您现在的位置是:首页 >技术教程 >计算卸载论文阅读01-理论梳理网站首页技术教程

计算卸载论文阅读01-理论梳理

༺ཌ༈Dream&Light༈ད༻ 2023-06-30 12:00:02
简介计算卸载论文阅读01-理论梳理

标题:When Learning Joins Edge: Real-time Proportional Computation Offloading via Deep Reinforcement Learning

会议:ICPADS 2019

一、梳理

问题:在任务进行卸载时,往往忽略了任务的特定的卸载比例。

模型:针对上述问题,我们提出了一种创新的强化学习(RL)方法来解决比例计算问题。我们考虑了一种常见的卸载场景,该场景具有时变带宽和异构设备,并且设备不断生成应用程序。对于每个应用程序,客户端必须选择本地或远程执行该应用程序,并确定要卸载的比例。我们将该问题制定为一个长期优化问题,然后提出一种基于RL的算法来解决该问题。基本思想是估计可能决策的收益,其中选择收益最大的决策。我们没有采用原来的深度Q网络(DQN),而是通过添加优先级缓冲机制专家缓冲机制,提出了Advanced DQN(ADQN),分别提高了样本的利用率和克服了冷启动问题。

目的:最小化时延和能耗的权值和(优化参数:用户卸载策略卸载的比例)。

算法:ADQN。

二、模型细节与算法

2.1系统模型

计算卸载主要包含: 本地计算和边缘计算

1、本地计算:用户U_i在时隙t_j时的计算时延(计算时延和排队时延)和能耗为

T_{ij}^{l}=D_{ij}^{l}+t_{q}^{l}=frac{omega B_{ij}Q_{y_{ij}}}{f_{ij}^{l}}+t_{q}^{l}

E_{ij}^{l}=e_lB_{ij}Q_{y_{ij}}

其中B_{ij}表示任务大小,Q_{y_{ij}}表示本地计算任务的比例

2、边缘计算:包含时延(上下行传输时延+执行时延+排队时延)和能耗
T_{ij}^{r}=D_{ij}^{r}+t_q^r=frac{omega B_{ij}P_{y_{ij}}}{F_{ij}^r}+frac{B_{ij}P_{y_{ij}}}{r_{ul}^{ij}}+frac{omega^s B_{ij}}{r_{dl}^{ij}} + t_q^r

E_{ij}^r=e_r B_{ij}P_{y_{ij}}

其中P_{y_{ij}}表示卸载到边缘服务器的比例,omega^s表示每单位计算任务处理后得到的结果大小。

3、总的时延与能耗成本

W_{ij}=lambda max (T_{ij}^r,T_{ij}^l)+eta(E_{ij}^r+E_{ij}^l)

2.2优化目标

min sum_{t_j=1}^{T}[W_{ij}]

       {	t s.t.}   f_{x_{ij}}^r leqslant F_{x_{ij}}^r, f_{x_{ij}}^l leqslant F_{x_{ij}}^l

                r_{ul}^{ij} leqslant R_{ul}^i, r_{dl}^{ij} leqslant R_{dl}^i

                x_{ij} in {1,2,...,m},y_{ij} in {1,2,...,n}

2.3 State、Action和Reward

1、状态空间:

状态空间包含:(数据大小、本地CPU频率,local可用计算资源,上行速率,下行速率,ES可用计算资源)

2、动作空间:

动作空间包含:(用户选择ES的策略,任务卸载比例)

3、奖励:

奖励:R=frac{W_l - W(s,a)}{W_l}

W_l表示本地计算成本,W(s,a)表示部分卸载产生的成本。

2.4代码

计算卸载环境代码:

import math
import copy
import numpy as np


class ENV():
    def __init__(self, UEs=3, MECs=7, k=33, lam=0.5):
        self.UEs = UEs
        self.MECs = MECs
        self.k = k

        q = np.full((k, 1), 0.)
        p = np.linspace(0, 1, k).reshape((k, 1))
        # Create Action
        for i in range(MECs - 1):
            a = np.full((k, 1), float(i + 1))
            b = np.linspace(0, 1, k).reshape((k, 1))
            q = np.append(q, a, axis=0)
            p = np.append(p, b, axis=0)  # 231(33 * 7) * 33

        self.actions = np.hstack((q, p))  # 231 * 2
        self.n_actions = len(self.actions)  # 231
        self.n_features = 3 + MECs * 3  # 3 + 7 * 3
        self.discount = 0.01

        # 基本参数
        # 频率
        self.Hz = 1
        self.kHz = 1000 * self.Hz
        self.mHz = 1000 * self.kHz
        self.GHz = 1000 * self.mHz
        self.nor = 10 ** (-7)
        self.nor1 = 10 ** 19

        # 数据大小
        self.bit = 1
        self.B = 8 * self.bit
        self.KB = 1024 * self.B
        self.MB = 1024 * self.KB

        # self.task_cpu_cycle = np.random.randint(2 * 10**9, 3* 10**9)

        self.UE_f = np.random.randint(1.5 * self.GHz * self.nor, 2 * self.GHz * self.nor)  # UE的计算能力
        self.MEC_f = np.random.randint(5 * self.GHz * self.nor, 7 * self.GHz * self.nor)  # MEC的计算能力
        # self.UE_f = 500 * self.mHz     # UE的计算能力
        # self.MEC_f = np.random.randint(5.2 * self.GHz, 24.3 * self.GHz)  # MEC的计算能力
        self.tr_energy = 1  # 传输能耗
        self.r = 40 * math.log2(1 + (16 * 10)) * self.MB * self.nor  # 传输速率
        # self.r = 800 # 传输速率
        self.ew, self.lw = 10 ** (-26), 3 * 10 ** (-26)  # 能耗系数
        # self.ew, self.lw = 0.3, 0.15 # 能耗系数
        self.et, self.lt = 1, 1
        self.local_core_max, self.local_core_min = 1.3 * self.UE_f, 0.7 * self.UE_f
        self.server_core_max, self.server_core_min = 1.3 * self.MEC_f, 0.7 * self.MEC_f
        self.uplink_max, self.uplink_min = 1.3 * self.r, 0.7 * self.r
        self.downlink_max, self.downlink_min = 1.3 * self.r, 0.7 * self.r
        self.lam = lam
        self.e = 1

    def reset(self):
        # 初始化环境,状态空间
        obs = []
        servers_cap = []
        new_cap = True
        for i in range(self.UEs):
            uplink, downlink = [], []
            # np.random.seed(np.random.randint(1, 1000))
            # task_size = np.random.randint(2 * 10**8 * self.nor, 3 * 10**8 * self.nor) #   任务大小
            task_size = np.random.randint(1.5 * self.mHz, 2 * self.mHz)  # 任务大小
            # self.task_size = self.task_size * self.task_cpu_cycle                     # 处理一个任务所需要的cpu频率
            # task_cpu_cycle = np.random.randint(2 * 10**9 * self.nor, 3 * 10**9 * self.nor)
            task_cpu_cycle = np.random.randint(10 ** 3, 10 ** 5)
            local_comp = np.random.randint(0.9 * self.UE_f, 1.1 * self.UE_f)  # UE的计算能力
            for i in range(self.MECs):
                up = np.random.randint(0.9 * self.r, 1.1 * self.r)
                down = np.random.randint(0.9 * self.r, 1.1 * self.r)
                if new_cap:
                    cap = np.random.randint(0.9 * self.MEC_f, 1.1 * self.MEC_f)  # MEC计算能力
                    servers_cap.append(cap)
                uplink.append(up)
                downlink.append(down)
            observation = np.array([task_size, task_cpu_cycle, local_comp])
            observation = np.hstack((observation, servers_cap, uplink, downlink))
            obs.append(observation)
            new_cap = False
        return obs

    def choose_action(self, prob):
        """
        根据概率选择动作
        :param prob:
        :return:
        """
        action_choice = np.linspace(0, 1, self.k)
        actions = []
        for i in range(self.UEs):
            a = np.random.choice(a=(self.MECs * self.k), p=prob[i])
            target_server = int(a / self.k)
            percen = action_choice[a % self.k]
            action = [target_server, percen]
            actions.append(action)
        return actions

    def step(self, observation, actions_prob, is_prob=True, is_compared=True):
        if is_prob:
            actions = self.choose_action(actions_prob)
        else:
            actions = actions_prob
        new_cap = False
        obs_ = []
        rew, local, ran, mec = [], [], [], []
        dpg_times, local_times, ran_times, mec_times = [], [], [], []
        dpg_energys, local_energys, ran_energys, mec_energys = [], [], [], []
        total = []
        a, b, c, d = 0, 0, 0, 0
        for i in range(self.UEs):
            if i == self.UEs - 1:
                new_cap = True
            # 提取信息
            task_size, task_cpu_cycle, local_comp, servers_cap, uplink, downlink = 
                observation[i][0], observation[i][1], observation[i][2], observation[i][3:3+self.MECs], observation[i][3+self.MECs:3+self.MECs*2], observation[i][3+self.MECs*2:3+self.MECs*3]

            action = actions[i]
            target_server, percen = int(action[0]), action[1]

            # 计算奖励
            # 1=======部分卸载==========
            # 卸载及回传数据产生的时延和能耗
            tr_time = (percen * task_size) / uplink[target_server] + self.discount * (percen * task_size) / downlink[
                target_server]
            tr_energy = (self.tr_energy * percen * task_size) / uplink[target_server] + self.discount * (
                        self.tr_energy * percen * task_size) / downlink[target_server]

            # 本地计算时延和能耗
            comp_local_time = task_cpu_cycle * (1 - percen) / (local_comp)
            comp_local_energy = self.lw * task_cpu_cycle * (1 - percen) * local_comp ** 2

            # 边缘计算时延和能耗
            comp_mec_time = (percen * task_cpu_cycle) / servers_cap[target_server]
            comp_mec_energy = self.ew * percen * task_cpu_cycle * servers_cap[target_server] ** 2

            # 最大计算时延
            comp_time = max(comp_local_time, comp_mec_time)
            time_cost = (comp_time + tr_time) * self.et
            # 能耗成本
            energy_cost = (tr_energy + comp_local_energy + comp_mec_energy) * self.e
            # 总成本
            total_cost = self.lam * time_cost + (1 - self.lam) * energy_cost

            # 2、=======完全本地计算==========
            local_only_time = task_cpu_cycle / (local_comp) * self.et
            local_only_energy = self.lw * task_cpu_cycle * local_comp ** 2 * self.e
            # local_only_energy = task_size * local_comp
            local_only = self.lam * local_only_time + (1 - self.lam) * local_only_energy

            # 3、=======完全边缘计算==========
            mec_only_tr_time = task_size / uplink[target_server] + self.discount * task_size / downlink[target_server]
            mec_only_tr_energy = self.tr_energy * task_size / uplink[
                target_server] + self.discount * self.tr_energy * task_size / downlink[target_server]
            # print("mec_only_tr_time:", mec_only_tr_time)
            # print("mec_only_tr_energy:", mec_only_tr_energy)

            mec_only_comp_time = task_cpu_cycle / servers_cap[target_server]
            mec_only_comp_energy = self.ew * task_cpu_cycle * servers_cap[target_server] ** 2
            # mec_only_comp_energy = task_size * servers_cap[target_server]
            # print("mec_only_comp_time:", mec_only_comp_time)
            # print("mec_only_comp_energy:", mec_only_comp_energy)

            mec_only_time_cost = (mec_only_tr_time + mec_only_comp_time) * self.et
            mec_only_energy_cost = (mec_only_tr_energy + mec_only_comp_energy) * self.e

            mec_only = self.lam * mec_only_time_cost + (1 - self.lam) * mec_only_energy_cost

            # 4、=======随机卸载==========
            percen_ran = np.random.uniform()  # 随机卸载比例
            mec_ran = np.random.randint(self.MECs)  # 随机选择一个服务器进行卸载

            random_tr_time = (percen_ran * task_size) / uplink[mec_ran] + (self.discount * percen_ran * task_size) / 
                             downlink[mec_ran]
            random_tr_energy = (self.tr_energy * percen_ran * task_size) / uplink[mec_ran] + self.discount * (
                        self.tr_energy * percen_ran * task_size) / downlink[mec_ran]

            random_comp_local_time = (1 - percen_ran) * task_cpu_cycle / local_comp
            random_comp_local_energy = self.lw * (1 - percen_ran) * task_cpu_cycle * local_comp ** 2
            # random_comp_local_energy = (1 - percen_ran) * task_size * local_comp

            random_comp_mec_time = percen_ran * task_cpu_cycle / servers_cap[mec_ran]
            random_comp_mec_energy = self.ew * percen_ran * task_cpu_cycle * servers_cap[mec_ran] ** 2
            # random_comp_mec_energy = percen_ran * task_size * servers_cap[mec_ran]

            random_comp_time = max(random_comp_local_time, random_comp_mec_time)
            random_time_cost = (random_comp_time + random_tr_time) * self.et
            random_energy_cost = (random_tr_energy + random_comp_local_energy + random_comp_mec_energy) * self.e

            random_total = self.lam * random_time_cost + (1 - self.lam) * random_energy_cost
            random_total_cost2 = random_energy_cost

            reward = -total_cost

            # 得到下一个observation
            x = np.random.uniform()
            y = 0.5
            if x > y:
                local_comp = min(local_comp + np.random.randint(0, 0.2 * self.UE_f), self.local_core_max)
                for j in range(self.MECs):
                    cap = min(servers_cap[j] + np.random.randint(0, 0.3 * self.UE_f), self.server_core_max)
                    # MEC容量保持一致
                    if new_cap:
                        for x in range(self.UEs):
                            observation[x][2 + j] = cap
                    downlink[j] = min(downlink[j] + np.random.randint(0, 0.2 * self.r), self.downlink_max)
                    uplink[j] = min(uplink[j] + np.random.randint(0, 0.2 * self.r), self.uplink_max)
            else:
                local_comp = max(local_comp + np.random.randint(-0.2 * self.UE_f, 0), self.local_core_min)
                for j in range(self.MECs):
                    # MEC容量保持一致
                    if new_cap:
                        cap = max(servers_cap[j] + np.random.randint(0, 0.3 * self.UE_f), self.server_core_max)
                        for x in range(self.UEs):
                            observation[x][2 + j] = cap
                    downlink[j] = max(downlink[j] - np.random.randint(0, 0.2 * self.r), self.downlink_min)
                    uplink[j] = max(uplink[j] - np.random.randint(0, 0.2 * self.r), self.uplink_min)

            task_size = np.random.randint(10, 50)
            task_cpu_cycle = np.random.randint(10 ** 3, 10 ** 5)  # 处理任务所需要的CPU频率
            observation_ = np.array([task_size, task_cpu_cycle, local_comp])
            observation_ = np.hstack((observation_, servers_cap, uplink, downlink))
            obs_.append(observation_)

            rew.append(reward)
            local.append(local_only)
            mec.append(mec_only)
            ran.append(random_total)

            dpg_times.append(time_cost)
            local_times.append(local_only_time)
            mec_times.append(mec_only_time_cost)
            ran_times.append(random_time_cost)

            dpg_energys.append(energy_cost)
            local_energys.append(local_only_energy)
            mec_energys.append(mec_only_energy_cost)
            ran_energys.append(random_energy_cost)

            total.append(total_cost)

        if is_compared:
            return obs_, rew, local, mec, ran, dpg_times, local_times, mec_times, ran_times, dpg_energys, local_energys, mec_energys, ran_energys, total
        else:
            return obs_, rew, dpg_times, dpg_energys

风语者!平时喜欢研究各种技术,目前在从事后端开发工作,热爱生活、热爱工作。