from os import * import psutil from utils.log.trans_log import trans_print def print_memory_usage(detail=""): # 获取当前进程ID pid = getpid() # 获取进程信息 py = psutil.Process(pid) # 获取内存信息 memory_info = py.memory_info() # RSS (Resident Set Size) 是进程实际占用的物理内存大小 memory_usage_rss = memory_info.rss # VMS (Virtual Memory Size) 是进程使用的虚拟内存大小 memory_usage_vms = memory_info.vms # 将字节转换为更易读的单位 memory_usage_rss_mb = memory_usage_rss / (1024 ** 2) memory_usage_vms_mb = memory_usage_vms / (1024 ** 2) trans_print(f"{detail},Memory usage (RSS): {memory_usage_rss_mb:.2f} MB") trans_print(f"{detail},Memory usage (VMS): {memory_usage_vms_mb:.2f} MB") def get_cpu_count(): return psutil.cpu_count() def get_available_cpu_count_with_percent(percent: float = 1): cpu_count = get_cpu_count() return int(cpu_count * percent) def get_file_size(file_path): return path.getsize(file_path) def get_dir_size(dir_path): return sum(get_file_size(path.join(dir_path, file)) for file in listdir(dir_path) if path.isfile(path.join(dir_path, file))) def get_available_memory_with_percent(percent: float = 1): memory_info = psutil.virtual_memory() return int(memory_info.available * percent) def get_max_file_size(file_paths: list[str]): max_size = 0 for file_path in file_paths: file_size = get_file_size(file_path) if file_size > max_size: max_size = file_size return max_size def use_files_get_max_cpu_count(file_paths: list[str], memory_percent: float = 1 / 12, cpu_percent: float = 2 / 5): max_file_size = get_max_file_size(file_paths) free_memory = get_available_memory_with_percent(memory_percent) count = int(free_memory / max_file_size) max_cpu_count = get_available_cpu_count_with_percent(cpu_percent) result = count if count <= max_cpu_count else max_cpu_count if result == 0: result = 1 if result > len(file_paths): result = len(file_paths) trans_print("总文件数:", len(file_paths), ",获取最大文件大小:", str(round(max_file_size / 2 ** 20, 2)) + "M", "可用内存:", str(get_available_memory_with_percent(1) / 2 ** 20) + "M", "总CPU数:", get_cpu_count(), "CPU使用比例:", round(cpu_percent, 2), "CPU可用数量:", max_cpu_count, ",最终确定使用进程数:", result) return result def max_file_size_get_max_cpu_count(max_file_size, memory_percent: float = 1 / 6, cpu_percent: float = 2 / 5): free_memory = get_available_memory_with_percent(memory_percent) count = int(free_memory / max_file_size) max_cpu_count = get_available_cpu_count_with_percent(cpu_percent) result = count if count <= max_cpu_count else max_cpu_count if result == 0: result = 1 trans_print(",获取最大文件大小:", str(round(max_file_size / 2 ** 20, 2)) + "M", "可用内存:", str(get_available_memory_with_percent(1) / 2 ** 20) + "M", "总CPU数:", get_cpu_count(), "CPU使用比例:", round(cpu_percent, 2), "CPU可用数量:", max_cpu_count, ",最终确定使用进程数:", result) return result if __name__ == '__main__': from utils.file.trans_methods import read_files import datetime read_path = r"Z:\collection_data\1进行中\密马风电场-山西-大唐\收资数据\scada\秒级数据" begin = datetime.datetime.now() all_files = read_files(read_path) print(datetime.datetime.now() - begin) print(use_files_get_max_cpu_count(all_files)) print(get_available_memory_with_percent(1) / 2 ** 20) print(get_available_memory_with_percent(2 / 3) / 2 ** 20) begin = datetime.datetime.now() print(len(all_files)) print(get_max_file_size(all_files) / 2 ** 20) print(datetime.datetime.now() - begin)