Requisito previo: Métodos de asignación de particiones en la administración de memoria
En Asignación de particiones , cuando hay más de una partición disponible libremente para acomodar una solicitud de proceso, se debe seleccionar una partición. Para elegir una partición en particular, se necesita un método de asignación de partición. Un método de asignación de particiones se considera mejor si evita la fragmentación interna.
Considere los siguientes datos para el proceso:
Nro. de proceso | Tamaño del proceso |
---|---|
1 | 88 |
2 | 192 |
3 | 277 |
4 | 365 |
5 | 489 |
Considere los siguientes datos para las ranuras de memoria:
Número de bloque de memoria | Tamaño del bloque de memoria |
---|---|
1 | 400 |
2 | 500 |
3 | 300 |
4 | 200 |
5 | 100 |
A continuación se muestran los diversos esquemas de asignación de particiones con su implementación con respecto a los datos proporcionados anteriormente.
1. Primer ajuste
Este método mantiene la lista de trabajos libres/ocupados organizada por ubicación de memoria, de orden bajo a memoria de orden alto. En este método, el primer trabajo reclama la primera memoria disponible con espacio mayor o igual a su tamaño. El sistema operativo no busca la partición adecuada, sino que simplemente asigna el trabajo a la partición de memoria más cercana disponible con tamaño suficiente.
A continuación se muestra la implementación del algoritmo First Fit :
// C++ program for the implementation // of the First Fit algorithm #include <iostream> #include <queue> #include <vector> using namespace std; // Process Class class process { public: // Size & number of process size_t size; pid_t no; }; // Memory Class class memory { public: size_t size; // Number of memory & queue of space // occupied by process pid_t no; queue<process> space_occupied; // Function to push process in a block void push(const process p) { if (p.size <= size) { space_occupied.push(p); size -= p.size; } } // Function to pop and return the // process from the block process pop() { process p; // If space occupied is empty if (!space_occupied.empty()) { p = space_occupied.front(); space_occupied.pop(); size += p.size; return p; } } // Function to check if block is // completely empty bool empty() { return space_occupied.empty(); } }; // Function to get data of processess // allocated using first fit vector<memory> first_fit(vector<memory> memory_blocks, queue<process> processess) { int i = 0; bool done, done1; memory na; na.no = -10; while (!processess.empty()) { done = 0; for (i = 0; i < memory_blocks.size(); i++) { done1 = 0; if (memory_blocks.at(i).size >= processess.front().size) { memory_blocks.at(i).push(processess.front()); done = 1; done1 = 1; break; } } // If process is done if (done == 0 && done1 == 0) { na.size += processess.front().size; na.push(processess.front()); } // pop the process processess.pop(); } if (!na.space_occupied.empty()) memory_blocks.push_back(na); return memory_blocks; } // Function to display the allocation // of all processess void display(vector<memory> memory_blocks) { int i = 0, temp = 0; process p; cout << "+-------------+--------------+--------------+" << endl; cout << "| Process no. | Process size | Memory block |" << endl; cout << "+-------------+--------------+--------------+" << endl; // Traverse memory blocks size for (i = 0; i < memory_blocks.size(); i++) { // While memory block size is not empty while (!memory_blocks.at(i).empty()) { p = memory_blocks.at(i).pop(); temp = to_string(p.no).length(); cout << "|" << string(7 - temp / 2 - temp % 2, ' ') << p.no << string(6 - temp / 2, ' ') << "|"; temp = to_string(p.size).length(); cout << string(7 - temp / 2 - temp % 2, ' ') << p.size << string(7 - temp / 2, ' ') << "|"; temp = to_string(memory_blocks.at(i).no).length(); cout << string(7 - temp / 2 - temp % 2, ' '); // If memory blocks is assigned if (memory_blocks.at(i).no != -10) { cout << memory_blocks.at(i).no; } // Else memory blocks is assigned else { cout << "N/A"; } cout << string(7 - temp / 2, ' ') << "|" << endl; } } cout << "+-------------+--------------+--------------+" << endl; } // Driver Code int main() { // Declare memory blocks vector<memory> memory_blocks(5); // Declare first fit blocks vector<memory> first_fit_blocks; // Declare queue of all processess queue<process> processess; process temp; // Set sample data memory_blocks[0].no = 1; memory_blocks[0].size = 400; memory_blocks[1].no = 2; memory_blocks[1].size = 500; memory_blocks[2].no = 3; memory_blocks[2].size = 300; memory_blocks[3].no = 4; memory_blocks[3].size = 200; memory_blocks[4].no = 5; memory_blocks[4].size = 100; temp.no = 1; temp.size = 88; // Push the process processess.push(temp); temp.no = 2; temp.size = 192; // Push the process processess.push(temp); temp.no = 3; temp.size = 277; // Push the process processess.push(temp); temp.no = 4; temp.size = 365; // Push the process processess.push(temp); temp.no = 5; temp.size = 489; // Push the process processess.push(temp); // Get the data first_fit_blocks = first_fit(memory_blocks, processess); // Display the data display(first_fit_blocks); memory_blocks.clear(); memory_blocks.shrink_to_fit(); first_fit_blocks.clear(); first_fit_blocks.shrink_to_fit(); return 0; }
+-------------+--------------+--------------+ | Process no. | Process size | Memory block | +-------------+--------------+--------------+ | 1 | 88 | 1 | | 2 | 192 | 1 | | 3 | 277 | 2 | | 4 | 365 | N/A | | 5 | 489 | N/A | +-------------+--------------+--------------+
2. Siguiente ajuste
El siguiente ajuste es una versión modificada de ‘primer ajuste’. Comienza como el primer ajuste para encontrar una partición libre, pero cuando se le llama la próxima vez, comienza a buscar desde donde lo dejó, no desde el principio. Esta política hace uso de un puntero itinerante. El puntero se mueve a lo largo de la string de memoria para buscar el siguiente ajuste. Esto ayuda a evitar el uso de la memoria siempre desde la cabeza (comienzo) de la string de bloques libre.
A continuación se muestra la implementación del algoritmo Next Fit :
// C++ program for the implementation // of the Next Fit algorithm #include <iostream> #include <queue> #include <vector> using namespace std; // Process Class class process { public: // Size & number of process size_t size; pid_t no; }; // Memory Class class memory { public: size_t size; // Number of memory & queue of space // occupied by process pid_t no; queue<process> space_occupied; // Function to push process in a block void push(const process p) { if (p.size <= size) { space_occupied.push(p); size -= p.size; } } // Function to pop and return the // process from the block process pop() { process p; // If space occupied is empty if (!space_occupied.empty()) { p = space_occupied.front(); space_occupied.pop(); size += p.size; return p; } } // Function to check if block is // completely empty bool empty() { return space_occupied.empty(); } }; // Function to get data of processess // allocated using Next Fit vector<memory> next_fit(vector<memory> memory_blocks, queue<process> processess) { int i = 0; bool done, done1; memory na; na.no = -10; // Loop till process is empty while (!processess.empty()) { done1 = 0; // Traverse memory_blocks for (i = 0; i < memory_blocks.size(); i++) { done = 0; // If process is not empty if (!processess.empty() && memory_blocks.at(i).size >= processess.front().size) { memory_blocks.at(i).push(processess.front()); done = 1; done1 = 1; processess.pop(); } } if (!processess.empty() && done == 0 && done1 == 0) { na.size += processess.front().size; na.push(processess.front()); processess.pop(); } } // If space is not occupied push // the memory_blocks na if (!na.space_occupied.empty()) { memory_blocks.push_back(na); } return memory_blocks; } // Function to display the allocation // of all processess void display(vector<memory> memory_blocks) { int i = 0, temp = 0; process p; cout << "+-------------+--------------+--------------+" << endl; cout << "| Process no. | Process size | Memory block |" << endl; cout << "+-------------+--------------+--------------+" << endl; // Traverse memory blocks size for (i = 0; i < memory_blocks.size(); i++) { // While memory block size is not empty while (!memory_blocks.at(i).empty()) { p = memory_blocks.at(i).pop(); temp = to_string(p.no).length(); cout << "|" << string(7 - temp / 2 - temp % 2, ' ') << p.no << string(6 - temp / 2, ' ') << "|"; temp = to_string(p.size).length(); cout << string(7 - temp / 2 - temp % 2, ' ') << p.size << string(7 - temp / 2, ' ') << "|"; temp = to_string(memory_blocks.at(i).no).length(); cout << string(7 - temp / 2 - temp % 2, ' '); // If memory blocks is assigned if (memory_blocks.at(i).no != -10) { cout << memory_blocks.at(i).no; } // Else memory blocks is assigned else { cout << "N/A"; } cout << string(7 - temp / 2, ' ') << "|" << endl; } } cout << "+-------------+--------------+--------------+" << endl; } // Driver Code int main() { // Declare memory blocks vector<memory> memory_blocks(5); // Declare next fit blocks vector<memory> next_fit_blocks; // Declare queue of all processess queue<process> processess; process temp; // Set sample data memory_blocks[0].no = 1; memory_blocks[0].size = 400; memory_blocks[1].no = 2; memory_blocks[1].size = 500; memory_blocks[2].no = 3; memory_blocks[2].size = 300; memory_blocks[3].no = 4; memory_blocks[3].size = 200; memory_blocks[4].no = 5; memory_blocks[4].size = 100; temp.no = 1; temp.size = 88; // Push the process processess.push(temp); temp.no = 2; temp.size = 192; // Push the process processess.push(temp); temp.no = 3; temp.size = 277; // Push the process processess.push(temp); temp.no = 4; temp.size = 365; // Push the process processess.push(temp); temp.no = 5; temp.size = 489; // Push the process processess.push(temp); // Get the data next_fit_blocks = next_fit(memory_blocks, processess); // Display the data display(next_fit_blocks); memory_blocks.clear(); memory_blocks.shrink_to_fit(); next_fit_blocks.clear(); next_fit_blocks.shrink_to_fit(); return 0; }
+-------------+--------------+--------------+ | Process no. | Process size | Memory block | +-------------+--------------+--------------+ | 1 | 88 | 1 | | 2 | 192 | 2 | | 3 | 277 | 3 | | 4 | 365 | N/A | | 5 | 489 | N/A | +-------------+--------------+--------------+
3. Peor ajuste
Worst Fit asigna un proceso a la partición que es lo suficientemente grande entre las particiones disponibles gratuitamente en la memoria principal. Si un proceso grande llega en una etapa posterior, la memoria no tendrá espacio para acomodarlo.
A continuación se muestra la implementación del algoritmo de peor ajuste :
// C++ program for the implementation // of the Worst Fit algorithm #include <iostream> #include <queue> #include <vector> using namespace std; // Process Class class process { public: // Size & number of process size_t size; pid_t no; }; // Memory Class class memory { public: size_t size; // Number of memory & queue of space // occupied by process pid_t no; queue<process> space_occupied; // Function to push process in a block void push(const process p) { if (p.size <= size) { space_occupied.push(p); size -= p.size; } } // Function to pop and return the // process from the block process pop() { process p; // If space occupied is empty if (!space_occupied.empty()) { p = space_occupied.front(); space_occupied.pop(); size += p.size; return p; } } // Function to check if block is // completely empty bool empty() { return space_occupied.empty(); } }; // Function to get data of processess // allocated using Worst Fit vector<memory> worst_fit(vector<memory> memory_blocks, queue<process> processess) { int i = 0, index = 0, max; memory na; na.no = -10; // Loop till process queue is not empty while (!processess.empty()) { max = 0; // Traverse the memory_blocks for (i = 0; i < memory_blocks.size(); i++) { if (memory_blocks.at(i).size >= processess.front().size && memory_blocks.at(i).size > max) { max = memory_blocks.at(i).size; index = i; } } if (max != 0) { memory_blocks.at(index).push(processess.front()); } else { na.size += processess.front().size; na.push(processess.front()); } // Pop the current process processess.pop(); } // If space is not occupied if (!na.space_occupied.empty()) { memory_blocks.push_back(na); } // Return the memory return memory_blocks; } // Function to display the allocation // of all processess void display(vector<memory> memory_blocks) { int i = 0, temp = 0; process p; cout << "+-------------+--------------+--------------+" << endl; cout << "| Process no. | Process size | Memory block |" << endl; cout << "+-------------+--------------+--------------+" << endl; // Traverse memory blocks size for (i = 0; i < memory_blocks.size(); i++) { // While memory block size is not empty while (!memory_blocks.at(i).empty()) { p = memory_blocks.at(i).pop(); temp = to_string(p.no).length(); cout << "|" << string(7 - temp / 2 - temp % 2, ' ') << p.no << string(6 - temp / 2, ' ') << "|"; temp = to_string(p.size).length(); cout << string(7 - temp / 2 - temp % 2, ' ') << p.size << string(7 - temp / 2, ' ') << "|"; temp = to_string(memory_blocks.at(i).no).length(); cout << string(7 - temp / 2 - temp % 2, ' '); // If memory blocks is assigned if (memory_blocks.at(i).no != -10) { cout << memory_blocks.at(i).no; } // Else memory blocks is assigned else { cout << "N/A"; } cout << string(7 - temp / 2, ' ') << "|" << endl; } } cout << "+-------------+--------------+--------------+" << endl; } // Driver Code int main() { // Declare memory blocks vector<memory> memory_blocks(5); // Declare worst fit blocks vector<memory> worst_fit_blocks; // Declare queue of all processess queue<process> processess; process temp; // Set sample data memory_blocks[0].no = 1; memory_blocks[0].size = 400; memory_blocks[1].no = 2; memory_blocks[1].size = 500; memory_blocks[2].no = 3; memory_blocks[2].size = 300; memory_blocks[3].no = 4; memory_blocks[3].size = 200; memory_blocks[4].no = 5; memory_blocks[4].size = 100; temp.no = 1; temp.size = 88; // Push the process processess.push(temp); temp.no = 2; temp.size = 192; // Push the process processess.push(temp); temp.no = 3; temp.size = 277; // Push the process processess.push(temp); temp.no = 4; temp.size = 365; // Push the process processess.push(temp); temp.no = 5; temp.size = 489; // Push the process processess.push(temp); // Get the data worst_fit_blocks = worst_fit(memory_blocks, processess); // Display the data display(worst_fit_blocks); memory_blocks.clear(); memory_blocks.shrink_to_fit(); worst_fit_blocks.clear(); worst_fit_blocks.shrink_to_fit(); return 0; }
+-------------+--------------+--------------+ | Process no. | Process size | Memory block | +-------------+--------------+--------------+ | 3 | 277 | 1 | | 1 | 88 | 2 | | 2 | 192 | 2 | | 4 | 365 | N/A | | 5 | 489 | N/A | +-------------+--------------+--------------+
4. Mejor ajuste
Este método mantiene la lista de disponibilidad ordenada por tamaño, de menor a mayor. En este método, el sistema operativo primero busca en toda la memoria de acuerdo con el tamaño del trabajo dado y lo asigna a la partición libre más cercana en la memoria, lo que le permite usar la memoria de manera eficiente. Aquí los trabajos están en el orden del trabajo más pequeño al trabajo más grande.
A continuación se muestra la implementación del algoritmo Best Fit :
// C++ program for the implementation // of the Best Fit algorithm #include <iostream> #include <queue> #include <vector> using namespace std; // Process Class class process { public: // Size & number of process size_t size; pid_t no; }; // Memory Class class memory { public: size_t size; // Number of memory & queue of space // occupied by process pid_t no; queue<process> space_occupied; // Function to push process in a block void push(const process p) { if (p.size <= size) { space_occupied.push(p); size -= p.size; } } // Function to pop and return the // process from the block process pop() { process p; // If space occupied is empty if (!space_occupied.empty()) { p = space_occupied.front(); space_occupied.pop(); size += p.size; return p; } } // Function to check if block is // completely empty bool empty() { return space_occupied.empty(); } }; // Function to get data of processess // allocated using Best Fit vector<memory> best_fit(vector<memory> memory_blocks, queue<process> processess) { int i = 0, min, index = 0; memory na; na.no = -10; // Loop till processe is not empty while (!processess.empty()) { min = 0; // Traverse the memory_blocks for (i = 0; i < memory_blocks.size(); i++) { if (memory_blocks.at(i).size >= processess.front().size && (min == 0 || memory_blocks.at(i).size < min)) { min = memory_blocks.at(i).size; index = i; } } if (min != 0) { memory_blocks.at(index).push(processess.front()); } else { na.size += processess.front().size; na.push(processess.front()); } // Pop the processe processess.pop(); } // If space is no occupied then push // the current memory na if (!na.space_occupied.empty()) { memory_blocks.push_back(na); } // Return the memory_blocks return memory_blocks; } // Function to display the allocation // of all processess void display(vector<memory> memory_blocks) { int i = 0, temp = 0; process p; cout << "+-------------+--------------+--------------+" << endl; cout << "| Process no. | Process size | Memory block |" << endl; cout << "+-------------+--------------+--------------+" << endl; // Traverse memory blocks size for (i = 0; i < memory_blocks.size(); i++) { // While memory block size is not empty while (!memory_blocks.at(i).empty()) { p = memory_blocks.at(i).pop(); temp = to_string(p.no).length(); cout << "|" << string(7 - temp / 2 - temp % 2, ' ') << p.no << string(6 - temp / 2, ' ') << "|"; temp = to_string(p.size).length(); cout << string(7 - temp / 2 - temp % 2, ' ') << p.size << string(7 - temp / 2, ' ') << "|"; temp = to_string(memory_blocks.at(i).no).length(); cout << string(7 - temp / 2 - temp % 2, ' '); // If memory blocks is assigned if (memory_blocks.at(i).no != -10) { cout << memory_blocks.at(i).no; } // Else memory blocks is assigned else { cout << "N/A"; } cout << string(7 - temp / 2, ' ') << "|" << endl; } } cout << "+-------------+--------------+--------------+" << endl; } // Driver Code int main() { // Declare memory blocks vector<memory> memory_blocks(5); // Declare best fit blocks vector<memory> best_fit_blocks; // Declare queue of all processess queue<process> processess; process temp; // Set sample data memory_blocks[0].no = 1; memory_blocks[0].size = 400; memory_blocks[1].no = 2; memory_blocks[1].size = 500; memory_blocks[2].no = 3; memory_blocks[2].size = 300; memory_blocks[3].no = 4; memory_blocks[3].size = 200; memory_blocks[4].no = 5; memory_blocks[4].size = 100; temp.no = 1; temp.size = 88; // Push the processe to queue processess.push(temp); temp.no = 2; temp.size = 192; // Push the processe to queue processess.push(temp); temp.no = 3; temp.size = 277; // Push the processe to queue processess.push(temp); temp.no = 4; temp.size = 365; // Push the processe to queue processess.push(temp); temp.no = 5; temp.size = 489; // Push the processe to queue processess.push(temp); // Get the data best_fit_blocks = best_fit(memory_blocks, processess); // Display the data display(best_fit_blocks); memory_blocks.clear(); memory_blocks.shrink_to_fit(); best_fit_blocks.clear(); best_fit_blocks.shrink_to_fit(); return 0; }
+-------------+--------------+--------------+ | Process no. | Process size | Memory block | +-------------+--------------+--------------+ | 4 | 365 | 1 | | 5 | 489 | 2 | | 3 | 277 | 3 | | 2 | 192 | 4 | | 1 | 88 | 5 | +-------------+--------------+--------------+
Publicación traducida automáticamente
Artículo escrito por jaspreetsinghpal18 y traducido por Barcelona Geeks. The original can be accessed here. Licence: CCBY-SA