WO2021003704A1

WO2021003704A1 - Method and apparatus for performing geographically distributed process mapping employing privacy constraint condition, and terminal

Info

Publication number: WO2021003704A1
Application number: PCT/CN2019/095458
Authority: WO
Inventors: 周池; 肖遥; 毛睿
Original assignee: 深圳大学
Priority date: 2019-07-10
Filing date: 2019-07-10
Publication date: 2021-01-14

Abstract

Disclosed are a method and apparatus for performing geographically distributed process mapping employing a privacy constraint condition, and a terminal. The method comprises: determining whether the number of privacy levels of processing a graph G awaiting segmentation is greater than 1; if so, using the maximum privacy level in the processing of the graph G as a feature privacy level I _m; placing a process having a privacy level lower than I _m into a subgraph G ₁, placing a process having a privacy level equal to I _m into a subgraph G ₂, placing a data center having a privacy level lower than I _m into a set S ₁, and placing a data center having a privacy level equal to I _m into a set S ₂; determining whether the number of processes in the subgraph G ₁ is less than the number of available nodes in the set S ₁; if not, storing the subgraph G ₂ in a list for storing segmented subgraphs; and configuring the subgraph G ₁ to be a new graph G to be segmented, and returning to the step of determining whether the number of privacy levels of the processes in the graph G is greater than 1. Thus, the invention is suitable for process mapping employing a privacy constraint condition in a geographically distributed environment.

Description

Geographically distributed process mapping method, device and terminal containing privacy constraint conditions

Technical field

The invention relates to the technical field of process mapping algorithms, in particular to a geographically distributed process mapping method, device and terminal containing privacy constraint conditions.

Background technique

For the process mapping problem, the current leading level algorithm is a heuristic greedy algorithm (Greedy algorithm) proposed by Heofler et al. This algorithm is used to solve the process mapping problem in heterogeneous networks. This method is based on the greedy algorithm. Mapping strategy, this method has low overhead and can achieve better optimization results; the other is a mapping optimization method (MPIPP) proposed by Chen et al. for arbitrary message-passing applications, which is based on k- A method of way graph division. This method has a relatively large search space and is more applicable to the process mapping problem of any message-passing application.

However, these two methods are both researches on traditional mapping problems in cluster or grid computing, and neither of them considers the unique characteristics of process mapping in geographically distributed environments. The first is the network characteristics in a geographically distributed environment: the network bandwidth in the data center is much higher than the network bandwidth between the data centers; the network bandwidth between the data centers is highly correlated with the geographic distance between the data centers. The second is the data migration constraint in a geographically distributed environment: due to the varying degrees of data privacy protection in various countries and regions in the world, data is not allowed to migrate from areas with high data privacy protection levels to areas with low data privacy protection levels for processing. Therefore, the process mapping problem in a geographically distributed environment is a process mapping problem with constraints.

Since the Greedy algorithm and the MPIPP algorithm do not consider the characteristics of the process mapping in the geographically distributed environment, these two algorithms may not be suitable for the process mapping problem in the geographically distributed environment.

technical problem

The embodiments of the present invention provide a geographically distributed process mapping method, device and terminal with privacy constraints, aiming to solve the problem that the existing process mapping algorithm is not suitable for process mapping in a geographically distributed environment.

Technical solutions

In the first aspect, an embodiment of the present invention provides a geographically distributed process mapping method with privacy constraints, which includes:

Obtain the process map of the application, and set the process map of the application as the to-be-segmented map G;

Determine whether the number of privacy levels of the processes in the to-be-segmented graph G is greater than 1;

If the number of privacy levels of the processes in the graph G to be divided is greater than 1, the largest privacy level in the processes in the graph G to be divided is used as the characteristic privacy level _Im ;

The graph G to be segmented privacy level is less than the process wherein the privacy level I _m is put into the sub-G in FIG. _1, the graph G to be segmented privacy level is equal to the privacy level I _m wherein the discharge process into the sub-G in FIG. _2, the application of the privacy level is less than I _m privacy level characteristic data center put into the set S _1, the privacy level is equal to the application wherein the privacy level I _m The data center is placed in set S ₂ ;

Judging whether the number of processes in the subgraph G ₁ is less than the number of available nodes in the set S ₁ ;

If the number of processes in the sub-graph G ₁ is not less than the number of available nodes in the set S ₁ , the sub-graph G ₂ is saved as a divided sub-graph in the storage for storing the divided sub-graph List;

Set the sub-graph G ₁ as a new graph G to be segmented, and return to the step of judging whether the number of privacy levels of processes in the graph G to be segmented is greater than one.

Its further technical solution is that the method further includes:

If the number of privacy levels of the processes in the to-be-segmented graph G is equal to 1, the to-be-segmented graph G is saved in a list for storing divided sub-graphs.

Its further technical solution is that the method further includes:

If the number of processes in the subgraph of G ₁ is less than the set number of available nodes S1 to obtain the sub-processes each of G ₂ and G all the processes of FIG. ₁ of the sub-total traffic;

The subgraph of G _{2 is} moved to the maximum in the course of total traffic of all the processes the subgraph of G ₁ to the subgraph of G _1, provided the privacy level of the sub-process is _a G in FIG. The highest privacy level among all processes, and return to the step of judging whether the number of processes in the subgraph G ₁ is less than the number of available nodes in the set S ₁ .

A further technical solution is that, after storing the sub-graph G ₂ as a divided sub-graph in the list for storing the divided sub-graph, the method further includes:

Leave the subgraph G ₂ blank.

In the second aspect, embodiments of the present invention also provide a geographically distributed process mapping device with privacy constraints, which includes a unit for executing the above method.

In a third aspect, an embodiment of the present invention also provides a terminal, which includes a memory and a processor, the memory stores a computer program, and the processor implements the above method when the computer program is executed.

In a fourth aspect, an embodiment of the present invention also provides a computer-readable storage medium, the storage medium stores a computer program, and the computer program can implement the foregoing method when executed by a processor.

Beneficial effect

In the embodiment of the present invention, the original process graph is divided into a series of subgraphs and the data center set of each subgraph, and the corresponding data center set can meet the privacy protection requirements of all processes in the corresponding subgraph (that is, the data center set The privacy protection level is equal to or lower than the privacy protection level of the process), thereby solving the privacy protection constraint problem of process mapping in a geographically distributed environment.

Description of the drawings

In order to explain the technical solutions of the embodiments of the present invention more clearly, the following will briefly introduce the drawings used in the description of the embodiments. Obviously, the drawings in the following description are some embodiments of the present invention. Ordinary technicians can obtain other drawings based on these drawings without creative work.

Figure 1 is a schematic diagram of an application scenario of a geographically distributed process mapping method with privacy constraints provided by an embodiment of the present invention;

Fig. 2 is a schematic block diagram of a terminal according to an embodiment of the present invention.

Embodiments of the invention

The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of them. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative work shall fall within the protection scope of the present invention.

It should be understood that when used in this specification and the appended claims, the terms "including" and "including" indicate the existence of the described features, wholes, steps, operations, elements and/or components, but do not exclude one or The existence or addition of multiple other features, wholes, steps, operations, elements, components, and/or collections thereof.

It should also be understood that the terms used in this specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

It should be further understood that the term "and/or" used in the specification and appended claims of the present invention refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations .

As used in this specification and the appended claims, the term "if" can be interpreted as "when" or "once" or "in response to determination" or "in response to detection" depending on the context . Similarly, the phrase "if determined" or "if detected [described condition or event]" can be interpreted as meaning "once determined" or "response to determination" or "once detected [described condition or event]" depending on the context ]" or "in response to detection of [condition or event described]".

Example 1

The parameters used in this embodiment are shown in Table 1 below.

Table 1

The technical problem to be solved by this embodiment is described below.

A series of processes are allocated to each machine so that the communication between processes can effectively utilize the physical links in the network. Such a process is called process mapping. Considering that in a geo-distributed cloud environment, data movement is subject to many constraints such as privacy protection, and some processes can only be mapped to specific data center machines. Therefore, a problem different from traditional process mapping arises: constrained process mapping.

Define two matrices of size M*M, L _T and B _T , which respectively represent the delay and bandwidth between different data centers. The elements are L _T (k',l') and B _T (k',l') Respectively represent the delay and bandwidth between data center k'and l'. The elements on the diagonal of the matrix represent the delay and bandwidth in the data center.

Define the communication pattern matrix C _G , where the element C _G (i,j) represents the communication volume between process i and process j. Define a counting matrix A _G , where the element A _G (i,j) represents the number of times that process i sends information to process j.

According to the above definition, when process i is mapped to data center k’ and process j is mapped to data center l’, the communication overhead can be calculated according to the following formula:

Among them, w _i,j represents the communication volume between process i and process j, d _k',l' represents the network performance between data center k'and l', and f(w,d) is the cost function.

Define a constraint matrix C of size N*M, where each element C(i,j) indicates whether process i can be mapped to data center j (C(i,j) = 1 indicates that process i can be mapped to data center j, C(i,j)=0 means not possible); if process i can be mapped to all data centers (C(i,j)=1, j=1, 2,...,M), it means that process i does not Restrictions.

Define N-dimensional vector

Represents the mapping result, where the i-th element represents the data center to which process i is mapped.

Define M-dimensional vector

The i-th element represents the number of available nodes in data center i.

Define function

For counting vectors

The number of elements whose median is equal to m.

Based on the above definition, the problem can be described by the following formula:

minimize

To meet the conditions

as well as

among them,

Use formula (1) to calculate.

The problem to be solved in this embodiment is the optimization problem with constraints described in formulas (2) and (3).

In order to solve the above technical problems, referring to FIG. 1, the technical solution proposed in this embodiment includes the following steps.

S1: Obtain a process map of the application, and set the process map of the application as the to-be-splitting map G.

In a specific implementation, the process map of the application is acquired, and the acquired process map of the application is set as the to-be-divided graph G.

S2: Determine whether the number of privacy levels of the processes in the graph G to be divided is greater than 1.

In specific implementation, it is determined whether the number of privacy levels of the processes in the graph G to be divided is greater than one.

S3: If the number of privacy levels of the processes in the to-be-segmented graph G is equal to 1, the to-be-segmented graph G is saved in a list for storing divided sub-graphs.

In specific implementation, if the number of privacy levels of the processes in the graph G to be divided is equal to 1, it indicates that the privacy levels of all processes in the graph G to be divided are the same. The list of subgraphs. Specifically, the list may be a subgraphlist.

S4: If the number of privacy levels of the processes in the graph G to be divided is greater than 1, use the largest privacy level in the processes of the graph G to be divided as the characteristic privacy level _Im .

In specific implementation, if the number of privacy levels of the processes in the graph G to be divided is greater than 1, the largest privacy level in the processes of the graph G to be divided is taken as the characteristic privacy level _Im .

S5, the division of the graph G to be smaller than the privacy level I _m wherein the privacy level to the process into the sub-G in FIG. _1, the graph G to be divided is equal to the privacy level wherein the privacy level I _m to process into the sub-G in FIG. _2, the application will be less than the privacy level privacy level I _m wherein the data center put into the set S _1, the application level is equal to the privacy feature privacy level I _m to the data center into the collection S _2.

In particular embodiments, the privacy level is set at the maximum process will be described later is divided in G of FIG characterized privacy level after I _m, the graph G to be segmented is smaller than the privacy level privacy level I _m wherein the process into the ₁ to the sub-graph G, the graph G to be divided is equal to the privacy level I _m wherein privacy level to put the process of the sub-G in FIG. _2.

Meanwhile, the application of the privacy level is less than I _m privacy level characteristic data center put into the set S _1, the application of the privacy level equal privacy level I _m wherein the data center into the collection S ₂ in.

S6, the determination of the number of G sub-set of available S ₁ whether the number of nodes in the process are less than _1.

In a specific implementation, it is determined whether the number of processes in the sub-graph G ₁ is less than the number of available nodes in the set S ₁ .

S7, if the number of G sub-set of available S _a number of nodes in the process are less than ₁ to obtain the subgraph of G ₂ total communication with all the processes of the processes is _a subgraph of G the amount.

In specific implementation, if the number of processes in the sub-graph G ₁ is less than the number of available nodes in the set S ₁ , then each process in the sub-graph G ₂ and all processes in the sub-graph G ₁ are calculated Total traffic.

S8, the said sub-G in FIG. _{2 is} moved to the maximum total traffic process all the processes of the subgraph of G ₁ to G in FIG. ₁ of the sub, the privacy level provided a process for the subgraph of G The highest privacy level among all the processes in ₁ and return to step S6.

In particular embodiments, the subgraph of G _{2 is} moved in the process of maximum total traffic amount of all the processes the subgraph of G ₁ to G in FIG. ₁ of the sub, the privacy level provided for said sub-process G in FIG. ₁ the highest privacy level in all processes, and returns to step S6 again circulating the above steps until the subgraph of G ₁ is equal to the number of processes and _a set S of the data center until the total available number of nodes.

S9. If the number of processes in the subgraph G ₁ is not less than the number of available nodes in the set S ₁ , save the subgraph G ₂ as a divided subgraph to store the divided subgraph. Figure in the list.

In specific implementation, if the number of processes in the sub-graph G ₁ is not less than the number of available nodes in the set S ₁ , the sub-graph G ₂ is saved as a divided sub-graph for storing the divided sub-graph In the list of subgraphs.

It should be noted that, after saving the sub-graph G ₂ as a divided sub-graph in the list for storing the divided sub-graphs, the sub-graph G ₂ is left blank.

SlO, the sub-G in _{FIG. 1} G is provided in FIG divided to be new, and returns to step 2.

In specific implementation, set the subgraph G ₁ as the new to-be-divided graph G, and return to step 2 to repeat the above steps until the process graph of the application is divided into k (k is equal to the value of the process graph of the application). The number of privacy levels in all processes) subgraphs, and each subgraph has a corresponding set of data centers.

It should be noted that after the sub-graph G _{1 is} set as the new to-be-divided graph G, the sub-graph G ₁ needs to be blanked.

Example 2

This embodiment provides a specific example. In this embodiment, the application contains 8 processes, numbered 1-8; there are four data centers, divided into four groups, numbered 1-4, and each data center contains two nodes, namely

Refer to Table 2 below, the communication mode matrix C _{G is} used to represent the process communication diagram.

Table 2 Communication mode matrix C _G

	11	22	33	44	55	66	77	88
11		22
22	22		44
33		44		44
44			44		88
55				88		22
66					22		44
77						44		44
88							44

In Table 2, the first row and the first column are the process numbers, the unit of communication volume is MB, and empty means 0.

See Table 3 below, which is the process constraint condition matrix C of this embodiment.

Table 3 Process constraint condition matrix C

a

b

c

d

11	11	00	00	00
22	11	11	00	00
33	11	11	00	00
44	11	11	11	00
55	11	11	11	00
66	11	11	11	11
77	11	11	11	11
88	11	11	11	11

From the process constraint matrix C, the privacy protection level of each process can be obtained, as shown in Table 4. Each process includes a total of four privacy protection levels, namely level 1 to level 4. The lower the level, the stricter the privacy protection requirements of the process. , The fewer data centers can be mapped to.

Table 4 Privacy protection level of the process

Table 4 Privacy protection level of the process

Further, from the process constraint condition matrix C, the privacy protection level of each data center can also be obtained. Refer to Table 5, Level 1 to Level 4, which respectively indicate the lowest privacy levels of processes that can be mapped to the data center. For example, the privacy level of data center b is 2, which means that processes with privacy protection levels 2, 3, and 4 can be mapped to the data center, and processes with privacy protection level 1 cannot be mapped to the data center. The higher the privacy protection level of the data center, the better the privacy protection provided by the data center, which can accommodate processes with stricter privacy protection requirements.

Table 5 Privacy protection level of data center

数据中心data center	aa	bb	cc	dd
隐私等级Privacy level	11	22	33	44

The detailed steps of applying the process mapping method with privacy constraints in this embodiment are as follows:

Step 1. Select the highest privacy protection level 4 among the processes of the graph G to be divided, put processes with a privacy protection level lower than 4 into the subgraph G ₁ , and put the remaining processes into the subgraph G ₂ . .

In the specific implementation, the highest privacy protection level 4 among the processes of the graph G to be divided is selected, the processes with the privacy protection level lower than 4 are put into the subgraph G ₁ , and the remaining processes are put into the subgraph G ₂ . G in FIG. ₁ the child process with a 1,2,3,4,5; G in FIG. ₂ there are sub 6,7,8 processes. Put data centers with a privacy protection level lower than 4 into set S ₁ , and put the remaining data centers into set S ₂ . The data collection center S ₁ a, b, c; data collection center S ₂ d. The number of available nodes in the data center in the set S ₁ is calculated by the following formula:

I(s ₁ )=I(a)+I(b)+I(c)=6

Since the number of processes in the sub-graph G ₁ is 5, which is less than the number of available nodes in the data center in the set S ₁ , it is necessary to select a part of the processes in the sub-graph G ₂ into the sub-graph G ₁ .

Step 2. Calculate the total communication volume between each process in the sub-graph G _{2 and} all processes in the sub-graph G ₁ .

Specifically, calculate the total communication volume between process 6 and processes 1, 2, 3, 4, and 5:

q ₆ ＝C _G (5,6)+C _G (6,5)=2+2=4

Calculate the total communication volume between process 7 and processes 1, 2, 3, 4, and 5:

q ₇ =0

Calculate the total communication volume between process 8 and processes 1, 2, 3, 4, and 5:

q ₈ = 0

Select the process 6 with the largest total communication volume and put it into the subgraph G ₁ , and set the privacy protection level of the process 6 to the highest privacy protection level among all the processes in the subgraph G ₁ , that is, level 3. Compare the number of processes in the subgraph G _{1 with the} number of available nodes in the data center in the set S ₁ .

Since the number of processes in the subgraph G ₁ is 6, which is equal to the number of available nodes in the data center in the set S ₁ . Therefore, the process in the sub-graph G ₂ is no longer selected and placed in the sub-graph G ₁ .

Step 3. Put the subgraph G ₂ into the subgraphlist subgraphlist, mark the data center d in the set S ₂ as its corresponding data center, and set the subgraph G ₁ and the subgraph G _{2 to} be empty.

Since the processes in subgraph G ₁ are 1, 2, 3, 4, 5, and 6, the privacy protection levels are 1, 2, 2, 3, 3, and 3 respectively, and the number of privacy levels is greater than 1, so subgraph G ₁ Continue to divide. At this time, set the sub-graph G ₁ as the new to-be-divided picture G, and set the sub-graph G ₁ and the sub-graph G _{2 to} be empty.

Step 4. Select the highest privacy protection level 3, put processes with a privacy protection level lower than 3 into the subgraph G ₁ , and put the remaining processes into the subgraph G ₂ .

In the specific implementation, the highest privacy protection level 3 is selected, the processes with the privacy protection level lower than 3 are put into the subgraph G ₁ , and the remaining processes are put into the subgraph G ₂ . G in FIG. ₁ the child process with a 1,2,3, 1,2,2 privacy level, respectively; G in FIG. ₂ there are sub-processes 4,5,6, 3,3,3 privacy level respectively. Put data centers with a privacy protection level lower than 3 into set S ₁ , and put the remaining data centers into set S ₂ . The data collection center _{S. 1} a, b; data set S ₂ center c. Calculate the number of available nodes in the data center in the set S ₁ :

I(s ₁ )=I(a)+I(b)=4

Since the number of processes in the sub-graph G ₁ is 3, which is less than the number of available nodes 4 in the data center in the set S ₁ , it is necessary to select a part of the processes in the sub-graph G _{2 to} be placed in the sub-graph G ₁ .

Step 5. Calculate the total communication volume between each process in the sub-graph G _{2 and} all processes in the sub-graph G ₁ .

Specifically, calculate the total communication volume between process 4 and processes 1, 2, and 3:

q ₄ ＝C _G (3,4)+C _G (4,3)=4+4=8

Calculate the total communication volume between process 5 and processes 1, 2, and 3:

q ₅ =0

Calculate the total communication volume between process 6 and processes 1, 2, and 3:

q ₆ = 0

Select the process 4 with the largest total communication volume and put it into the subgraph G ₁ , and set the privacy protection level of process 4 to the highest privacy protection level among all processes in G ₁ , that is, level 2. Comparative subgraph of G having ₁ Process ₁ in the set S of nodes of the data center is available: the number of sub-G ₁ in FIG. 4 is a process, set S ₁ is equal to the data center available nodes. Therefore, the process in the sub-graph G ₂ is no longer selected and placed in the sub-graph G ₁ .

Step 6. Put the subgraph G ₂ into the subgraphlist subgraphlist, mark the data center c in the set S ₂ as its corresponding data center, and set the subgraph G ₁ and the subgraph G _{2 to} be empty.

In particular embodiments, the sub-G ₁ in the process of FIG 1,2,3,4, 1,2,2,2 privacy level respectively, the number of privacy level greater than 1, it is necessary to continue dividing _a sub-G in FIG. Set the subgraph G ₁ as a new segmented graph G, and at the same time, set the subgraph G ₁ and the subgraph G _{2 to} be empty.

Step 7. Select the highest privacy protection level 2, put processes with a privacy protection level lower than 2 into the subgraph G ₁ , and put the remaining processes into the subgraph G ₂ .

Specifically, the processes with a privacy protection level lower than 2 are put into the subgraph G ₁ , and the remaining processes are put into the subgraph G ₂ . Then there is process ₁ in subgraph G1, and the privacy protection level is 1, respectively; there are processes 2, 3, and 4 in subgraph G2, and the privacy protection level is 2, ₂ , and 2 respectively. Put data centers with a privacy protection level lower than 2 into set S ₁ , and put the remaining data centers into set S ₂ . The data collection _{center. 1} S a; S set in the data center ₂ b. Calculate the number of available nodes in the data center in the set S ₁ :

I(s ₁ )=I(a)=2

The number of processes in the sub-graph G ₁ is 1, which is less than the number of available nodes in the data center in the set S _1. Therefore, it is necessary to select a part of the processes in the sub-graph G ₂ into the sub-graph G ₁ .

Step 8. Calculate the total communication volume between each process in the sub-graph G _{2 and} all processes in the sub-graph G ₁ .

Specifically, calculate the total communication volume between process 2 and process 1:

q ₂ = C _G (1,2)+C _G (2,1)=2+2=4

Calculate the total communication volume between process 3 and process 1:

q ₃ =0

Calculate the total communication volume between process 4 and process 1:

q ₄ = 0

Select the process 2 with the largest total communication volume and put it into the subgraph G ₁ , and set the privacy protection level of the process 2 to the highest privacy protection level among all the processes in the subgraph G ₁ , that is, level 1. Comparative subgraph of G having ₁ Process ₁ in the set S of nodes of the data center is available: the number of sub-G in FIG. 2 is _a process, set S ₁ is equal to the data center available nodes. Therefore, the process in the sub-graph G ₂ is no longer selected and placed in the sub-graph G ₁ .

Step 9. Put the subgraph G ₂ into the subgraphlist subgraphlist, mark the data center b in the set S ₂ as its corresponding data center, and set the subgraph G ₁ and the subgraph G _{2 to} be empty.

In particular embodiments, _a G sub-processes in FIG. 1,2, 1,1, respectively privacy level, privacy level number is equal to 1, then there is no need to continue _a G sub picture divided into _a sub-G in FIG. In the subgraphlist subgraphlist, mark the data center a in the set S ₁ as its corresponding data center, and set the subgraph G ₁ and the subgraph G _{2 to} be empty. End the algorithm flow.

Finally, through the above method, a series of subgraphs and corresponding data center sets can be obtained, and the corresponding data center sets can meet the privacy protection requirements of all processes in the corresponding subgraphs (that is, the privacy protection level of the data center set is equal to or lower The privacy protection level of the process). The resulting subgraph and the corresponding data center set are shown in Table 6.

Table 6 Process mapping results

Technical effect of this embodiment

In the embodiment of the present invention, the above method is tested on two cloud platforms of Amazon EC2 and Windows Azure, and the methods including BT (Block Tri-diagonal solver algorithm for solving block tri-diagonal equations) and SP (Scalar Penta-diagonal Solver scalar five-diagonal equations solving algorithm), LU (Lower-upper Gauss-Seidel solver bottom-up Gauss-Seidel iterative algorithm), K-means clustering algorithm, and DNN (deep neutral network deep neural network) The five applications used our proposed method to test the performance of the algorithm. Experiments show that our proposed algorithm can achieve an average performance improvement of 40% compared to the current best process mapping algorithm.

Example 3

Corresponding to the process mapping method with privacy constraints proposed in the above embodiments, this embodiment also provides a geographically distributed process mapping device with privacy constraints. The geographically distributed process mapping device with privacy constraints includes a unit for executing the above-mentioned geographically distributed process mapping method with privacy constraints. The device can be configured in a desktop computer, a tablet computer, a laptop computer, and other terminals. Specifically, the process mapping device includes a setting unit, a first judging unit, a confirming unit, a putting unit, a second judging unit, a first saving unit, and a returning unit.

The setting unit is used to obtain the process graph of the application, and set the process graph of the application as the graph G to be divided.

The first judging unit is used to judge whether the number of privacy levels of the processes in the graph G to be divided is greater than one.

The confirmation unit is configured to, if the number of privacy levels of the processes in the graph G to be divided is greater than 1, use the largest privacy level in the processes of the graph G to be divided as the characteristic privacy level _Im .

Into the unit to be divided for the graph G is less than the privacy level I _m wherein the privacy level to the process into the sub-G in FIG. _1, the graph G to be segmented privacy level is equal to the privacy feature level I _m to the process into the sub-G in FIG. _2, the application of the privacy level is less than I _m privacy level characteristic data center put into the set S _1, the application level is equal to the privacy privacy level I _m wherein the data collection center was placed in S _2.

The second judging unit is used to judge whether the number of processes in the subgraph G ₁ is less than the number of available nodes in the set S ₁ ;

A first storage unit configured to, if the number of ₁ in the process S is not less than the set of available nodes in _a sub-graph G, the map G ₂ as a sub-divided into sub FIG saved for Store the list of divided subgraphs;

The returning unit is configured to set the subgraph G ₁ as a new to-be-segmented graph G, and return to the step of judging whether the number of privacy levels of the processes in the to-be-segmented graph G is greater than one.

In an embodiment, the process mapping device further includes a second saving unit.

The second saving unit is configured to, if the number of privacy levels of the processes in the to-be-segmented graph G is equal to 1, save the to-be-segmented graph G in a list for storing the divided sub-graphs.

In an embodiment, the process mapping device further includes an acquiring unit and a moving unit.

The obtaining unit is configured to obtain information about each process in the sub-graph G ₂ and all processes in the sub-graph G ₁ if the number of processes in the sub-graph G ₁ is less than the number of available nodes in the set S1 Total traffic

A mobile unit for the sub-graph G _{2 is} moved to the maximum total traffic amount of the process all the processes of the subgraph of G ₁ to G in FIG. ₁ of the sub, the privacy level provided to the process subgraph of G ₁ highest privacy level in all processes, and returns the determination of the number of sub-G in _{FIG. 1} whether the number of processes in the set S ₁ in step is less than the available nodes.

In an embodiment, the process mapping device further includes a first blanking unit.

The first blanking unit is used to blank the sub-graph G ₂ .

In an embodiment, the process mapping device further includes a second blanking unit.

Second blanking means for the sub-G ₁ blanking FIG.

It should be noted that those skilled in the art can clearly understand that the specific implementation process of the above process mapping device and each unit can refer to the corresponding description in the foregoing method embodiment. For the convenience and brevity of the description, it will not be omitted here. Repeat.

The above geographically distributed process mapping device with privacy constraints may be implemented in the form of a computer program, and the computer program may run on the terminal as shown in FIG. 2.

Example 4

Please refer to FIG. 2, which is a schematic block diagram of a terminal 300 according to another embodiment of the present invention. As shown in the figure, the terminal 300 in this embodiment may include: one or more processors 301; one or more input devices 302, one or more output devices 303, and a memory 304. The aforementioned processor 301, input device 302, output device 303, and memory 304 are connected via a bus 305. The memory 302 is used to store instructions, and the processor 301 is used to execute instructions stored in the memory 302. Among them, the processor 301 is used to execute:

In an embodiment, the processor 301 is further configured to execute:

Leave the subgraph G ₂ blank.

In an embodiment, the processor 301 is further configured to execute:

Leave the subgraph G ₁ blank.

It should be understood that, in the embodiment of the present application, the processor 301 may be a central processing unit (Central Processing Unit, CPU), and the processor 301 may also be other general-purpose processors, digital signal processors (Digital Signal Processors, DSPs), Application Specific Integrated Circuit (ASIC), Field-Programmable Gate Array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor.

Those of ordinary skill in the art can understand that all or part of the processes in the methods of the foregoing embodiments can be implemented by computer programs instructing relevant hardware. The computer program may be stored in a storage medium, and the storage medium is a computer-readable storage medium. The computer program is executed by at least one processor in the computer system to implement the process steps of the foregoing method embodiment.

Example 5

The invention also provides a storage medium. The storage medium may be a computer-readable storage medium. The storage medium stores a computer program. When the computer program is executed by the processor, the processor executes the following steps:

In an embodiment, the processor further implements the following steps when executing the computer program:

Leave the subgraph G ₂ blank.

Leave the subgraph G ₁ blank.

The storage medium may be a U disk, a mobile hard disk, a read-only memory (Read-Only Memory, ROM), a magnetic disk or an optical disk, and other computer-readable storage media that can store program codes.

A person of ordinary skill in the art may realize that the units and algorithm steps of the examples described in the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of the two, in order to clearly illustrate the hardware and software Interchangeability. In the above description, the composition and steps of each example have been generally described in terms of function. Whether these functions are executed by hardware or software depends on the specific application and design constraint conditions of the technical solution. Professionals and technicians can use different methods for each specific application to implement the described functions, but such implementation should not be considered as going beyond the scope of the present invention.

In the several embodiments provided by the present invention, it should be understood that the disclosed device and method may be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of each unit is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be omitted or not implemented.

The steps in the method of the embodiment of the present invention can be adjusted, merged, and deleted in order according to actual needs. The units in the device of the embodiment of the present invention can be combined, divided, and deleted according to actual needs. In addition, the functional units in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units may be integrated into one unit.

If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the existing technology, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium. It includes several instructions to make a terminal (which may be a personal computer, a terminal, or a network device, etc.) execute all or part of the steps of the method described in each embodiment of the present invention.

In the above-mentioned embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in an embodiment, reference may be made to related descriptions of other embodiments.

Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. In this way, even if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these modifications and variations.

The above are only specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily think of various equivalents within the technical scope disclosed in the present invention. Modifications or replacements, these modifications or replacements should be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

The geographically distributed process mapping method with privacy constraints is characterized by including:

Obtain the process map of the application, and set the process map of the application as the to-be-segmented map G;

Determine whether the number of privacy levels of the processes in the to-be-segmented graph G is greater than 1;

If the number of privacy levels of the processes in the graph G to be divided is greater than 1, the largest privacy level in the processes in the graph G to be divided is used as the characteristic privacy level Im ;

The graph G to be segmented privacy level is less than the process wherein the privacy level I m is put into the sub-G in FIG. 1, the graph G to be segmented privacy level is equal to the privacy level I m wherein the discharge process into the sub-G in FIG. 2, the application of the privacy level is less than I m privacy level characteristic data center put into the set S 1, the privacy level is equal to the application wherein the privacy level I m The data center is placed in set S 2 ;

Judging whether the number of processes in the subgraph G 1 is less than the number of available nodes in the set S 1 ;

If the number of processes in the sub-graph G 1 is not less than the number of available nodes in the set S 1 , the sub-graph G 2 is saved as a divided sub-graph in the storage for storing the divided sub-graph List;

Set the sub-graph G 1 as a new graph G to be segmented, and return to the step of judging whether the number of privacy levels of processes in the graph G to be segmented is greater than one.
The method of claim 1, wherein the method further comprises:

If the number of privacy levels of the processes in the to-be-segmented graph G is equal to 1, the to-be-segmented graph G is saved in a list for storing divided sub-graphs.
The method of claim 1, wherein the method further comprises:

If the number of processes in the subgraph of G 1 is less than the set number of available nodes S1 to obtain the sub-processes each of G 2 and G all the processes of FIG. 1 of the sub-total traffic;

The subgraph of G 2 is moved to the maximum in the course of total traffic of all the processes the subgraph of G 1 to the subgraph of G 1, provided the privacy level of the sub-process is a G in FIG. all processes highest privacy level, and returns the determination of the number of sub-G in FIG. 1 whether the number of processes in the set S 1 in step is less than the available nodes.
After the method according to claim 1, wherein, in the said map G 2 as a sub-divided into a sub-list is saved into the storage sub-graph partitioning, the method further comprising:

Leave the subgraph G 2 blank.
The method according to claim 1, characterized in that, after the setting the sub-graph G 1 as a new to-be-divided graph G, the method further comprises:

Leave the subgraph G 1 blank.
The geographically distributed process mapping device with privacy constraints is characterized by comprising a unit for executing the method according to any one of claims 1-5.
A terminal, wherein the terminal includes a memory and a processor, and a computer program is stored on the memory, and the processor implements the computer program according to any one of claims 1-5 when the computer program is executed. method.
A computer-readable storage medium, wherein the storage medium stores a computer program, which when executed by a processor can realize the method according to any one of claims 1-5.