Skip to content

GCRI · Foundry

算力与边缘

Risk analysis needs compute in very different places: a large cloud allocation for a national ensemble, a ministry’s own servers for records that may not leave, a field laptop that keeps a model running when the network fails. This page sets out how compute risk develops, from capacity shortfall and concentration in a few providers to jurisdiction, energy and degraded connectivity, and how cloud, local nodes and edge devices are designed to be scheduled as one capacity.

01登记

算力与边缘计算记录在案。

02问题

算力与边缘:风险如何形成。

云端、本地节点和边缘的技术能力。下方档案说明风险如何发展、衡量什么、数据来自何处、如何建模以及由谁决定。

机制

需要时容量失效

算力需求在事件期间达到峰值,而此时恰是共享容量被争用、网络受损、电力不稳之际。队列变长,集合缩小,预报在决策窗口关闭后才到达。

机制

集中度

Many institutions rely on a few cloud providers, regions and chip suppliers. An outage, sanction, price change or supply shortage reaches all of them at once, which makes compute concentration a systemic dependency in its own right.

机制

司法管辖区与数据所在地

Where data may be processed is set by law and contract. Moving analysis to the data keeps records in place, but constrains which methods can run and can still leave results open to inference about the inputs.

机制

断开连接与对账

Edge nodes keep working offline and reconcile later. Conflicting edits, clock skew and stale models accumulate during an outage, and reconciliation after a long disconnection may need human judgement rather than an automatic merge.

03方法与证据

如何对计算与边缘计算进行建模、测量和测试。

模型、其假设、数据和不确定性,以科学家或工程师会核查的形式呈现。结构遵循已发表的方法。所取数值仅为示意,不是研究发现。

模型、方程与假设

用于算力和边缘的模型结构、其所作假设以及各项假设如何检验。

要素 i 的风险Ri = f(Hi, Ei, Vi, Ci)危害、风险暴露、脆弱性和应对能力
预期年度损失EAL = ∑k pk · Lkover events k with annual probability pk
耦合系统xt+1 = A xt + B ut + εtA carries the couplings, u the shock, ε the error
假设为何制作如何检验
在一次事件中,危害与暴露相互独立使损失积分保持可计算与基于联合记录拟合的 copula 进行比较
耦合在工作点附近是线性的让冲击通过 A 传播在线性范围之外施加扰动并比较
脆弱性曲线可以在不同地点之间迁移使用很少有本地曲线可用在有观测损失的地方据此进行验证
输入
带溯源信息的版本化数据集
成果
分布,绝不是单一数字
代码
连同模型卡和测试一起发布
复现
相同的数据、代码、环境和随机种子
标准的风险与状态空间表述;所列假设是该方法的典型假设,供审查之用,并非结果。
FAIR 指导原则 · GO FAIR 和 FORCE11W3C PROV · 万维网联盟关于一致处理不确定性的指南 · IPCCCEOS 分析就绪数据 · 地球观测卫星委员会
04在五个领域中

Compute is designed to serve each domain’s heaviest models.

05指标

衡量什么,以及衡量得如何。

每个指标都注明其单位、频率、背后的来源类型以及数值将附带的证据标签。此处不显示数值:数值由计划依据其自身权限提供。

指标单位节奏来源类型标签
Queue wait timeminutes, median and tail每日Scheduler accounting观测
服务可用性占计划时间的比例每月可用性监测观测
部署在许可司法管辖区之外的工作负载计数per placementPlacement policy logs观测
Edge reconciliation conflicts每次重新同步的计数每个事件同步日志观测
每项工作负载的能耗kilowatt-hours每次运行电力遥测数据或供应方估算估算
Carbon intensity of electricity usedgrams CO2 equivalent per kilowatt-hour每小时Grid operator and provider data估算
提供的容量与分配的容量对比核时或 GPU 时每月Offer and allocation register已声明
Recovery time in failover exercises分钟每次演练演练记录观测
06来源

证据来自何处。

具名公共数据集与机构记录,各自注明许可类别。依协议获得的来源仅按其条款使用,绝不再次发布。

来源它提供什么许可类别
TOP500 和 Green500 榜单已公布的大型 系统Free, terms of the project
Ember 电力数据各国发电结构和电力碳强度CC BY 4.0
Copernicus Data Space EcosystemEarth observation archives with processing next to the data免费开放,须署名
NASA Earthdata Cloud云对象存储中的地球科学档案,用于区域内分析Free and open, NASA data policy
公有云区域和状态记录区域位置、可用性历史和事件报告提供方条款
Kubernetes 发布和弃用记录Version support windows and API removals that affect workload portabilityCC BY 4.0
07决定

由谁决定,以及证据服务于什么。

Foundry 负责准备证据。以下决定由所列机构依其自身规则作出,绝不由平台作出。

谁来决定问题本页面服务的证据
数据所有者和数据保护机构数据可在何处处理,以及结果是否可以出境Placement logs, the jurisdiction of each tier, federated execution records
项目负责人事件期间如何在分配额度内确定工作负载的优先次序队列状态、剩余配额、截止日期
基础设施贡献方和运营方What capacity to offer and on what terms供给登记册、利用率、能源和支持负荷
Ministries responsible for digital infrastructure国家 风险分析 在多大程度上依赖外国或集中的提供方服务提供方与区域依赖图;退出与故障切换计划

该主题产生的产出

  • 按数据类别、司法管辖区、紧迫性和成本确定的放置政策
  • Workload packages portable across cloud, private and edge tiers
  • 带演练记录的故障切换和降级运行手册
  • 已提供、已分配和已使用容量的登记册,分别保存
08核心问题

计算工程师和数据所有者提出的问题。

  • Where is the data legally and contractually allowed to be processed?
  • Which workloads need burst capacity, and which need steady, local availability?
  • 在连接性下降或中断时,工作如何继续?
  • 如何区分提供的承保能力与实际分配和使用的承保能力?
09机制

横跨三个层级的统一控制平面。

NXS-COR (compute, nodes and runtimes) is designed as the compute control plane across cloud, private infrastructure and edge. Workloads are packaged with their dependencies so they can run on any tier that the project’s policy allows, and the scheduler places them according to data location, sensitivity, cost and urgency. Where data cannot move, the analysis is designed to run where the data is held and only permitted results leave.

Edge capacity matters most when it is hardest to provide: during the events risk work is about. Field kits, observatory nodes managed through NXS-OPS (instruments, sensors and calibration) and offline operation of NXS-NST (civilian common operating picture) are designed so that coordination continues through outages, synchronising when connectivity returns. Energy use, cooling and hardware supply are part of the engineering, and the platform’s computing questions connect to the energy and data-centre analysis at Data and Compute.

10方法与局限

哪些已经工程化,哪些依然困难。

Federation

Compute to data

Federated execution keeps sensitive data in place and moves code and permitted results.

韧性

降级运行

边缘节点在本地继续运行,随后再进行同步,冲突会被标记,而不会被悄然覆盖。

Accounting

Offer is not allocation

Contributed infrastructure is recorded as an offer until it is separately allocated.

局限:联邦分析限制了可运行的方法,边缘设备能力有限,长时间断连后的对账可能需要人工判断。不将各项硬件规格汇总为性能声明。

11方法

如何研究,以及方法无法说明的内容。

  1. 高性能计算、云计算与边缘计算跨层级的集合调度 · 贴近数据的联合执行在此应用
  2. 系统工程规范系统清单 · 契约优先的接口 · 可追溯至证据的需求在此应用
  3. 软件与安全关键系统工程失败即关闭的策略关口 · 构建来源追溯与物料清单 · 保证论证在此应用
  4. 数据工程、互操作性与数据管护导入边界处的验证 · 用于可复现产品的冻结数据切片在此应用
  5. 地理信息系统通过 STAC 的影像目录 · 注明汇总方式的风险暴露叠加在此应用
12Where this page’s references

Where this page’s references lead.

阅读本页时所对照的机构和文件。状态和版本会变化;每个链接都指向发布机构。

13问题

问题与解答。

Why move analysis to the data instead of the reverse?
Because some data may not legally or contractually leave its location, and some is too large to move. The code goes to the data, runs under the owner’s policy, and only permitted results come back.
What happens to edge work during an outage?
节点使用其持有的模型和数据在本地继续运行。连接恢复后,它会同步并标记冲突,交由人员处理,而不是覆盖任何一方。
我的机构可以提供服务器吗?
Offers can be described to the Consortium Secretariat. An offer is recorded as an offer, not as allocated or used capacity, until a separate allocation is agreed.
是否报告能源使用?
The design records energy and carbon per workload where telemetry or provider estimates exist, with a label stating which. Estimates are never presented as measurements.
Does membership include compute?
不会。会员资格和赞助均不能确保获得算力配额。
是否有开放的算力分配?
尚未。待条款和审查能力具备后,将开放分配。
报告贡献

从本页贡献证据。

Reports 是联合体的季刊。您对该主题的了解可以成为下一期中的证据、案例或更正,署名归您,并在属于您本人作品时以 CC BY 4.0 发布。

GCRI区域任务最常用的 FoundryFoundry 构建与准备度
区域任务最常用的 Foundry

带来一种方法、一个数据集或一个问题。

信息和咨询现已开放。每项咨询都会获得一个编号,并告知下一步;申请、付款、算力分配和注册只有在相应团队、条款和审查能力就绪时才会开放。

The Global Centre for Risk and Innovation (GCRI Canada) is a Canadian not-for-profit organisation, not a registered charity, with its registered office in Ontario.