CN102479111A - Data storage method under Linux operating system and Linux computer system thereof - Google Patents

Abstract

The present invention provides a data storage method under a Linux operating system and a Linux computer system. The method comprises the following steps: determining whether a shared memory is in use. If not, a first virtual machine obtains the right to use the shared memory. The first virtual machine determines whether a second virtual machine is in a dead state based on the right to use. If so, the first virtual machine stores the first data and the second data stored in the shared memory of the first virtual machine and the second virtual machine respectively in a storage device based on the right to use and according to the dead state.

Description

Data storage method under Linux operating system and Linux computer system thereof

technical field

The invention relates to a data storage method under a Linux operating system and a Linux computer system thereof.

Background technique

The operating system (Operation System, OS) is equivalent to the central processing unit (Central Process Unit, CPU) in computer equipment, and is the heart of the software program. Each execution command (command) and application program (application program) and other software are driven by the operating system to use hardware components in computer equipment, such as motherboards, memory, hard disks, optical drives, display cards, sound cards, network cards, Various adapter cards and so on.

For current personal or family use, Microsoft's window operating system (Microsoft Windows) and Macintosh's MacOS operating system are competing against each other. The computer equipment used in enterprises must shoulder huge data traffic and high-speed computing performance. Therefore, the stability and operation performance of the operating system are important considerations. In addition to the development of the Unix operating system that has been developed for decades, today The Linux operating system, which is similar in function to the Unix operating system, is gradually favored by enterprises.

The Linux operating system can be said to be a derivative of the Unix operating system. It was designed by a programmer in Helsinki around October 1991. It inherits the advantages of the good stability of the Unix operating system, coupled with the open source code, so Enterprises can tailor appropriate functions for different needs. The development of related application software is also quite rapid. Today, the Linux operating system has become one of the most important operating systems in the industry.

The Linux operating system is positioned as a network operating system. Its operation mode is different from that of the window interface. It is operated by the user typing in command data in the form of a command line. For those familiar with graphical environments, Linux may appear primitive at first glance. But the operation of the Linux operating system focuses on its intrinsic functions rather than superficial things. Even in a text-only environment, it also has very advanced network, script (script) and security capabilities. The Linux operating system is expected to perform these tasks in cooperation with other systems on the network. The automatic execution ability of the Linux operating system is also very strong. Only need to design batch file files, the system can complete very detailed tasks.

In today's Internet era, under the Linux operating system, users can add many parameters after the command to complete many different tasks. The powerful network management performance of the Linux operating system is beyond doubt. Having said that, the Linux operating system still has some room for improvement when it comes to writing data asynchronously. Asynchronous writing is an important technology for improving the performance of the storage device, and its main function is to accelerate the writing speed of the storage device through the memory. The method of asynchronous writing under the current Linux operating system is to use the aio_write command to allocate a specific memory block to use the cache to speed up writing data. However, when the virtual machine crashes in this way, the memory used for the cache by the aio_write command will disappear, causing an error in writing data to the storage device.

Contents of the invention

In order to solve the problems of the known technology, an object of the present invention is to provide a data storage method under the Linux operating system, by sharing the same memory block technology of multiple virtual machines common in new virtual machines, providing a parallel use The memory allocation mechanism allows the virtual machine using the Linux operating system as the virtual operating system (Guest OS) to have the function of preventing asynchronous write errors on the advanced disk array card with a battery as usual, thereby eliminating the need to crash without warning , an error written asynchronously.

According to an embodiment of the present invention, a data storage method under a Linux operating system includes the following steps. Determine whether shared memory is in use. If not, the first virtual machine obtains the right to use the shared memory. Make the first virtual machine judge whether there is a second virtual machine in a dead state based on the obtained usage right. If so, make the first virtual machine store the first data and the second data respectively stored in the shared memory of the first virtual machine and the second virtual machine in the storage device based on the obtained usage right and according to the crash state.

Another object of the present invention is to provide a Linux computer system.

According to another embodiment of the present invention, a Linux computer system includes a shared memory, a first virtual machine, a second virtual machine, a storage device, and a management module. Shared memory provides usage rights. The first virtual machine has first data. The first data is stored in the shared memory. Wherein the first virtual machine obtains the usage right. The second virtual machine has second data. The second data is stored in the shared memory. Wherein the second virtual machine is in a dead state. The management module is used for allowing the first virtual machine to store the first data in the storage device based on the acquired usage right, and store the second data in the storage device according to the crash state.

The data storage method under the Linux operating system of the present invention and the Linux computer system thereof mainly provide a memory allocation mechanism used in parallel through the common technology of multiple virtual machines sharing the same memory block in new virtual machines, so that Linux can be used As a virtual machine of the virtual operating system (Guest OS), the operating system has the function of preventing asynchronous writing errors on an advanced disk array card with a battery, thereby eliminating asynchronous writing errors when there is no warning crash.

Description of drawings

FIG. 1 is a flow chart illustrating the steps of a data storage method under a Linux operating system according to an embodiment of the present invention;

FIG. 2 is a schematic diagram illustrating a Linux computer system according to an embodiment of the present invention.

[Description of main component symbols]

1: Linux computer system 10: Management module

12: The first virtual machine 120: The first private memory block

14: The second virtual machine 140: The second private memory block

16: Spinlock module 18: Status record module

20: Shared memory 22: Storage device

S100～S114: process steps

Detailed ways

Several implementations of the present invention will be disclosed below with the accompanying drawings. For the sake of clarity, many practical details will be described together in the following description. It should be understood, however, that these practical details should not be used to limit the invention. That is, in some embodiments of the present invention, these practical details are unnecessary. In addition, for the sake of simplifying the drawings, some known and commonly used structures and components will be shown in a simple and schematic way in the drawings.

A technical solution of the present invention is a data storage method under the Linux operating system. More specifically, it mainly provides a memory allocation mechanism used in parallel through the common technology of multi-virtual machines sharing the same memory block in new virtual machines, so that the Linux operating system as the virtual operating system (Guest OS) The virtual machine has the function of preventing asynchronous writing errors on an advanced disk array card with a battery as usual, thereby eliminating the asynchronous writing errors when crashing without warning.

Please refer to FIG. 1 , which is a flowchart illustrating the steps of a data storage method under the Linux operating system according to an embodiment of the present invention.

As shown in FIG. 1 , if the first virtual machine wants to use the shared memory, the method for storing data under the Linux operating system according to this embodiment can execute step S100 .

S100: Determine whether the shared memory is in use.

In the asynchronous write operation under the Linux system, if you want to avoid the race problem (that is, the data overwrite error caused by the shared memory being written by multiple virtual machines at the same time), you must first determine the shared memory used by all virtual machines. Whether the memory is occupied by any virtual machine, and it must also be ensured that the shared memory can only be used by at most one virtual machine at any time. Therefore, according to the data storage method under the Linux operating system in this embodiment, step S100 must be executed before the first virtual machine uses the shared memory.

Wherein, if the judgment result of step S100 is yes, step S102 is further executed.

S102: Re-execute step S100 after waiting for a predetermined time.

In other words, if the determination result of step S100 is that the shared memory is being used by a certain virtual machine, the first virtual computer cannot be used. If the first virtual computer wants to use the shared memory, it must wait for the virtual machine using the shared memory to be used up, so it is possible to wait for a predetermined time to confirm whether the shared memory is not in use. The aforementioned predetermined time is not limited, and can be flexibly changed according to requirements or limitations in practical applications.

In contrast, if the judgment result of step S100 is negative, it means that no virtual computer is currently using the shared memory, and the first virtual machine wants to use the shared memory, and step S104 can be further executed.

S104: Obtain the right to use the shared memory by the first virtual machine.

After the first virtual machine has obtained the right to use the shared memory, it means that the first virtual machine has obtained the dominant right to use the shared memory, and other virtual machines cannot use it at the same time. If other virtual machines want to use the shared memory, they still have to wait for the first virtual machine to finish using it and return the right to use it.

In addition, under the architecture of asynchronous writing under the Linux system, a virtual machine crash may occur. If the virtual machine crashes when it obtains the right to use and uses the shared memory, the data it was originally using in the shared memory It may be too late to be stored in the storage device, and disappear with the crash of the virtual machine. Therefore, according to the data storage method under the Linux operating system in this embodiment, after the first virtual machine obtains the right to use the shared memory, step S106 can be executed again.

S106: Make the first virtual machine determine whether there is a second virtual machine in a dead state based on the obtained usage right.

Wherein, if the judgment result of step S106 is yes, step S108 is further executed.

S108: Make the first virtual machine store the first data and the second data respectively stored in the shared memory of the first virtual machine and the second virtual machine in the storage device based on the obtained usage right and according to the shutdown state.

In other words, after the first virtual machine obtains the right to use the shared memory, according to step S106, it is found that the second virtual machine is in a dead state, then the first virtual machine can enter the shared memory to search for the second virtual machine according to the dead state of the second virtual machine. Second data in use before the crash exists in the second memory. Then, according to step S108, the first virtual machine can store the first data being used in the shared memory in the storage device based on the obtained ownership, and replace the second virtual machine with the The second data temporarily stored in the shared memory before the crash is also stored in the storage device together.

After the first virtual machine finishes using the shared memory (that is, both the first data of the first virtual machine and the second data of the second virtual machine are stored in the storage device), the data under the Linux operating system according to this embodiment In the storage method, step S110 may be performed next.

S110: Make the first virtual machine return the use right after the first data and the second data are stored in the storage device.

Relatively, if the judgment result of step S106 is negative, it means that before the first virtual machine obtains the right to use the shared memory, no virtual machine has crashed. method, continue to execute step S112.

S112: Make the first virtual machine store the first data stored in the shared memory by the first virtual machine in the storage device based on the obtained usage right.

In other words, if the first virtual machine finds that no virtual machine has crashed based on the usage right, then the first virtual machine only needs to focus on processing the first data used by itself in the shared memory, and use After completion (that is, storing the first data of the first virtual machine to the storage device), the data storage method under the Linux operating system according to this embodiment executes step S114.

S114: Make the first virtual machine return the usage right after the first data is stored in the storage device.

Another technical solution of the present invention is a Linux computer system.

Please refer to FIG. 2 , which is a schematic diagram illustrating a Linux computer system 1 according to an embodiment of the present invention. As shown in FIG. 2 , the Linux computer system 1 of this embodiment may include a shared memory 20 , a first virtual machine 12 , a second virtual machine 14 , a storage device 22 and a management module 10 . Shared memory 20 may provide usage rights. Of course, as mentioned in the previous embodiment, if the asynchronous writing operation under the Linux system is to avoid the racing problem, it must be ensured that the right to use the shared memory 20 can only be used by at most one virtual machine at any time.

As shown in FIG. 2 , in this embodiment, the right to use the shared memory 20 is obtained by the first virtual machine 12 . In addition, the first virtual machine 12 includes a first private memory block 120 . The first private memory block 120 can be used to temporarily store the first data being processed by the first virtual machine 12, and after obtaining the ownership of the shared memory 20, the first data of the first virtual machine 12 can be further temporarily stored in the shared memory 20 in. The second virtual machine 14 includes a second private memory block 140 . The second private memory block 140 can be used to temporarily store the second data. However, in this embodiment, since the second virtual machine 14 is in a shutdown state, it means that the second data of the second virtual machine 14 is still temporarily stored in the shared memory 20 before the second virtual machine 14 crashes.

As shown in FIG. 2 , the above-mentioned management module 10 can allow the first virtual machine 12 to store the first data temporarily stored in the shared memory 20 to the storage device 22 based on the use right when the first virtual machine 12 obtains the use right, And when the second virtual machine 14 is in a crash state, the first virtual machine 12 is allowed to store the second data temporarily stored in the shared memory 20 to the storage device 22 according to the crash state. In this way, the second virtual machine 14 in the crash state can store the second data temporarily stored in the shared memory 20 before the crash in the storage device 22 with the assistance of the first virtual machine 12, and the second data is It will not disappear with the shutdown of the second virtual machine 14. In addition, the above-mentioned management module can make the first virtual machine 12 return the right to use the shared memory 20 after the first data and the second data are stored in the storage device 22 .

As shown in FIG. 2 , the Linux computer system 1 of this embodiment may further include a spinlock module 16 . The spinlock module 16 can mark the shared memory 20 as being in use when the first virtual machine 12 obtains the use right, so that other virtual machines cannot use the shared memory 20 at the same time. Moreover, the spinlock module 16 may also cancel the use state of the shared memory 20 after the first virtual machine 12 returns the use right. Subsequently, if other virtual machines want to use the shared memory 20 , they can repeat the steps of obtaining the right to use the shared memory 20 first, and then mark the shared memory 20 as being in use through the spinlock module 16 .

Also shown in FIG. 2 , the Linux computer system 1 of this embodiment may further include a stateful recording module 18 . The state recording module 18 can be used to regularly create a comparison table according to the survival states of the first virtual machine 12 and the second virtual machine 14 . In this way, the management module 10 can allow the first virtual machine 12 to determine that the second virtual machine 14 is in a dead state based on the usage rights and the regularly updated comparison table. In other words, the first virtual machine 12 judges whether the second virtual machine 14 is in a dead state through the comparison table created by the state recording module 18 .

In practical applications, the number of the above-mentioned second virtual machines 14 in a dead state is not limited to this embodiment, as long as they are virtual machines in a dead state, they can be regarded as the second virtual machines 14 of this embodiment, and The first virtual machine 12 that has obtained the right to use the shared memory 20 can further store the second data stored in the shared memory 20 by all the second virtual machines 14 that are in a shutdown state in the storage device 22 .

From the above detailed description of specific embodiments of the present invention, it can be clearly seen that the data storage method under the Linux operating system of the present invention and its Linux computer system mainly share the same virtual machine through common multi-virtual machines in new virtual machines. The memory block technology provides a memory allocation mechanism used in parallel, so that the virtual machine using the Linux operating system as the virtual operating system (Guest OS) has the ability to prevent asynchronous writing on the advanced disk array card with a battery as usual. The error function can be added to eliminate the error of asynchronous writing when there is no warning crash.

Although the present invention has been disclosed above in terms of implementation, it is not intended to limit the present invention. Any skilled person can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the present invention The scope of protection should be based on the scope defined by the appended claims.

Claims (9)

1. the data storage method under the (SuSE) Linux OS is characterized in that, comprises the following step:

(a) judge that one shares internal memory and whether is in a user mode, if not, execution in step (b) then;

(b) obtain the right to use of this shared drive by one first virtual machine;

(c) make this first virtual machine judge whether that through the right to use that obtains one second virtual machine is in a deadlock state based on this, if, execution in step (d) then; And

(d) make this first virtual machine based on this through the right to use that obtains and according to this deadlock state, this first virtual machine and this second virtual machine are stored in respectively in one first data and one second data storing to a storage device in this shared drive.

2. the data storage method under the (SuSE) Linux OS according to claim 1 is characterized in that, if the judged result of step (a) is for being then to carry out the following step:

(a1) wait for a schedule time after execution in step (a) again.

3. the data storage method under the (SuSE) Linux OS according to claim 1 is characterized in that, if the judged result of step (c) is that then step (c) does not further comprise the following step:

(d1) make this first virtual machine based on this right to use, this first virtual machine is stored in this first data storing in this shared drive to this storage device through obtaining.

4. the data storage method under the (SuSE) Linux OS according to claim 3 is characterized in that, step (d1) further comprises the following step:

(e1) in this first data storing after this storage device, make this first virtual machine give back this right to use.

5. the data storage method under the (SuSE) Linux OS according to claim 3 is characterized in that, further comprises the following step:

(e2) in these first data and this second data storing to this storage device, make this first virtual machine give back this right to use.

6. a Linux computer system is characterized in that, comprises:

One shares internal memory, and a right to use is provided;

One first virtual machine has one first data, can be stored in this shared drive;

One second virtual machine has one second data, can be stored in this shared drive;

One storage device; And

One administration module; When being used to this first virtual machine and obtaining this right to use; Allow this first virtual machine based on this right to use with this first data storing to this storage device; And when being used to this second virtual machine and being in a deadlock state, allow this first virtual machine according to this deadlock state with this second data storing to this storage device.

7. Linux computer system according to claim 6 is characterized in that, further comprises:

One spin lock module is used to this first virtual machine and obtains this this shared drive of right to use tense marker and be in a user mode, and after this first virtual machine is given back this right to use, cancels this user mode.

8. Linux computer system according to claim 7 is characterized in that, this management mould to this storage device, makes this first virtual machine give back this right to use in these first data and this second data storing.

9. Linux computer system according to claim 6 is characterized in that, further comprises:

One state recording module; In order to regularly to make comparison list according to the existing state of this first virtual machine and this second virtual machine, wherein this administration module allows this first virtual machine to judge that based on this right to use and according to this table of comparisons this second virtual machine is in this deadlock state.

Images