什么是思科 ACI

原创 Luke 网工笔记本
和大学同学在聊天时候发现,即使大家都是从网络工程专业毕业,有着相似的背景,不过由于工作经验差别较大,在讨论问题时候时常也会先问一些基础的问题,比如说:什么是思科 ACI ?

之前写过不少关于 ACI 的技术分享,不过确实是忘了写这个基础的信息介绍。为了更好的解释 ACI,我整理了一些信息仅供大家参考。

中英文对照,看不懂中文时候,读英文文档反而更容易理解。

(hardware based) SDN
Cisco Application Centric Infrastructure (ACI) is Cisco’s software-defined networking (SDN) solution for data centers. It aims to simplify, optimize, and accelerate the deployment and management of applications in a highly scalable and agile network environment. ACI achieves this by providing a holistic architecture with centralized automation and policy-driven application profiles.

思科以应用为中心的基础设施 (ACI) 是思科面向数据中心的软件定义网络 (SDN) 解决方案。市场定位是“简化、优化和加速在高度可扩展和敏捷的网络环境中部署和管理应用”。

ACI 通过提供具有集中自动化 (centralized automation) 和策略驱动型应用程序 (policy-driven application profiles) 配置文件的整体架构来实现这一目标。

思科 ACI 的亮点:
Centralized Management and Automation:

集中管理和自动化:

ACI uses the Application Policy Infrastructure Controller (APIC) as the centralized management console. APIC enables network administrators to automate and centrally manage the entire data center network.

ACI 使用应用程序策略基础结构控制器 (APIC) 作为集中式管理控制台。APIC 使网络管理员能够自动化和集中管理整个数据中心网络。

从实际使用体验来说,图形化管理很友好,并且应该是未来的趋势。单纯的徒手敲命令行已经无法很好的管理大规模的网络设备了。

Policy-Driven Architecture:

策略驱动架构:

ACI allows the definition of policies that specify how applications interact with the network. These policies are abstracted from the underlying hardware, enabling easier management and consistency across different environments.

ACI 允许定义指定应用程序如何与网络交互的策略。这些策略是从底层硬件中抽象出来的,从而可以更轻松地跨不同环境进行管理和一致性。 从实际应用来看,这部分可以参考思科 ACI 的 managed objects

Scalability and Flexibility:

可扩展性和灵活性:

The ACI architecture can scale to accommodate thousands of devices and millions of endpoints. It supports a wide range of physical and virtual devices, providing flexibility in deployment.

ACI 架构可以扩展以容纳数千台设备和数百万个端点。它支持各种物理和虚拟设备,提供部署灵活性。

ACI 能够支持数百万个 endpoint 的原因,可以从 ACI Endpoint Manager / EPM 介绍 里面了解一下,ACI 是如何学习/同步 endpoints,如何节省 TCAM 空间。

Security and Micro-Segmentation:

安全和微分段:

ACI provides built-in security features, including micro-segmentation, which allows fine-grained control over network traffic between application components. This enhances security by isolating workloads and limiting the attack surface.

ACI 提供内置的安全功能,包括微分段,允许对应用程序组件之间的网络流量进行细粒度控制。这通过隔离工作负载和限制攻击面来增强安全性。

ACI 要求不同 EPG 之间通信必须配置 contract,同时对于相同 EPG 的 endpoints 如果需要隔离,也可以打开 Intra-EPG contract 功能。

Multi-Cloud and Hybrid Cloud Support:

多云和混合云支持:

ACI extends its policy-driven model to multi-cloud and hybrid cloud environments. This ensures consistent networking and security policies across on-premises data centers and public cloud services.

ACI 将其策略驱动型模型扩展到多云和混合云环境。这可确保跨本地数据中心和公有云服务实现一致的网络和安全策略。

Simplified Troubleshooting and Monitoring:

简化的故障排除和监控:

With tools like Cisco Network Assurance Engine and Telemetry, ACI provides advanced monitoring, troubleshooting, and analytics capabilities. These tools help in identifying and resolving issues proactively.

借助思科网络感知引擎和遥测等工具,ACI 可提供高级监控、故障排除和分析功能。这些工具有助于主动识别和解决问题。

Integration with DevOps Tools:

与 DevOps 工具集成:

ACI integrates with various DevOps tools and platforms, such as Kubernetes, OpenStack, and VMware. This enables seamless deployment and management of containerized and virtualized workloads.

ACI 与各种 DevOps 工具和平台集成,例如 Kubernetes、OpenStack 和 VMware。这样可以无缝部署和管理容器化和虚拟化工作负载。

Enhanced Performance and Efficiency:

增强的性能和效率:

The architecture of ACI optimizes network performance by dynamically adjusting to the needs of applications. This results in improved resource utilization and lower operational costs.

ACI 的架构通过动态调整应用需求来优化网络性能。这样可以提高资源利用率并降低运营成本。

Support for Open Standards:

支持开放标准:

ACI is built on open standards, which ensures interoperability with third-party solutions and protects against vendor lock-in. This provides flexibility in choosing the best-of-breed components for the network.

ACI 建立在开放标准之上,可确保与第三方解决方案的互操作性,并防止供应商锁定。这为为网络选择同类最佳组件提供了灵活性。

300-610:Designing Cisco Data Center Infrastructure (DCID)

300-610:Designing Cisco Data Center Infrastructure (DCID)
设计思科数据中心基础设施
时长:90 分钟
语言:英语
价格:300 美元
考試概覽
相關認證:
CCNP 數據中心
思科認證專家 – 數據中心設計
該考試測試您對數據中心基礎設施設計的了解,包括:

網絡
計算
存儲網絡
自動化
2019 Cisco Systems, Inc. This document is Cisco Public. Page 1
Designing Cisco Data Center Infrastructure v1.0 (300-610)
Exam Description: Designing Cisco Data Center Infrastructure v1.0 (DCID 300-610) is a 90-minute exam
associated with the CCNP Data Center Certification. This exam certifies a candidate’s knowledge of data
center infrastructure design including network, compute, storage network, and automation. The course,
Designing Cisco Data Center Infrastructure, helps candidates to prepare for this exam.
The following topics are general guidelines for the content likely to be included on the exam. However,
other related topics may also appear on any specific delivery of the exam. To better reflect the contents
of the exam and for clarity purposes, the guidelines below may change at any time without notice.
35% 1.0 Network Design
1.1 Evaluate options for Layer 2 connectivity
1.1.a Endpoint mobility
1.1.b Redundancy/high availability
1.1.c Convergence
1.1.d Services insertion
1.2 Evaluate options for Layer 3 connectivity
1.2.a IP mobility
1.2.b Redundancy / high availability (graceful restart, NSF)
1.2.c Convergence
1.2.d Services insertion (load balancing, security)
1.3 Evaluate data center technologies such as vPC/vPC+
1.4 Evaluate options for interconnecting data centers such as VXLAN EVPN and OTV
1.5 Evaluate options for device and routing virtualization
1.5.a VDC
1.5.b VRF (standard, VXLAN EVPN tenant)
1.6 Evaluate in-band and out-of-band options for management
1.7 Evaluate redundancy options such as active/active and disaster recovery site
25% 2.0 Compute Design
2.1 Evaluate options for Ethernet connectivity
2.1.a Redundancy / high availability
2.1.b Bandwidth (oversubscription)
2.1.c Fabric interconnect operation mode (switch mode, end host mode)
2.2 Evaluate options for storage connectivity
2019 Cisco Systems, Inc. This document is Cisco Public. Page 2
2.2.a Bandwidth (port-channels, oversubscription)
2.2.b Fabric interconnect operation mode (switch mode, end host mode)
2.2.c Direct-attached storage (appliance, FC storage, and FCoE ports)
2.3 Evaluate options for network device virtualization in a data center (Cisco VIC adapters)
2.3.a Number of interfaces vs. IOM uplinks
2.3.b vCon placement policies
2.3.c Ethernet adaptor policies
2.3.d Fibre Channel adapter policies
2.4 Evaluate options for hyperconverged infrastructure
2.4.a Cluster mode
2.4.b HX for virtual server
2.4.c Desktop virtualization
20% 3.0 Storage Network Design
3.1 Plan for iSCSI deployment in a data center (Multipathing and addressing schemes)
3.2 Evaluate QoS requirements in a data center
3.2.a Fibre Channel
3.2.b FCoE
3.2.c FCIP
3.2.d iSCSI
3.3 Determine FCoE/ Fibre Channel interface
3.3.a Dedicated and shared mode
3.3.b Port types
3.3.c ISL
3.3.d Oversubscription
3.4 Evaluate SAN topology options
20% 4.0 Automation Design
4.1 Evaluate options for network orchestration and automation
4.1.a DCNM
4.1.b Intersight
4.1.c NX-API
4.1.d Model-driven programmability
4.1.e Ansible
4.1.f Puppet
4.1.g Python
4.2 Evaluate options for compute orchestration and automation
4.2.a Service profile templates
4.2.b vNIC templates
4.2.c vHBA templates
4.2.d Global policies vs. local policies