《算力網路詳解卷1：算網大腦》是2023年清華大學出版社出版的圖書，作者是馬雷明、孫傑、歐陽曄。
基本介紹
- 中文名：算力網路詳解卷1：算網大腦
- 作者：馬雷明、孫傑、歐陽曄
- 出版時間：2023年1月1日
- 出版社：清華大學出版社
- ISBN：9787302623700
- 定價：89 元
內容簡介,圖書目錄,
內容簡介
本書從算力網路的頂層系統設計出發，首先介紹了算力網路的核心功能，然後圍繞算網大腦功能對涉及的關鍵技術實現進行了詳細描述。第 1、2 章介紹算力網路技術發展背景與趨勢，並深入講解算力網路與雲網的關係；第 3 章系統性地介紹算力網路的體系結構；第 4 章介紹與算網大腦有關聯的算網運營交易技術；第 5~9 章詳細講解算網大腦的核心功能：算網一體編排、算網管理調度、算網智慧型運維、算網智慧型引擎和算網數字孿生；第 10 章介紹算力網路套用場景；第11 章對算力網路的發展趨勢做了展望。
圖書目錄
第1章 算力網路概述 ··············1
1.1 算力網路概念······················2
1.1.1 算力網路的定義 ····················2
1.1.2 算力網路與電力網路 ··············3
1.2 新型數位化經濟需求·············4
1.2.1 人工智慧產業化需求 ··············4
1.2.2 分散式多級算力需求 ··············5
1.3 算力網路發展的基礎·············8
1.3.1 算力網路的技術底座 ··············8
1.3.2 “雙碳”社會的國際共識 ·······10
1.3.3 算力網路發展的國家政策 ·······11
1.4 算力網路的新套用形態介紹 ·······12
1.4.1 新型算力組合 ······················12
1.4.2 新型算力模式 ······················12
1.5 算力網路產業規劃···············13
1.5.1 算力網路的標準規劃 ·············13
1.5.2 中國運營商的算力網路規劃 ····15
第2章 算力網路與雲網融合的關係 ·············20
2.1 主流運營商與雲商的雲網融合發展戰略 ·····20
2.2 雲網融合技術特點···············24
2.2.1 雲基礎設施技術 ···················24
2.2.2 網基礎設施技術 ···················28
2.2.3 雲網融合管理編排技術 ··········29
2.3 雲網融合典型場景···············33
2.4 從雲網融合到算力網路·········35
2.5 算力網路目標·····················37
第3章算力網路技術體系 ······40
3.1 算力網路標準體系進展·········40
3.1.1 ITU標準中的算力網路體系框架 ·······40
3.1.2 CCSA標準中算力網路體系架構 ···········43
3.2 亞信科技算力網路系統參考體系··············44
3.2.1 系統功能架構 ······················45
3.2.2 系統接口設計 ······················46
3.3 算網運營交易中心···············50
3.3.1 中心簡介 ····························50
3.3.2 功能介紹 ····························50
3.3.3 集成關係 ····························54
3.4 算力網路編排中心···············54
3.4.1 中心簡介 ····························54
3.4.2 功能介紹 ····························55
3.4.3 集成關係 ····························57
3.5 網路管理調度中心···············57
3.5.1 中心簡介 ····························57
3.5.2 功能介紹 ····························57
3.5.3 集成關係 ····························59
3.6 算力管理調度中心···············59
3.6.1 中心簡介 ····························59
3.6.2 功能介紹 ····························60
3.6.3 集成關係 ····························61
3.7 算力網路智慧型引擎 ··············62
3.7.1 簡介 ··································62
3.7.2 功能介紹 ····························62
3.7.3 集成關係 ····························63
3.8 算力網路數字孿生中心·········64
3.8.1 中心簡介 ····························64
3.8.2 功能介紹 ····························64
3.8.3 集成關係 ····························66
3.9 算力網路基礎設施···············66
3.9.1 網路基礎設施 ······················66
3.9.2 算力基礎設施 ······················68
3.10 業務流程介紹 ···················70
3.10.1 算力網路業務開通 ···············70
3.10.2 算力網路業務變更 ···············72
3.10.3 算力網路業務撤銷 ···············74
3.10.4 算力註冊併網 ·····················75
3.10.5 套用服務部署 ·····················76
3.10.6 套用服務自智保證 ···············77
第4章 算力網路運營交易關鍵技術 ·········79
4.1 多量綱的模型體系催生多形態的算力網路商品 ········79
4.1.1 三要素體系定義 ···················79
4.1.2 多量綱模型體系 ···················81
4.1.3 多形態算力網路商品 ·············81
4.2 多樣化的商業模式促成雙驅動的算力網路受理 ················82
4.2.1 商業模式定義 ······················82
4.2.2 雙驅動算力網路受理 ·············84
4.3 區塊鏈的技術底座實現可信任的算力網路交易 ················85
4.3.1 算力網路存證 ······················86
4.3.2 算力網路通證 ······················87
4.3.3 智慧型契約 ····························88
4.3.4 交易溯源 ····························89
4.4 沉浸式的用戶體驗發掘算力網路的新價值 ······90
第5章算力網路一體編排關鍵技術··················91
5.1 算力網路節點評估···············91
5.1.1 算力網路節點定義 ················91
5.1.2 算力網路節點性能指標 ··········93
5.1.3 算力網路節點資源指標 ··········98
5.1.4 性能與資源關係模型 ···········103
5.2 算力網路融合資源視圖·······107
5.2.1 算力網路鏈路 ····················107
5.2.2 算力節點感知 ····················110
5.2.3 算力網路拓撲 ····················113
5.3 算網融合感知···················114
5.3.1 數據統一採集 ····················115
5.3.2 數據統一存儲 ····················117
5.3.3 數據時間粒度 ····················120
5.3.4 跨域數據關聯 ····················121
5.3.5 數據開放使用 ····················123
5.4 算力網路一體化編排流程····126
5.4.1 業務SLA ··························127
5.4.2 節點編排決策 ····················129
5.4.3 路徑編排決策 ····················133
5.4.4 算網編排方案 ····················135
第6章算力網路管理調度關鍵技術 ··············138
6.1 SRv6網路管理調度 ···········138
6.1.1 技術架構 ··························138
6.1.2 調度機制 ··························140
6.1.3 能力開放 ··························148
6.2 SD-OTN網路管理調度·······150
6.2.1 技術架構 ··························150
6.2.2 能力開放 ··························154
6.3 中心雲管理調度················156
6.3.1 多算力雲管理 ····················157
6.3.2 異構算力資源納管 ··············158
6.3.3 算力資源管理 ····················160
6.3.4 算力資源調度 ····················161
6.4 MEC管理調度 ·················162
6.4.1 MEC組成架構 ···················163
6.4.2 MEC服務 ·························168
6.4.3 MEC能力開放 ···················169
6.5 5G切片管理調度 ··············171
6.5.1 技術簡介 ··························171
6.5.2 切片架構 ··························173
6.5.3 關鍵能力 ··························175
6.6 O-RAN管理調度 ··············180
6.6.1 O-RAN 架構 ······················181
6.6.2 O-RAN 資源控制調度機制 ····182
6.6.3 O-RAN 能力開放 ················183
6.7 5GC網路管理調度 ············185
6.7.1 5GC系統架構 ····················185
6.7.2 5GC 管理調度機制 ··············186
6.7.3 5GC能力開放 ····················189
6.8 算力資源彈性調度·············191
6.8.1 資源適配 ··························191
6.8.2 資源管理 ··························193
6.8.3 套用管理 ··························195
第7章 算力網路智慧型運維關鍵技術 ···············198
7.1 算力網路自智運維管理·······198
7.1.1 算力網路資源監控 ··············198
7.1.2 算力網路運行智慧型預警 ········203
7.1.3 算力網路業務故障自處理 ·····205
7.2 算力網路業務質量保障·······210
7.2.1 端到端保障體系 ·················210
7.2.2 用戶體驗模型 ····················211
7.2.3 服務質量模型 ····················212
第8章算力網路智慧型引擎關鍵技術 ·············214
8.1 算力網路智慧型引擎的基本概念··················214
8.2 算力網路智慧型引擎的系統組件·················215
8.3 算力網路引擎的研究現狀····216
8.4 算力網路智慧型引擎關鍵技術················217
8.4.1 算力網路節點評估算法 ········217
8.4.2 算力網路鏈路評估技術 ········224
8.4.3 算力網路綜合代價函式評估 ·················226
8.5 算力網路智慧型引擎套用場景示例···················228
第9章算力網路數字孿生關鍵技術 ······················231
9.1 套用價值·························231
9.2 問題現狀·························233
9.3 業務目標·························234
9.4 核心算法·························236
9.4.1 分級多層布局 ····················236
9.4.2 混合空間布局 ····················236
9.4.3 縱向多級穿透 ····················237
9.5 關鍵技術·························238
9.5.1 3D地理可視化技術 ·············238
9.5.2 網路拓撲可視化技術 ···········239
9.5.3 遊戲引擎技術 ····················239
9.6 套用場景·························240
第10章 套用案例分析與設計 ······················241
10.1 雲—邊—端算力協同 ········242
10.1.1 AR/VR業務算網方案 ·········242
10.1.2 公安機關嫌疑犯識別 ··········245
10.2 多雲算力協同 ·················248
10.2.1 基於聯邦學習的醫療推薦業務場景 ···················248
10.2.2 “東數西算”數據遷移場景 ······················252
10.3 雲邊算力協同 ·················255
第11章 未來展望 ···············259
