is in the process that starts, at first will own copy memory headroom to from external memory, arrive the external memory location lookup operating system nucleus mirror image of appointment then, if the operating system nucleus mirror image exists, so the operating system nucleus mirror image is written into kernel from the external memory position of appointment, and decompresses, in start-up course, if boot is found to have the operating system mirror image that starts fast, operating system is written into internal memory with this mirror image, and decompresses;
Operating system is founded a virtual terminal earlier, be used for showing the system information of loading in the mirrored procedure, the correctness of checked operation system image at first, if it is correct, then the operating system mirror image is written into, the page of all preservations is obtained in circulation, and recovery memory content, value by the internal memory barrier restore processor register that inserts, discharge the memory headroom of applying in the rejuvenation at last, and recover notice by the equipment transmitting system in power management module, registered, finish the loading of whole operation system image.
CNB2006100498559A 2006-03-15 2006-03-15 Mirror starting optimization of built-in operation system CN100383739C (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CNB2006100498559A CN100383739C (en) 2006-03-15 2006-03-15 Mirror starting optimization of built-in operation system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CNB2006100498559A CN100383739C (en) 2006-03-15 2006-03-15 Mirror starting optimization of built-in operation system

Publications (2)

Publication Number Publication Date
CN1818869A CN1818869A (en) 2006-08-16
CN100383739C true CN100383739C (en) 2008-04-23

Family

ID=36918897

Family Applications (1)

Application Number Title Priority Date Filing Date
CNB2006100498559A CN100383739C (en) 2006-03-15 2006-03-15 Mirror starting optimization of built-in operation system

Country Status (1)

Country Link
CN (1) CN100383739C (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101373437B (en) * 2007-04-10 2012-09-05 标准微系统公司 Accessing safety memory by embedded controller to increase system safety

Families Citing this family (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8001336B2 (en) * 2007-03-02 2011-08-16 International Business Machines Corporation Deterministic memory management in a computing environment
CN101216772B (en) * 2008-01-15 2011-05-11 中兴通讯股份有限公司 Embedded equipment startup method and software upgrading method
CN101515194B (en) 2008-02-21 2011-11-09 鸿富锦精密工业(深圳)有限公司 Computer operating state converting method
CN101576829B (en) * 2009-06-12 2011-08-24 宋志飞 Trusteeship method and system for application process in embedded linux system
CN101593122B (en) * 2009-06-24 2012-09-05 中兴通讯股份有限公司 Method and device for starting embedded system
CN102012823A (en) * 2009-09-04 2011-04-13 中兴通讯股份有限公司 Method and device for quickly restarting computer
CN103995711B (en) * 2010-03-30 2018-02-02 鸿富锦精密工业(深圳)有限公司 Electronic calculator device and its method for restarting
WO2012048459A1 (en) * 2010-10-12 2012-04-19 北京中科院软件中心有限公司 Method and systeme for loading running environments of remote terminal
CN102455917B (en) * 2010-10-22 2016-06-29 康佳集团股份有限公司 The startup control method of a kind of Web TV and device
KR101813641B1 (en) * 2011-11-15 2018-01-02 에스프린팅솔루션 주식회사 Image forming apparatus and method for booting the same having hibernation function
CN103514065B (en) * 2012-06-29 2016-08-10 国基电子(上海)有限公司 Tool automatically detects and recovers electronic installation and the method for function
US9600369B2 (en) 2013-12-10 2017-03-21 Huawei Technologies Co., Ltd. Operating system recovery method and apparatus, and terminal device
CN103729211B (en) * 2013-12-10 2017-08-04 华为技术有限公司 A kind of operating system recovery method, device and terminal device
CN105786547A (en) * 2014-12-26 2016-07-20 中兴通讯股份有限公司 Operation system restarting method and apparatus
CN104950703B (en) * 2015-03-31 2017-10-17 南通艾利特自动化有限公司 Industrial robot controller quick start method based on embedded system
CN107977233A (en) * 2016-10-19 2018-05-01 华为技术有限公司 The quick loading method of kernel mirror image file and device

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1334510A (en) * 2001-09-07 2002-02-06 清华大学 Remoteboot method of computer in network environment
US20020078338A1 (en) * 2000-12-15 2002-06-20 Ibm Corporation Method and apparatus for fast computer initialization
DE202004009434U1 (en) * 2004-06-15 2004-08-12 Eltec Elektronik Ag Computer system for cloning a process operating system including all applications programs and applications data has a working memory, a non-volatile memory and other hardware as required
WO2005029325A1 (en) * 2003-08-21 2005-03-31 Ti Technologies Limited (A New Zealand Company) Method and apparatus for booting a computer system

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020078338A1 (en) * 2000-12-15 2002-06-20 Ibm Corporation Method and apparatus for fast computer initialization
CN1334510A (en) * 2001-09-07 2002-02-06 清华大学 Remoteboot method of computer in network environment
WO2005029325A1 (en) * 2003-08-21 2005-03-31 Ti Technologies Limited (A New Zealand Company) Method and apparatus for booting a computer system
DE202004009434U1 (en) * 2004-06-15 2004-08-12 Eltec Elektronik Ag Computer system for cloning a process operating system including all applications programs and applications data has a working memory, a non-volatile memory and other hardware as required

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101373437B (en) * 2007-04-10 2012-09-05 标准微系统公司 Accessing safety memory by embedded controller to increase system safety

Also Published As

Publication number Publication date
CN1818869A (en) 2006-08-16

Similar Documents

Publication Publication Date Title
US9547542B1 (en) System and method for application isolation with live migration
US10394621B1 (en) Method and computer readable medium for providing checkpointing to windows application groups
US9460028B1 (en) Non-disruptive and minimally disruptive data migration in active-active clusters
DK3008600T3 (en) Backup of a virtual machine from a storage snapshot
US8694828B2 (en) Using virtual machine cloning to create a backup virtual machine in a fault tolerant system
US9870288B2 (en) Container-based processing method, apparatus, and system
US20160055062A1 (en) Systems and Methods for Maintaining a Virtual Failover Volume of a Target Computing System
US8433948B2 (en) Method and apparatus for realizing application high availability
KR101534540B1 (en) Method and device for implementing memory migration
US20150089117A1 (en) Computer system, memory management method and program thereof
JP5724477B2 (en) Migration program, information processing apparatus, migration method, and information processing system
US8032707B2 (en) Managing cache data and metadata
US8966315B2 (en) System-directed checkpointing implementation using a hypervisor layer
US6698017B1 (en) Software migration on an active processing element
US8417928B2 (en) Turbo boot systems and methods for subsequent booting from a captured data stored in a non-volatile semiconductor memory
US8296601B2 (en) Take over method for computer system
US7620784B2 (en) High speed nonvolatile memory device using parallel writing among a plurality of interfaces
US20140310708A1 (en) Mechanism for providing virtual machines for use by multiple users
US8656386B1 (en) Method to share identical files in a common area for virtual machines having the same operating system version and using a copy on write to place a copy of the shared identical file in a private area of the corresponding virtual machine when a virtual machine attempts to modify the shared identical file
US20150178119A1 (en) Hypervisor-based server duplication system and method and storage medium storing server duplication computer program
US6336161B1 (en) Computer configuration system and method with state and restoration from non-volatile semiconductor memory
US7657888B2 (en) Method for forking or migrating a virtual machine
JP3454492B2 (en) Window-based multi-user computer system and method
US7809975B2 (en) Recovering from a storage processor failure using write cache preservation
US10621052B1 (en) System and method for hybrid kernel and user-space checkpointing using a character device

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
C17 Cessation of patent right
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20080423

Termination date: 20120315