ES2330604A1 - Method for managing computation resources which is applicable to a software-defined radio system - Google Patents

Abstract

Method to manage computing resources applicable to a radio system defined by program.

It includes mapping functions, processes or parts of at least one application on processors with certain computing resources, which are selected using cost functions through a series of stages in which a pre-mapping of each of M functions is performed. ( f 1 -f M) of said application to each of N processors ( P 1 -P N), thus obtaining a matrix of NxM candidate nodes, said mapping being a mapping dynamic that is apt to use said plurality of cost functions ( f 1 -f M) and radio scenarios, traveling said matrix through a series of candidate paths, dynamically managing the computing resources available for meet and optimize the requirements of program-defined radio computing through a selection of nodes connected by a path that provides an optimal cost.

Description

Method to manage computing resources applicable to a radio system defined by program.

Technical sector

The present invention concerns a method for manage computing resources applicable to a radio system defined by program, which generally includes mapping functions, processes or parts of an application about processors with certain computing resources, selected using cost functions, and in particular at a method comprising performing a pre-mapping of said functions on processors, constituting nodes, each of them associated with a function and a processor, which form an array of nodes pre-connected by means of a few candidate paths, traverse all the nodes of said matrix to evaluate your costs, make a pre-selection of roads and candidate nodes, and finally make a selection or mapping final based on the different costs accrued and the preselection done.

Prior art

Program-defined radio (Software-defined radio (SDR)) can be considered as a generalization of radio by program, or software, because it characterizes a transceiver that implements one or more signal processing blocks by software. SDR is an emerging concept that influences the design of software-defined and independent signal processing chains in relation to hardware for radio communications. It introduces flexibility in wireless systems, facilitating the dynamic passage from one radio access technology (RAT) to another or, in other words, the reallocation and reallocation of computing resources of an SDR application to
other.

The context of SDR computing consists mainly on heterogeneous multiprocessor platforms, whether base stations (BSs) or mobile terminals (MTs), as well as chains of signal processing defined by software. The terminals mobile phones are very limited in terms of energy resources and of computing, in flexibility, and in support for RAT implementations concurrent On the contrary, the computing resources of base stations are less limited, their energy consumption is not a limitation, and the number of concurrent RAT implementations It can be as tall as desired. However, the potentially large number of users and high degrees of flexibility, modularity and reconfigurability of the platforms make the Resource management at base stations is equally important but more complex than in mobile terminals.

There is a need to define a structure of flexible software that interfaces between resources SDR computing, on the one hand, and the wireless system, represented by SDR computing requirements, on the other side. The structure must be able, by applying a corresponding method, to map dynamically and in a way Efficient SDR applications with precedence restrictions, over SDR platforms while satisfying the restrictions of SDR computing These restrictions are, first of all, the real-time computing requirements of applications SDR, defined by the demands regarding minimum bit rate and maximum latency, and the limited computing resources of SDR platforms

There are many contributions regarding mapping and programming of tasks on computer platforms heterogeneous multiprocessor. These contributions are focused to a wide variety of problems in computing contexts of general or special purpose Some jointly address  mapping problems and task scheduling and present optimal or less than optimal solutions with different objectives, such as minimizing the programming length of tasks or communication header, the objective of satisfying Real-time response deadlines, as well as other objectives Additional or alternative.

By US6768901 it is known to manage computing resources of a radio system defined by program, mapping some functions, processes or parts of a application on processors with computing resources determined, selected using cost functions. Be proposes in this document to make an abstraction of resources computational in order to use a language of capabilities common, both in terms of available resources and to those required. A specific type of abstraction is not specified, downplaying, in fact, the kind of abstraction to use.

On the other hand by the article "Modeling for the management of computing resources in SDR systems ", of XXI National Symposium of the International Scientific Radio Union, of the same inventors of the present invention, a model of an SDR system based on an abstraction that has in account for limited computing resources.

The application in SDR systems is also known of voracious mapping, or "greedy- or g-mapping", the what is a local mapping approach that does not require any postprocessed This is based on selecting from a grid or matrix of candidate nodes, each associated with a function and a processor, based on the results of a cost function applied to each or of the nodes. The algorithm g-mapping first select a node from the first column of said matrix, after which it takes into account the cumulative cost in each of the nodes of the second column taking as origin the selected node of the first column, in base at whose cost you select a node from the second column, and so successively for the rest of the columns, so that at each step only one node is selected, which is the only one that remains active when performing the next step of selecting a node of the next column.

It seems necessary to offer an alternative to state of the art that allows, for the same array of nodes candidates to map, get better results than the obtained with the g-mapping, thanks to consider for the selection of each node, many more possible combinations of nodes that are taken into account by g-mapping, as well as the fact of making a series of presets prior to The final selection of nodes.

Explanation of the invention.

To achieve this purpose the present invention concerns a method to manage computing resources applicable to a program-defined radio system, of the type that comprises mapping functions, processes or parts of at least one application on processors with computing resources determined, selected using cost functions by performing the following stages:

a) pre-map each of M functions of said application to each of N processors, thus obtaining a matrix of NxM candidate nodes, each of them  associated with a function and a processor,

b) pre-connect each of the candidate nodes belonging to a column of said matrix with all candidate nodes in the next column, enabling each of said connections a data communication through a candidate path that includes at least one origin candidate node, one of those connections and a destination candidate node,

c) apply at least one cost function to each one of the N candidate nodes of the first column of said matrix, or nodes associated with a first of these functions, obtaining the cost of each of said candidate nodes,

d) apply at least one cost function to the candidate paths that include the candidate nodes of the first column in connection with at least those of the second, and calculate the cumulative cost associated with each of these roads candidates that includes at least the cost of the destination node depending on of previous decisions and the cost of the origin node obtained in the stage c), in whole or in part, thus obtaining the cost of each of said candidate paths, and:

e) preselect for each candidate node of the second column at least part of the candidate path of optimal cost which leads to said candidate node of the second column or passes through he, obtaining, respectively, N first paths or segments of preselected roads, discarding the connections that unite the candidate nodes of the first column with those of the second not they are part of said first roads or segments of roads preselected

For a first embodiment, which will be referenced in a later section as indicative of the use of a calculation window w = 1, for which each of said candidate paths is a two-node path (or path of length 1) which includes only one of said connections, a node origin candidate and a two-column destination candidate node Consecutive, the method includes carrying out:

- said step d) to apply said function of cost, which is at least one, to the candidate paths that include the candidate nodes of the first column in connection with, only, the candidate nodes of the second column, and

- said step e) to pre-select entirely, for each candidate candidate node of the second column, the path candidate of optimal cost that leads to him, obtaining said N first shortlisted paths that start from a node of the first column and end in a node of the second column.

For said first embodiment if M = 2, the method comprises selecting the two nodes that are part of the optimal cost path, from among these first paths preselected, and therefore the processor or associated processors to each of said two nodes selected as processor or mapped processors to execute the first function, the associate to the node of the first column, and a second function the one associated with node of the second column.

Instead for a more real case where M \ geq3, the method comprises, for said first embodiment, perform the following stages, analogous to these stages d) and e), respectively:

d ') apply at least one cost function to the candidate paths that include the candidate nodes of the second column in connection with those of the third, to calculate the accumulated cost associated with each of these candidate paths that  includes at least the cost of the destination node of the third column in function of previous decisions and the cost of one of said first pre-selected roads in stage e), each of the which includes only one node of the first connected column with one node of the second column, thus obtaining the cost of each one of said candidate paths destined for a respective node third column candidate, and

e ') preselect for each candidate node destination of the third column the candidate path of optimal cost leading to it from a source node of the second column, obtaining N seconds preselected paths, discarding the connections that link the candidate nodes of the second column with those of the third that are not part of these second paths preselected

According to the first embodiment when M = 3, the method includes selecting, from among said N seconds preselected roads, the optimum cost, and travel it towards back along with the first preselected path to the one found  joined, to select the three nodes that are part of said route, and therefore the processor or processors associated with each one of said three nodes selected as mapped processors to execute the first function, the one associated with the node of the first column, the second function, the one associated with the second node column, and a third function, the one associated with the node of the third column.

If instead M> 3, the method comprises, for the first example of realization, perform, for each of the columns following the third column, two by two and moving forward column by column, stages analogous to said stages d ') and e') to finally preselect for each destination candidate node of each additional column to the third column, the candidate path of optimal cost that leads to it from a column origin node immediately above, obtaining N preselected paths additional per column, discarding connections that do not form part of said additional preselected roads.

Finally for that case where M> 3, the method comprises, for the first embodiment, select from among the N additional preselected paths which include the nodes of the last column, the optimum cost, and  traverse it back along with all the preselected roads to which it is attached, to select the M nodes that they are part of that route, and therefore the processor or processors associated with each of said M selected nodes as mapped processors to execute the M functions.

For a second embodiment, which will be referenced in a later section as indicative of the use of a calculation window w = 2, for which each of said candidate paths includes an origin candidate node, a node intermediate candidate and a three-column destination candidate node consecutive respectively interconnected by two of said connections, that is a three-node path, the method comprises carry out:

- said step d) to apply at least said cost function to the candidate roads, each of which includes, interconnected by two respective connections, a node Candidate origin of the first column, a candidate node intermediate of the second column and a candidate node destination of the third column, to calculate the cumulative cost of each of said candidate paths of three nodes and that of their segments initials, or path segments of two nodes that include a node of the first column and one of the second, and

- said step e) to preselect, for each intermediate candidate node of the second column, only part of the candidate path of three nodes of optimal cost that passes through it, said pre-selected parts being said first N segments Two way path.

For said second embodiment, when M> 3, the method comprises performing the next stage, analogous to said stage d):

d '') apply at least one cost function to the candidate paths of three nodes that include, interconnected by two respective connections, a candidate node origin of the second column, an intermediate candidate node of the third column and a destination candidate node of the fourth column, to calculate the cost accumulated associated to each of said three candidate paths nodes, which includes at least the cost of the destination node of the fourth column based on previous decisions and cumulative cost of one of said first two node path segments preselected in stage e).

Specifically when M = 4, the method comprises, for said second embodiment, perform the following stages, after said stage d ''):

- preselect for each candidate node intermediate of the third column, the complete candidate path of three nodes of optimal cost that passes through it and ends in a node of the fourth column, obtaining N paths of three nodes preselected, and

- select, from among said N paths of three preselected nodes, the optimal cost one, and travel that path selected from its intermediate node, from the third column, towards forward and backward along with the first segment of two road nodes preselected in stage e) to which it is attached, for select the four nodes that are part of said path, and therefore the processor or processors associated with each of said four nodes selected as mapped processors for execute the first function, the one associated with the node of the first column, the second function, the one associated with the second node column, the third function, the one associated with the third node column, and a fourth function, the one associated with the node of the fourth column.

Alternatively when M> 4, the method comprises, for said second embodiment, perform the following stage after said stage d ''):

e '') preselect, for each candidate node intermediate of the third column, only part of the way candidate of three nodes of optimal cost that passes through it and ends in a candidate node of the fourth column, obtaining N seconds Road segments of two preselected nodes.

The method comprises performing, for that case in that M> 4 of the second embodiment, for each of the columns following the fourth column, three to three and moving forward column by column, some stages analogous to those stages d '') and e '') for each additional three-node path that ends in each of the following or additional columns to the fourth column, to exception of the additional three-node path that ends in the column M, to preselect for each intermediate candidate node of each additional three-node path, only part of the path candidate of three nodes of optimal cost that passes through it, obtaining N path segments of two additional preselected nodes per each additional column to the fourth column.

As regards these three ways additional nodes ending in column M, the method includes performing the following steps (for M> 4 of the second exemplary embodiment), after said stage analogous to said stage and''):

- preselect for each candidate node intermediate of column M-1, the candidate path complete of three nodes of optimal cost that passes through it and ends in  a node of column M, obtaining N paths of three nodes preselected, and

- select, from among said N paths of three preselected nodes, the optimal cost one, and travel that path selected from its intermediate node, from the column M-1, forward and back along with all path segments of two preselected nodes to which find attached, to select the M nodes that are part of said path, and therefore the processor or associated processors to each of said M nodes selected as processors mapped to execute the M functions.

For some additional embodiments for which w> 2, or in other words when each of said candidate paths is a path of more than three nodes, the method It includes carrying out:

- said step d) to calculate the cost accumulated of each of said candidate paths of more than three nodes and the one of its initial path segments of two nodes that include a node of the first column and one of the second,

- said step e) to preselect, for each candidate node of the second column, only part of the path candidate of more than three nodes of optimal cost that passes through it, said pre-selected parts being said first N segments two-way path, and

- perform stages analogous to these stages described for said three-node paths, preselecting to each candidate node that occupies the second position on each path of plus three additional nodes ending in each of the columns following or additional to the fourth column, except for the path of more than three additional nodes ending in column M, the initial segment of two nodes of the candidate path of more than three nodes of optimal cost that passes through it, obtaining N segments of path of two additional preselected nodes per column additional to the fourth column.

The method comprises performing, for said paths of more than three additional nodes that end in the column M, the following stages:

- preselect for each candidate node that occupies the second position on each path of three more nodes additional ending in column M, the candidate path complete with more than three nodes of optimal cost that passes through it and ends in a node of column M, obtaining N paths of more three preselected nodes, and

- select, from among said N paths of more of three preselected nodes, the optimum cost, and traverse said path selected from its respective node that occupies the second position on the road, forward and backward together with all the path segments of two preselected nodes to those that are united, to select the M nodes that form part of said route, and therefore the processor or processors associated to each of said M nodes selected as mapped processors to execute the M functions.

As regards the aforementioned previous decisions, these influence said accumulated cost by less as regards the remaining resources in each node after having made those decisions or preselections.

The cost functions used by the method proposed by the invention evaluate the cost of computing that relates processing requirements to resources processing available, and the cost of communication puts that relates bandwidth requirements to transfer data from one process to another with the corresponding bandwidths available between processors.

The method comprises, for an example of realization, determine said computing resources of each of said processors based on a temporary segment, or "time slot", of a plurality of equal time segments,  of sufficient duration to ensure that each processor is capable of executing the assigned functions or processes and of transferring data to another processor before said segment ends temporary, thus guaranteeing the real-time processing required by SDR applications.

On the other hand in order to work with such temporal segments, the method comprises mapping those functions to finally run the application through a process of instruction segmentation, establishing them computing requirements for each of these segments Temporary

         \ vskip1.000000 \ baselineskip
      

Brief description of the drawings

The above and other advantages and features will be more fully understood from the following detailed description of some embodiments with reference to the attached drawings, which should be taken as a title illustrative and not limiting, in which:

Fig. 1 is a schematic representation of a platform model with its initially available resources (a) and an application model with its resource requirements (b), usable to apply the method proposed herein invention;

Fig. 2 schematically shows a node matrix (a, c) configured according to the proposed method, with initial nodes drawn with continuous line object of analysis according to a first part of the proposed method, for a exemplary embodiment, as well as a table of costs (b) of said nodes;

Fig. 3 shows the same matrix (a, c) of the Fig. 2, for the same embodiment, but for one step posterior, or second part of the method proposed by the invention, indicating the view (a) the paths examined in a continuous line and the view (c) the preselected path in bold, as well as a table of costs (b) indicating the cost accumulated in a first node of a second column of said matrix;

Fig. 4 shows views similar to those of Fig. 3, for the same second step, but applied to a second node of said second column of the matrix, indicating the view (a) the roads examined in a continuous line and sight (c) the road preselected in bold;

Fig. 5 shows views similar to those of Fig. 3, with the difference that the analysis referring to said second part of the method proposed by the invention, takes out with respect to the first node of the third column;

Fig. 6 shows views similar to those of Fig. 4, with the difference that the analysis referring to said second part of the method proposed by the invention, takes out with respect to the second node of the third column;

Fig. 7 shows views analogous to those of Fig. 3 and 5, with the difference that the analysis referring to said second part of the method proposed by the invention is carried out with respect to the first node of the fourth column;

Fig. 8 shows views analogous to those of Fig. 4 and 6, with the difference that the analysis referring to said second part of the method proposed by the invention is carried out with respect to the second node of the fourth column, and that in (c) no line appears in bold because there is no preselected no path leading to the second node of the fourth column due to the infinite cost of both;

Fig. 9 shows in view (a) the matrix of nodes of Figs. 2 to 7, with selected nodes definitely that they are part of the route indicated by some dashed arrows, as well as a view (b) indicating the mapping final represented by said selected nodes;

Fig. 10 shows views similar to those of Fig. 3 but for another exemplary embodiment for which examine paths of three nodes that pass through the gray node of the view (a), indicating said view (a) the roads examined online continue and view (c) the segment of one of said paths preselected, in bold, as well as a cost table (b) indicating the costs accrued in each of these roads examined;

Fig. 11 shows views analogous to those of Fig. 10, for the same second step, but applied to the roads that pass through the gray node of the view (a);

Fig. 12 shows views similar to those of Fig. 10, but where the nodes under analysis have been shifted one position to the right, and where in view (c) it has preselected, marked in bold, a full path of three nodes covering the last three columns;

Fig. 13 shows similar views to those of Fig. 12, but where both the paths of three nodes examined (view (a)) as the preselected (view (c)) go through the second node of the third column; Y

Fig. 14 shows in view (a) the matrix of nodes with selected nodes definitely that are part of the route indicated by dashed arrows, for exemplary embodiment of Figs. 10 to 13, as well as a view (b) indicating the final mapping represented by said nodes selected;

Fig. 15 is a schematic representation of an array of NxM nodes generated and used by the method proposed by the present invention;

Fig. 16 illustrates a sequence of instructions or pseudo-code representative of the implementation of the method proposed by the present invention to an example of embodiment; Y

Fig. 17 is a table containing the most important variables and expressions used in the following detailed description section of some examples of realization.

         \ vskip1.000000 \ baselineskip
      

Detailed description of some embodiments

This section will refer, with reference to the attached figures, the method proposed as " t w-mapping" algorithm or dynamic window programming algorithm w, where w indicates the size of said window, for analysis and calculation .

Although a particular problem of mapping, the concepts shown are valid for any trouble.

To exemplify the mapping process according to the t w-mapping algorithm, a simple cost function has been assumed. This consists of the sum between the cost of computing and the cost of communication. The cost of computing relates the processing requirements to the available processing resources, while the communication cost relates the bandwidth requirements (to transfer data from one process to another) with the corresponding available bandwidths between the processors. Internal bandwidths to a processor are assumed infinite, so the cost of transferring data from one process to another within the same processing does not imply any communication cost. To ensure proper mapping, the available resources are updated after each decision of the t- w-mapping. The numbers (costs) come from this cost function, which is not specified in more detail because the objective is to exemplify the operation of the t- w-mapping; The origin of these numbers is not significant but how the algorithm behaves based on these numbers. Note that an infinite cost indicates the case of requiring more resources than are available. Therefore finite costs represent valid mappings, while infinite costs indicate invalid (partial) mappings. For this, as for any cost function applied, the objective is to optimize (minimize in this case) the cost of mapping; The algorithm makes the corresponding decisions, which will be seen below.

As regards the platform and application (Resources available and required), in Fig. 1a shows the model of a platform of 2 processors, indicating processing resources (bold numbers) and communication (normal numbers), and in Fig. 1b the resources are shown required processing (bold numbers) and data transfer (normal numbers) and the application model consisting of 4 processes (modules or functions).

         \ vskip1.000000 \ baselineskip
      

Operation of t w-step-by-step mapping

Window w = 1

A) Part I of the algorithm, or in other words first part of the proposed method

In the first part, the algorithm calculates the cost of mapping the process or function f 1 to all processors (Fig. 2a). Depending on the cost function chosen, mapping the process f 1 to the processor P 1 costs 0.5, while mapping the process f 1 to the processor P 2 costs 1 (Fig. 2b). These costs are saved in the corresponding nodes (Fig. 2c).

         \ vskip1.000000 \ baselineskip
      

B) Part II of the algorithm

In the second part, the t _ {1} -mapping examines nodes between step (step) 2 and (M - w +1). The nodes of each step can be processed in parallel, while step ( i +1) must be processed after the processing of the nodes of step i . For the node ( P 1, f 2), the paths [ P 1 P 1] 2 and [ P 2 P 1] _ are examined {2} (Fig. 3a). Its cumulative costs, that is the cost of the origin node plus the cost of mapping f 2 to P 1 and P 2, respectively, are calculated based on the remaining resources (Fig. 3b) for choose the one with the lowest cost that marks the way forward (Fig. 3c).

For the node ( P 2, f 2), the paths [ P 1 P 2] 2 and [ P 2 P 2] _ are examined {2} (Fig. 4a). Its accumulated costs, that is the cost of the origin node plus the cost of mapping f 2 to P 1 and P 2, respectively, are calculated based on the remaining resources (Fig. 3b) to choose the one with the lowest cost that marks the way forward (Fig. 4c).

Once you have finished processing the nodes of step 2, you continue with the nodes of step 3. These nodes are processed in the same way as was applied to the nodes of step 2, based on the costs saved and the resources available in nodes (P {1}, {2} f) and (P {2}, {2} f) (Fig. 5 and Fig. 6). Finally, the nodes of step 4 are processed (Fig. 7 and
Fig. 8).

C) Part III of the algorithm

Once part II of the algorithm is finished, part III is continued. In this part, t w-mapping first searches for the lowest cost node under those of step M - w +1. In the case of t _ {1} -mapping and the example, this is the node (P {1}, {4} f), with the cost of 2.83, opposed to the infinite cost node (P { 2, f 4) (Fig. 9a). Starting from this node diagram is transferred t _ {w} -mapping by boldfaced connections to reach a node in step 1, marking nodes crossed (Fig. 9a). This returns the final mapping according to the cost function applied and the window chosen. In this case the solution proposes to map f 1, f 2 and f 4 to processor P 1 and f 3 to P 2 (Fig. 9b). The total cost of this mapping is 2.83, as indicated by the node (P {1}, {4} f) (Fig. 9a).

         \ vskip1.000000 \ baselineskip
      

Window w = 2

A) Part I of the algorithm

It is the same as in the case of w = 1 (Fig. 2).

         \ vskip1.000000 \ baselineskip
      

B) Part II of the algorithm

The paths have a length of w = 2. There are N w = 2 2 = 4 paths that are compared on each node of part II of the algorithm. FIG. 10 illustrates the method for the node (P {1}, {2} f) and FIG. 11 for the node (P {2}, {2} f). Note that, although the cost of the path of length 2 is used to make the decision, only the first part of the path is marked and the cost corresponding to the passage of length 1 is saved (Fig. 10c and Fig. 11c).

Proceed with the 2 nodes under step 3 (see Figs. 12 and 13) in an equivalent manner, with the difference that the cost accumulated in the nodes of the fourth column is saved or assigned to the nodes of the third column and of that the entire path of length 2 is marked after the decision has been made. This is because it is the last step , part II t _ {2} -mapping: M - w +1 = 4-2 + 1 = 3.

         \ vskip1.000000 \ baselineskip
      

C) Part III of the algorithm

Minimum cost node under step (M - w + 1) is chosen = (4/2 + 1) = 3, in this case the node (P {2}, {3} f). From this diagram node t _ {w} -mapping is traversed in both directions by the connections marked until step 1 on one side and the step 4 on the other (Fig. 14a). The final mapping is obtained from the traversed nodes (Fig. 14b) during this procedure.

Because SDR systems are conceived different optimization criteria, such as the one related to meet real-time computing requirements in difficult conditions to satisfy, or relative to optimize the energy consumption, the method or algorithm proposed herein invention is general purpose, and can execute different cost functions and be dynamically adjusted to changes in the Radio signal environment.

It will be detailed below, with reference to Figs. 15 and 16, the t w-mapping algorithm described for a more complex embodiment than described above, and generalized for M functions, N processors and NxM nodes.

With reference to Fig. 15, there can be seen a t w-mapping diagram containing a matrix of NxM t-nodes. A t-node is identified as ( P k (l), f i) and represents the mapping of an SDR function f i to the processor P k (l). Any t-node in step i (column i in the diagram {w} t _ -mapping) is connected to all nodes t-step i +1. The sequence of processors [ P k (0) P k (1) ... P k (w)] i identifies the path-w, a path of length w , which is associated with the t-node ( P k (1), f i). P_ {k (0)} is the source processor of the path-w in step i -1 and P k (w)} the destination processor in step i + w -1. Thus, [ P 1 P 1 P N P 1] 2 represents the path of three connections or segments marked in bold in Fig. 6, and is associated with t-node (P {1}, {2} f). Table 1 in Fig. 17 contains the most important variables and expressions that appear in this specification.

The t w-mapping algorithm illustrated in Fig. 16 can be divided into three parts: processed in step 1 (lines 1-2 in Fig. 16), processed in steps i = 2, 3, ..., M - w +1 (lines 3-18 in Fig. 16), and postprocessed (lines 19-20 in Fig. 16). The t w-mapping and therefore its description does not assume a particular cost function.

         \ vskip1.000000 \ baselineskip
      

A) Part I: processed in step 1

The first part of the algorithm concerns the function SDR f {1}. For any processor P k (0), the t w-mapping ( w ≥ 1) pre-maps f 1 to P k (0) and stores the pre-mapping cost CT ( P k (0), f 1) at the t-node ( P k (0), f 1) (lines 1-2 in Fig. 16). The term pre-mapping indicates that the final mapping of the SDR application is not known until all its SDR functions have been processed, or pre-mapped.

B) Part II: processed in steps i = 2, 3, ..., M - w +1

The t 1 -mapping analyzes the N path segments, or paths of length 1, originating from the t-node ( P k (1), f i). These are [ P 1 P k (1)] i, [ P 2 P k (1)]] i, ..., and [ P _ {N} P k (1)] i (line 5 in Fig. 16). By way of 1 [P {k (0)} P {k (1)}] _ {i} is assigned the weight WT [P {k (0)} P {k (1)}] i (line 6 in Fig. 16). This weight represents the cost of premapear function SDR f {i} the processor P {k (1)} taking into account the previous decisions, which are provided by the t-source node of the path 1 (P { k (0), f i-1). The t 1 -mapping calculates the cumulative cost CT [ P k (0) P k (1)] i = CT ( P k (0)}, f i-1) + WT [ P k (0) P k (1)] i (bold elements of line 12 in Fig. 16) for each path of 1. The path of 1 [ P k (0) \ text {*} P k (1)}] is obtained from:

one

and represents the pre-mapping decision in the t-node ( P k (1), f i). (The argmin { x } function results in the argument (s) that lead to the minimum value of x ). The algorithm finally highlights the path of 1 [ P k (0) \ text {*}} P k (1)}] i (line 14, in bold, of Fig. 16) and stores its accumulated cost in the t-node ( P k (1), f i) (line 15).

The t w -mapping ( w > 1) analyzes the N w-paths that are associated with the t- node ( P k (1), f i). Any of these w-paths originates from the t-node in step i -1, runs through ( P k (1), f i), and ends at the t-node in the step i + w -1. The algorithm calculates the accumulated costs due to lines 5-12 of Fig. 16. Then it solves

2

which offers as a result the indices of the w-path that have the minimum accumulated cost. In the case that i < M - w +1, t w -mapping ( w > 1) highlights the path of 1 [ P k (0) \ text {*}} P _ {k ( 1)}] i and stores the corresponding cumulative cost in the t-node ( P k (1), f i) (lines 13-15 in Fig. 16). Otherwise, highlight the w-path [ P _ {k (0) \ text {*}} P _ {k (1)} P _ {k (2) \ text {*}} ... P _ { k (w) \ text {*}}] {M-w + 1} integer and stores its total cumulative cost in the t-node ( P k (1)}, f M-w + 1} ) (lines 16-18 of Fig. 16).

All N t -nodes in step i (line 4 of Fig. 16) can be processed in parallel. Once its processing has been completed, t w -mapping ( w ≥ 1), that is to say the method proposed by the present invention, proceeds with step i +1 in the event that i < M - w +1 and if not, proceed with part III (line 3).

C) Part III: postprocessed

Part III of the postprocessing algorithm the pre-mapping decisions of parts I and II. The t w-mapping ( w ≥ 1) first highlights the t-node in step M - w + 1 that retains the minimum cost (line 19 of Fig. 16). Starting with this t-node, the t w -mapping diagram, that is, the NxM matrix, is traversed backward (or forward and backward in case w > 1) along the paths of 1 highlighted, while highlighting all the t-nodes that are traversed during that route (line 20 of Fig. 16). This results in M t- highlighted nodes, which are finally selected to specify the proposed mapping proposal for the particular problem, the cost function and the window size chosen.

Parts II and III described indicate that the window w controls the level of local (vs. local global scope) decisions premapeado: Algorithm T _ {1} -mapping is based on local decisions. However, it keeps N pre-mapped at each step, one per t-node, which results in N (partially) different mapping options (bold expressions in Fig. 16).

Moreover, the algorithm _ {M t-1} -mapping examines all N {M} possible mappings M functions to N processors. Specifically, calculates the costs arising from different paths - (M -1), ie paths of length (M-1) of all N {M} -caminos- (M -1) crossing the t - node ( P k (1), f 2) (lines 5-12 in Fig. 16) and select the lowest cost (lines 17 and 18 of Fig. 16). This is done for all N t- nodes in step 2 (line 4 of Fig. 16). The algorithm then highlights the t -node ( P k (1) \ text {*}}, f 2) (line 19 of Fig. 16) and travels back and forth the diagram of t w -mapping, that is the NxM matrix of interconnected nodes, to obtain the final mapping (line 20 of Fig. 16). It is thus found through the application of the method proposed by the present invention, an optimal solution for the given problem and the cost function used.

The w parameter also controls the compromise between computing efficiency and mapping performance: the larger the size of the w window, the greater the complexity of the algorithm but the better the mapping results.

Although, as it has been commented above, the method proposed by the present invention does not It is limited to a particular type of cost function, for a embodiment example a cost function is proposed that manages adequately the limited computing resources of the SDR platforms under demanding real-time conditions, where computing resources constrain application mapping SDR

The proposed cost function is WT [ P k (l-1) P k (l)] h and therefore defines the cost function as an overlap between computing costs and costs of communication, as follows:

3

where WT_ {comp} indicates the cost computing, and WT_ {com} the cost of communication.

Any t-node ( P k (l), f i) stores the processing capacities C k (l), \ (i)} and the bandwidths B k (l) , \ (i)} remaining based on the previous pre-mapping decisions. The cost function first calculates the pre-mapping costs in the t-nodes ( P 1, f 1), ( P 2, f 1), ..., and ( P N, f 1) using the right part of (3b). For each t-node ( P k (0), f 1), the processing requirement c 1 of the SDR function f 1 is then subtracted from the corresponding initial processing capacity C k (0). This updates the remaining processing capabilities for all t-nodes in step 1.

Before processing the path of length 1 [ P k (l-1) P k (l)] h, C (k (l), \ h)} and B ^ (k (l), \ h)} are initialized with C (k (l-1), \ h-1)} and B (k (l-1), \ h-1)}. The algorithm then calculates WT [ P k (l-1) P k (l)] h before updating C k (l)} (k (l), \ h )} and B (k (l), \ h)}. In particular, c h is subtracted from C k (l)} {(k (l), \ h)} after calculating WT_ {comp} [ P k (l-1)} P _ {k (l)} h (3b). B ^ {(k (l) \ h)}, on the other hand, are updated dynamically, subtracting any required bandwidth b _ {jh} of the corresponding entry in B ^ {(k (l) \ h)} immediately after adding b jh / {\ cdot} to WT_ {com} [P_ {k (l-1)} P k (l)}] h (3c).

As mentioned above, the method proposed by the present invention, also referred to as the " t w-mapping" algorithm, is applicable for different types of modeling or abstraction of an SDR system, such as that described by each platform based on its computing or processing and communication resources (see Fig. 1a) and that of the SDR application based on the required processing and data transfer resources (see Fig. 1b).

For other embodiments, the parameters to consider to model both the SDR system and the SDR application may be different or complementary to the indicated regarding processing resources and communication.

An example of realization of the proposed method consider using the modeling proposed in the article "Modeling for the management of computing resources in SDR systems ", of XXI National Symposium of the International Scientific Union of Radio, of the same inventors of the present invention.

The data that is transmitted or received to through a wireless link they need to be processed while There is data to transmit or receive. An SDR application will will run in general during the entire user session, although there may be periods of time in which no one is transmitted User data The fact that an SDR application can be replaced by another during a single user session does not affect to this continuous data processing.

Because of this, the method proposed by the The present invention contemplates, for an exemplary embodiment, "break" continuous execution into periodic executions dividing the time of each computing resource into segments equidistant temporary computing, or "time slots", and the SDR application in stages of "pipelining".

The introduction of temporary segments of computing allows to identify the computing capacity of a processor based on these temporary segments, generating thus a modeling analogous to that of the mentioned article, but incorporating the aforementioned segmentation in "time slots".

This provides a basic mechanism for a efficient computing resource management, and it is especially useful to meet the maximum latency demands of SDR applications

The execution by segmentation process of instructions, or "pipelined", of an SDR application set that, in any "time slot", all SDR functions process and propagate some part of the data. That is, the same processed and the same data transfer is repeated for each temporal segment in a different piece of data. It introduces some synchronization requirements that PHAL (The Platform and Hardware Abstraction Layer '' satisfies (see article "Software radios: unifying the reconfiguration process over heterogeneous platforms ", EURASIP Journal on Applied Signal Processing, vol. 2005, no. 16, pp. 2626-2640, Sept. 2005, of X. Reves, A. Gelonch, V. Marojevic, R. Ferrus).

The "Pipelining" also introduces the latency, which must be maintained within the radio service and of the dependency limits of QoS (Quality of service).

From the discussion in the previous paragraphs, for system modeling and SDR applications based on the aforementioned segments temporary, or "time slots", the new units are derived MOPTS (millions of operations per time slot) and MBPTS (mega-bits per "time slot") as t_ {ts} \ cdotMOPS and t_ {ts} \ cdotMbps, where t_ {ts} is the duration of the "time slot". MOPTS and MBPTS synchronize the computing resources available with time-based management slots "and are the basic units for the referred modeling.

It is proposed to define tts according to the following equation:

4

where n_ {ts} are the number of "time slots" and L_ {max} is the maximum allowable latency for the efficient programming of tasks in each processor, which it is a function of the tolerable end-to-end delay of a radio communication link due to QoS and service between the radio service provider and the user final. In the proposed modeling it is assumed that tts is sufficiently great to allow (efficient) execution of any function SDR in the chain indicted.

While the description of the proposed method with reference to Figs. 15, 16 and 17, it has been done in this section by calling it the " t w-mapping" algorithm, and using equations and instructions (which are indicated in Fig. 16), Fig. 15 is also referenced from the description of the proposed method made in the explanation section of the invention, that is to say that the matrix referred to there as a matrix of NxM nodes is, for an exemplary embodiment, the matrix illustrated in Fig. 15, including all the elements mentioned in said description made in the aforementioned section above, such as the nodes of the different columns (P_ {1}, f_ {1} ... P_ {N}, f_ {1}) a (P_ {1}, f_ { M} ... P_ {N}, f_ {M}), as well as the different connections between nodes of two consecutive columns ([P_ {1} P_ {1}] 2} ... [P_ {1}
P_ {N}] 2 ... [P_ {N} P_ {1} 2} ... [P_ {N} P_ {N}] 2) ... ([P_ {1} P_ {1} M {... [P_ {1} P_ {N}] _ {M} ... [P_ {P} {1}] M {... P_ {N} P_ {N}} {M}), which make up what in this section have been called as longitude 1 roads, or road segments when they are part of a road runs along more than two columns

Serve references in parentheses included in all the attached claims as a guide clarification for a better understanding of these claims leaning on Fig. 15.

A subject matter expert could introduce changes and modifications in the described embodiments without departing from the scope of the invention as defined in the attached claims.

Claims (19)

1. Method to manage computing resources applicable to a program-defined radio system, of the type that comprises mapping functions, processes or parts of at least one application on processors with computing resources determined, selected using cost functions by performing the following stages: a) pre-mapping each of M functions (f_ {1} -f_ {M}) of said application to each of N processors (P_ {1} -P_ {N}), thus obtaining an array of NxM candidate nodes ( P k (l), f i), each associated with a function and a processor, b) pre-connect each of the candidate nodes belonging to a column of said matrix with all candidate nodes in the next column, enabling each of said connections a data communication through a candidate path that includes at least one origin candidate node, one of those connections and a destination candidate node, c) apply at least one cost function to each one of the N candidate nodes (P_ {1}, f_ {1} ... P_ {N}, f_ {1}) of the first column of said matrix, or nodes associated with a first (f_ {1}) of these functions, obtaining the cost of each of said candidate nodes (P_ {1}, f_ {1} ... P_ {N}, f_ {1}), d) apply at least one cost function to the candidate paths that include the candidate nodes of the first column (P_ {1}, f_ {1} ... P_ {N}, f_ {1}) in connection with at least those of the second (P 1, f 2) ... f_ {2}), and calculate the cumulative cost associated with each of said candidate paths that includes at least the cost of the node destination based on previous decisions and the cost of the node origin obtained in stage c), in whole or in part, thus obtaining the cost of each of said candidate paths, Y: e) preselect for each candidate node of the second column (P_ {{1}, f_ {2}… P_ {N}, f_ {}}) at least part of the candidate path of optimal cost that leads to said node second column candidate (P_ {{{}}}}, f_ {2}, P_ {N}, f_ {2}) or passes through it, obtaining, respectively, first N pre-selected roads or segments of roads, discarding the connections ([P_ {P} {1}] 2 ... [P_ {1} P_ {N}] 2 ... [P_ {P} {1}} 2 ... [P_ {N} P_ {N}] 2) which join the candidate nodes of the first column (P_ {1}, f_ {1} ... P_ {N}, f_ {1}) with the of the second (P_ {1}, f_ {2} ... P_ {N}, f_ {2}) that do not form part of said first roads or segments of roads preselected 2. Method according to claim 1, characterized in that when each of said candidate paths is a two-node path that includes only one of said connections, a source candidate node and a destination candidate node of two consecutive columns, comprises carrying out: - said step d) to apply said function of cost, which is at least one, to the candidate paths that include the candidate nodes of the first column (P_ {1}, f_ {1} ... P_ {N}, f_ {1}) in connection with only the candidate nodes of the second column (P_ {{{}}}}, f_ {2} ... P_ {N}, f_ {2}), and - said step e) to pre-select entirely, for each destination candidate node (P_ {1}, f_ {2} ... P_ {N}, f_ {2}) of the second column, the candidate path of optimal cost leading to him, obtaining said first N paths preselected from a node in the first column (P_ {1}, f_ {1} ... P_ {N}, f_ {1}) and end in a node of the second column (P 1, f 2 ... P N, f 2). 3. Method according to claim 2, characterized in that when M = 2 comprises selecting the two nodes that are part of the optimum cost path, from said first preselected paths, and therefore the processor or processors associated with each of said two nodes selected as a processor or mapped processors to execute the first function (f_ {1}), the one associated with the node of the first column (P_ {1}, f_ {1} ... P_ {N}, f_ {1}), and a second function (f_ {2}) associated to the node of the second column (P_ {1}, f_ {2} ... P_ {N}, f_ {2}). 4. Method according to claim 2, characterized in that when M ≥ 3 comprises performing the following steps, analogous to said stages d) and e), respectively: d ') apply at least one cost function to the candidate paths that include the candidate nodes of the second column (P_ {1}, f_ {2} ... P_ {N}, f_ {2}) in connection with those of the third (P_ {1}, f_ {3}… P_ {N}, f_ {3}), to calculate the cumulative cost associated with each of these roads candidates that includes at least the cost of the destination node of the third column (P_ {1}, f_ {3} ... P_ {N}, f_ {3}) depending of previous decisions and the cost of one of those first preselected paths in stage e), each of which includes only one node of the first column (P_ {1}, f_ {1} ... P_ {N}, f_ {1}) connected to one node of the second column (P_ {1}, f_ {2} ... P_ {N}, f_ {2}), thus obtaining the cost of each of said candidate paths destined for a respective candidate node of the third column (P_ {1}, f_ {3} ... P_ {N}, f_ {3}), and e ') preselect for each candidate node destination of the third column (P_ {1}, f_ {3} ... P_ {N}, f_ {3}) the candidate path of optimal cost that leads to it from a source node of the second column (P_ {1}, f_ {2} ... P_ {N}, f_ {2}), obtaining N seconds preselected paths, discarding the connections that bind the candidate nodes of the second column (P_ {1}, f_ {2} ... P_ {N}, f_ {2}) with those of the third (P_ {1}, f_ {3} ... P_ {N}, f_ {3}) that do not form part of these second preselected paths. 5. Method according to claim 4, characterized in that when M = 3, it comprises selecting, from said N second preselected paths, the optimum cost, and traveling it backwards together with the first preselected path to which it is attached, to select the three nodes that are part of said path, and therefore the processor or processors associated with each of said three nodes selected as mapped processors to execute the first function (f_ {1}), the one associated with the node of the first column (P_ {1}, f_ {1} ... P_ {N}, f_ {1}), the second function (f_ {2}), the one associated with the node of the second column (P_ {1}, f_ {2} ... P_ {N}, f_ {2}), and a third function (f_ {3}), the one associated with the node of the third column (P_ {1}, f_ {3} ... P_ {N} , f_ {3}). Method according to claim 4, characterized in that when M> 3, comprises performing, for each of the columns following the third column, two to two and advancing column by column, steps analogous to said stages d ') and e ') to finally preselect for each candidate node destination of each additional column to the third column, the candidate path of optimal cost that leads to it from an origin node of the immediately preceding column, obtaining N additional preselected paths per column, discarding the connections that are not part of said additional preselected roads. Method according to claim 6, characterized in that for said case in which M> 3, comprises selecting, from among the N additional preselected paths that include the nodes of the last column (P1, fM). . P_ {N}, f_ {M}), the one with optimal cost, and travel it back along with all the preselected paths to which it is attached, to select the M nodes that are part of said route, and therefore the processor or processors associated with each of said M nodes selected as mapped processors to execute the M functions (f_ {1} -f_ {M}). Method according to claim 1, characterized in that when each of said candidate paths includes an origin candidate node, an intermediate candidate node and a destination candidate node, of three consecutive columns respectively, interconnected by two of said connections, comprises carrying out : - said step d) to apply at least said cost function to the candidate roads, each of which includes, interconnected by two respective connections, a node origin candidate of the first column (P_ {1}, f_ {1} ... P_ {N}, f_ {1}), an intermediate candidate node of the second column (P_ {1}, f_ {2} ... P_ {N}, f_ {2}) and a candidate node destination of the third column (P_ {1}, f_ {3} ... P_ {N}, f_ {3}), to calculate the accumulated cost of each of these candidate paths of three nodes and that of their initial segments, or path segments of two nodes that include a node of the first column (P_ {1}, f_ {1} ... P_ {N}, f_ {1}) and one of the second (P_ {1}, f_ {2} ... P_ {N}, f_ {2}), and - said step e) to preselect, for each intermediate candidate node of the second column (P_ {1}, f_ {2} ... P_ {N}, f_ {2}), only part of the candidate path of three nodes of optimal cost that passes through it, said parts being preselected said first N road segments of two nodes 9. Method according to claim 8, characterized in that when M> 3 comprises performing the following stage, analogous to said stage d): d '') apply at least one cost function to the candidate paths of three nodes that include, interconnected by two respective connections, a candidate node origin of the second column (P_ {1}, f_ {2} ... P_ {N}, f_ {2}), a candidate node intermediate of the third column (P1, f3) ... f_ {3}) and a destination candidate node of the fourth column (P_ {1}, f_ {4} ... P_ {N}, f_ {4}), to calculate the accumulated cost associated to each of these candidate paths of three nodes, which includes at least the cost of the destination node of the fourth column (P_ {1}, f_ {4} ... P_ {N}, f_ {4}) based on decisions precedents and the accumulated cost of one of said first path segments of two preselected nodes in the stage and). Method according to claim 9, characterized in that when M = 4, it comprises performing the following steps, after said step d ''): - preselect for each candidate node intermediate of the third column (P1, f3) ... f_ {3}), the complete candidate path of three nodes of optimal cost that passes through it and ends in a node of the fourth column (P_ {1}, f_ {4} ... P_ {N}, f_ {4}), obtaining N paths of three preselected nodes, and - select, from among said N paths of three preselected nodes, the optimal cost one, and travel that path selected from its intermediate node, from the third column (P_ {1}, f_ {3} ... P_ {N}, f_ {3}), forward and backward along with the first preselected two-node path segment in step e) to which it is attached, to select the four nodes that are part of this route, and therefore the processor or processors associated with each of said four selected nodes as mapped processors to run the first function (f_ {1}), the one associated with the node of the first column (P_ {1}, f_ {1} ... P_ {N}, f_ {1}), the second function (f_ {2}), the one associated with the node of the second column (P_ {1}, f_ {2} ... P_ {N}, f_ {2}, the third function (f_ {3}), the associated to the node of the third column (P_ {1}, f_ {3} ... P_ {N}, f_ {3}), and a fourth function (f_ {4}), the one associated with fourth column node (P_ {1}, f_ {4} ... P_ {N}, f_ {4}). 11. Method according to claim 9, characterized in that when M> 4, comprises performing the following stage after said step d ''): e '') preselect, for each candidate node intermediate of the third column (P1, f3) ... f_ {3}), only part of the candidate path of three nodes of optimal cost that passes through it and ends in a candidate node of the fourth column (P_ {1}, f_ {4} ... P_ {N}, f_ {4}), obtaining N second path segments of two preselected nodes. 12. Method according to claim 11, characterized in that it comprises performing, for each of the columns following the fourth column (P 1, f 4, P 14, f 4), three three and advancing column by column, stages analogous to said stages d '') and e ''). for each additional three-node path that ends in each of the following or additional columns to the fourth column (P_ {1}, f_ {4} ... P_ {N}, f_ {4}), except for the path of three additional nodes ending in column M, to preselect for each intermediate candidate node of each additional three-node path, only part of the candidate path of three nodes of optimal cost that passes through it, obtaining N path segments of two nodes additional preselected for each additional column to the fourth column (P1, f4) ...
f_ {4}). 13. Method according to claim 12, characterized in that it comprises performing, for said paths of three additional nodes ending in column M, the following stages, after said stage analogous to said stage e ''): - preselect for each candidate node intermediate of column M-1 (P1), f_ {M-1} ... P_ {N}, f_ {M-1}), the complete candidate path of three nodes of optimal cost that passes through it and ends at a node in column M (P_ {1}, f_ {M} ... P_ {N}, f_ {M}), obtaining N paths of three nodes preselected, and - select, from among said N paths of three preselected nodes, the optimal cost one, and travel that path selected from its intermediate node (P_ {1}, f_ {M-1} ... P_ {N}, f_ {M-1}), of column M-1, forward and backward together with all the path segments of two preselected nodes to those that are united, to select the M nodes that form part of said route, and therefore the processor or processors associated to each of said M nodes selected as mapped processors to execute the M functions (f_ {1} -f_ {M}). 14. Method according to any of claims 8 to 13, characterized in that when each of said candidate paths is a path of more than three nodes, it comprises carrying out: - said step d) to calculate the cost accumulated of each of said candidate paths of more than three nodes and the one of its initial path segments of two nodes that include a node in the first column (P_ {1} f_ {1} ... P_ {N} f_ {1}) and one of the second (P_ {1}, f_ {2} ... P_ {N}, f_ {2}), - said step e) to preselect, for each candidate node of the second column (P_ {1}, f_ {2} ... P_ {N}, f_ {2}), only part of the candidate path of more than three nodes  of optimal cost that passes through it, said parts being preselected said first N road segments of two nodes, and - perform stages analogous to these stages described for said three-node paths, preselecting to each candidate node that occupies the second position on each path of plus three additional nodes ending in each of the columns following or additional to the fourth column, except for the path of more than three additional nodes ending in column M, the initial segment of two nodes of the candidate path of more than three nodes of optimal cost that passes through it, obtaining N segments of path of two additional preselected nodes per column additional to the fourth column. 15. Method according to claim 14, characterized in that it comprises performing, for said paths of more than three additional nodes ending in column M, the following steps: - preselect for each candidate node that occupies the second position on each path of three more nodes additional ending in column M, the candidate path complete with more than three nodes of optimal cost that passes through it and ends in a node of column M (P_ {1}, f_ {M} ... P_ {N}, f_ {M}), obtaining N paths of plus three preselected nodes, Y - select, from among said N paths of more of three preselected nodes, the optimum cost, and traverse said path selected from its respective node that occupies the second position on the road, forward and backward together with all the path segments of two preselected nodes to those that are united, to select the M nodes that form part of said route, and therefore the processor or processors associated to each of said M nodes selected as mapped processors to execute the M functions (f_ {1} -f_ {M}). 16. Method according to claim 1, characterized in that said preceding decisions influence said accumulated cost at least as regards the resources remaining in each node after having made said decisions. 17. Method according to any of the preceding claims, characterized in that said cost functions evaluate the computation cost that relates the processing requirements to the available processing resources, and the communication cost puts that relates the bandwidth requirements to transfer data from one process to another with the corresponding bandwidths available between the processors. 18. Method according to claim 1, characterized in that it comprises determining said computing resources of each of said processors (P 1 -P N) based on a time segment of a plurality of equal time segments, of duration ( t_ {ts}) sufficient to ensure that each processor (P_ {1} -P_ {N}) is capable of executing the functions (f_ {1} -f_ {M}) or processes assigned and transferring data to another processor before the end of said temporary segment. 19. Method according to claim 18, characterized in that it comprises mapping said functions (f_ {1} -f_ {M}) in order to finally execute the application through an instruction segmentation process, establishing the same computing requirements for each of said temporary segments.