CN102831016B - A cloud computing physical machine recycling method and device thereof - Google Patents

Abstract

The invention discloses a physical machine recycle method of a cloud computing and a device thereof, which can ensure the entire properties of a physical machine in a cloud computing condition to a greatest extent. The physical machine recycle method comprises the following steps of: A, carrying out level ordering on hardware properties of an idle physical machine in the cloud computing condition; and B, switching off the physical machine according to an order of the hardware properties from low to high according to a level ordering result, so that the physical machine which is not switched off ensures the processing request of low load. The device comprises a level ordering unit and an equipment management unit; the level ordering unit is used for carrying out the level ordering on the hardware properties of the idle physical machine in the cloud computing condition; and the equipment management unit is used for switching off the physical machine according to the order of the hardware properties from low to high according to the level ordering result, so that the physical machine which is not switched off ensures the processing request of low load. With the adoption of the physical machine recycle method of the cloud computing and the device thereof provided by the invention, the purposes of saving resources and reducing maintenance cost can be realized.

Description

A cloud computing physical machine recycling method and device thereof

technical field

The invention relates to the field of cloud computing, in particular to a cloud computing physical machine recycling method and device thereof.

Background technique

At present, the development and application of cloud computing are being carried out in more and more industries, and the number of tasks performed in the cloud computing environment is also increasing. Therefore, the number of physical devices managed in the cloud computing environment is also increasing. The cost also increases with the development of the cloud environment.

How to accurately and timely complete the recovery of physical equipment data in the cloud computing environment, reduce the maintenance cost of the cloud computing environment, and at the same time meet the hardware platform needs of various businesses in the cloud computing environment has become a physical equipment management in the cloud environment. key issues.

As shown in Figure 1, the current computer manufacturer’s solution is to maintain equipment in a cloud computing environment, and use shutdown operations on idle physical equipment, but the shutdown operation is random, so it is possible to use low-configuration physical equipment. The appliance acts as the primary computer that distributes the load. It is easy to cause problems such as system performance degradation and even business interruption.

Contents of the invention

In order to guarantee the overall performance of the physical machine in the cloud computing environment to the greatest extent, and realize the purpose of resource saving and cost saving, without affecting various services running in the cloud computing environment, the present invention proposes a cloud computing physical machine recycling Method and device thereof.

In order to solve the above technical problems, the present invention provides a physical machine recovery method for cloud computing, including

A. Rank the hardware performance of the idle physical machines in the cloud computing environment;

B. According to the ranking results, the physical machines are shut down according to the order of hardware performance from low to high, so that the physical machines that are not shut down can meet the processing requirements of low load.

Further, the hardware performance includes CPU performance, memory performance and hard disk performance.

Further, in step B, standby processing is also adopted for the remaining physical machines.

Further, the way of registering and sorting in step A is:

Quantify the hardware performance of the idle physical machines, and sort the hardware performance of the idle physical machines according to the quantized value.

In order to solve the above technical problems, the present invention also provides a cloud computing physical machine recycling device, including a ranking unit and a device management unit,

The ranking unit is used to rank the hardware performance of the idle physical machines in the cloud computing environment;

The device management unit is used for shutting down the physical machines according to the order of hardware performance from low to high according to the result of ranking, so that the physical machines that are not shut down can guarantee low-load processing requirements.

Preferably, the device management unit is further configured to take standby processing for the remaining physical machines.

Preferably, the ranking unit includes a quantization module and a comparison module,

The quantization module is used to quantify the hardware performance of the idle physical machine;

The comparison module is used to sort the hardware performance of the idle physical machines according to the values obtained by the quantization module.

Compared with the prior art, the present invention achieves the purpose of saving resources and reducing maintenance costs, and when shutting down some physical machines, the overall processing capability of the physical machines in the cloud computing environment is guaranteed to the greatest extent. By shutting down the physical machines with low performance, ensure that the physical machines with high performance are in standby state, so as to cope with the load increase and the transferred computing tasks in the cloud computing environment at any time, and avoid the random shutdown of some idle physical machines when the load is low It also makes cloud computing more economical and humane in the management of physical machines, avoiding waste of resources and energy.

Description of drawings

Fig. 1 is a schematic diagram of recycling idle resources in a traditional cloud environment; Fig. 2 is a schematic structural diagram of a physical machine recycling device for cloud computing according to an embodiment of the present invention; Fig. 3 is a schematic diagram of recycling idle resources under a cloud environment according to an embodiment of the present invention.

Detailed ways

In order to make the purpose, technical solution and advantages of the present invention more clear, the embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings. It should be noted that, in the case of no conflict, the embodiments in the present application and the features in the embodiments can be combined arbitrarily with each other.

Embodiments of the present invention are illustrated in conjunction with Fig. 2 and Fig. 3, the physical machine reclaiming device of the cloud computing of the embodiment of the present invention, the rank sorting unit 1 and the equipment management unit 2, and the rank sorting unit 1 is used for idle under the cloud computing environment The hardware performance of the physical machine is graded; the device management unit 2 is used to close the physical machine with low performance according to the order of hardware performance from low to high according to the result of the grade sort, so that the physical machine with high performance is in a standby state, Make the non-shutdown physical machine guarantee low-load processing requirements.

Wherein the rank sorting unit 1 further comprises a quantization module 11 and a comparison module 12, at first the hardware performance of the idle physical machine is quantified by the quantization module 11, and then the hardware performance of the idle physical machine is quantified by the comparison module 12 according to the numerical value after quantization Sort. The physical machine recovery method of cloud computing according to the embodiment of the present invention, the related process includes:

A. Rank the hardware performance of the idle physical machines in the cloud computing environment;

B. According to the ranking results, the physical machines are shut down according to the order of hardware performance from low to high, so that the physical machines that are not shut down can meet the processing requirements of low load.

The hardware performance includes CPU performance, memory performance and hard disk performance.

The cloud computing physical machine recovery method in the embodiment of the present invention ranks the idle physical machines in the cloud computing environment when the load is low, and adopts the method of quantifying the hardware performance of the idle physical machines, according to the overall quantified hardware performance The value divides the idle physical machines into Class A, Class B, and Class C, where Class A is a physical machine with high performance, Class C is a physical machine with low performance, and Class B is a physical machine with intermediate performance.

When performing shutdown operations on idle physical machines in a cloud computing environment, perform selective shutdown according to the classification of physical machines. In the case of class A, class B, and class C, class C physical machines are preferentially shut down. The second is Class B physical machines, and the last is some Class A physical machines, so that the physical machines that are not shut down can guarantee the overall processing capacity. In this way, it can ensure that the hardware performance in the cloud environment is guaranteed to the greatest extent while saving resources.

Example

As shown in Figure 3, the CPU performance, memory performance and hard disk performance of idle physical machines in the cloud computing environment are quantified. %, for the CPU performance of each idle physical machine, the quantitative value of the total number of cores above 16 cores is 60, the quantitative value between 8 cores and 16 cores is 50, and the quantitative value between 4 cores and 8 cores It is 40, and the quantization value below 2 cores is 20. For the memory performance of each idle physical machine, the quantized value above 32G is 20, the quantized value between 16G and 32G is 15, and the quantized value below 8G is 10. For the hard disk performance of each idle physical machine, the quantitative value is 20 for the capacity above 6T, and 10 for the capacity below 2T. According to the quantified sum of the three kinds of hardware performance as the standard, rank the idle physical machines, classify the quantized sum above 85 as a class A physical machine, and classify as a class B physical machine with a quantized sum of 60 to 85 , 60 points or less are class C physical machines.

Class A physical machines have excellent performance and good operating conditions, such as devices 6 and 8 in Figure 3; Class C physical machines have low performance, such as Devices 3 and 9 in Figure 3; the rest are Class B physical machines.

When some physical machines need to be shut down due to the low load in the negative environment of cloud computing, according to the results of ranking, class C physical machines are shut down first, followed by class B physical machines, and class A physical machines with excellent performance are reserved to prevent cloud computing The load task sharing when the pressure load increases, on the basis of effectively saving resources, achieves the purpose of optimizing the entire cloud computing environment.

Although the embodiments disclosed in the present invention are as above, the described content is only an embodiment adopted for the convenience of understanding the present invention, and is not intended to limit the present invention. Anyone skilled in the technical field to which the present invention belongs can make any modifications and changes in the form and details of the implementation without departing from the spirit and scope disclosed by the present invention, but the patent protection scope of the present invention, The scope defined by the appended claims must still prevail.

Claims (4)

1. A physical machine recycling method for cloud computing, characterized in that, comprising A. Rank the hardware performance of the idle physical machines in the cloud computing environment; B. According to the ranking results, the physical machines are shut down in the order of hardware performance from low to high, so that the physical machines that are not shut down can meet the processing requirements of low load; Described hardware performance comprises CPU performance, memory performance and hard disk performance; Wherein, The described ranking of the hardware performance of the idle physical machines under the cloud computing environment includes: Quantify the CPU performance, memory performance and hard disk performance of idle physical machines in the cloud computing environment. The quantified value is up to 100 points, of which CPU performance is up to 60 points, memory performance is up to 20 points, and hard disk performance is up to 20 points; For the CPU performance of each idle physical machine, the quantitative value of the total number of cores above 16 cores is 60 points, the quantitative value between 8 cores and 16 cores is 50 points, and the quantitative value between 4 cores and 8 cores For the memory performance of each idle physical machine, the quantized value above 32G is 20 points, the quantized value between 16G and 32G is 15 points, and the quantized value below 8G 10 points; for the hard disk performance of each idle physical machine, the quantitative value of the capacity above 6T is 20 points, and the quantitative value of the capacity below 2T is 10 points; According to the quantitative sum of the three kinds of hardware performance as the standard, rank the idle physical machines, classify the quantitative sum of 85 points or more as Class A physical machines, and classify between 60 and 85 points as Class B physical machines, Those with a score below 60 are Class C physical machines. 2. The recovery method according to claim 1, characterized in that in step B, standby processing is also adopted for the remaining physical machines. 3. A physical machine recycling device for cloud computing, characterized in that, the recycling device includes a rank sorting unit (1) and an equipment management unit (2), The ranking unit (1) is used to rank the hardware performance of the idle physical machines in the cloud computing environment; The device management unit (2) is used to shut down the physical machines according to the order of hardware performance from low to high according to the result of ranking, so that the physical machines that are not shut down can meet the processing requirements of low load; Described hardware performance comprises CPU performance, memory performance and hard disk performance; Wherein, The ranking unit (1) includes a quantization module (11) and a comparison module (12); The quantization module (11) is used to quantify the CPU performance, memory performance and hard disk performance of idle physical machines in the cloud computing environment. The quantification value is up to 100 points, wherein the CPU performance is the highest 60 points, and the memory performance is the highest. 20 points, hard disk performance takes up to 20 points; for the CPU performance of each idle physical machine, the quantitative value of the total number of cores above 16 cores is 60 points, and the quantitative value between 8 cores and 16 cores is 50 points. The quantitative value between 4 cores and 8 cores is 40 points, and the quantitative value below 2 cores is 20 points; for the memory performance of each idle physical machine, the quantitative value above 32G is 20 points, and the quantitative value between 16G and 32G is 20 points. The numerical value is 15 points, and the quantitative value below 8G is 10 points; for the hard disk performance of each idle physical machine, the quantitative value is 20 points if the capacity is above 6T, and the quantitative value is 10 points if the capacity is below 2T; The comparison module (12) is used to sort the idle physical machines according to the quantitative sum of the three hardware performances as a standard, and classify the physical machines with a quantized sum of more than 85 points as Class A physical machines, with 60 points to 85 points Those with a score between 1 and 2 are classified as Class B physical machines, and those with a score below 60 are classified as Class C physical machines. 4. The recovery device according to claim 3, characterized in that, the device management unit (2) is further configured to take standby processing for the remaining physical machines.