Microsoft Azure Overview

Day1

 

Overview

1. Cloud - Internet scaling / Internet connects / Resource pool

2. Why the cloud?  Rapidly setup environments / Scale to meet peak demands / increase daily activities. (Speed / Scale / Economics)

 

3. cloud computing patterns: on and off/ growing fast / unpredictable bursting / predictable bursting

 

4. cloud computing : IaaS(Host) / PaaS(Build) / SaaS(Consume)

 

Infrastructure as a Service

Platform as a service

Software as a service

 

5. How are Microsoft Azure Charges Incurred?

Pay only for what you use*

VMs usage is by the minute

VMs that are stopped in Microsoft Azure do not incur charges

Microsoft Azure Enterprise Agreement(EA) billing process differs

 

6. Microsoft Azure Compute : App service / MS Azure Cloud service / VMs - IaaS

 

7. How we differentiate with Azure

Hyper-scale / Enterprise Grade / Hybrid

 

8. Azure Site Recovery

 

9. MS Azure Active Directory

 

10. MS Azure Multi-Factor Authentication (多因素認證)

 

11. SQL Database (Day 2 afternoon)

 

12. HDInsight (Hadoop/Hive/Pig etc.)

 

13. Apache Storm for Azure HDInsight : Real-time event processing solution for large, fast steams of data.

 

14. Apache Spark for Azure HDInsight

 

15. Azure Machine Learning: 圖像識別/小冰、小娜等/分析票務信息 --算法 / CRM

Powerbi.com

 

16. Azure Stream Analytics : 智能樓宇/交通流量/物聯網/監控/計算機日誌 (根據時間窗口進行數據分析)

 

17. Azure Data Factory: 數據清洗

 

18. Azure App service: Web apps/ Mobile apps / logic apps(系統整合) /API apps

Azure App service Web Apps: 系統負荷擴展

Azure App Service Mobile Apps: 信息推送

Azure App Service Logic Apps:

https://tryappservice.azure.com

 

Azure Media Services: 視頻培訓(不須要客戶端播放器配置)

 

Internet of things (IOT)

 

Compute Scaling

 

1.Scale UP :縱向擴展

2.Scale Out :水平擴展

 

 

 

Azure Virtual Machine

Azure Network

 

Day2

 

 

微軟公有云和高級企業技術服務介紹

 

微軟存儲很便宜 0.98元/TB

 

Severity : A B C D

A: < 1 Hour

B: < 2 Hour or 4 Hour ( C or D)

 

 

 

三種計算模式：Vm/cloud service / app service

 

PaaS Cloud Service

 

Cloud Service Basic: Web role / Worker Role (A container of related service roles)

What can it run?  Language: C# / Java / VB / C++ / PHP /Python/ Node.js etc.

Framework: .NET / ExpressJS/ Rails/ Zend, etc.

General Rule: If it runs in MS windows, it runs in MS Azure.

 

Roles & Instances : 可修改池裏實例個數

 

Fault domain（故障域） / upgrade domain（更新域）

 

Role Lifecycle : Fabric Calls -- Role Lifetime -- Request Routed

 

Worker Role Patterns

Queue polling worker:

 

性能考慮（硬件增長一倍，性能不必定增長一倍）/ 安全性考慮（壓力測試，獲取系統負荷）/  

 

Paas Database Service

 

Paas Storage

 

Local Storage and Diagnostics

Role Instance:

Role --> Diagnostic monitors <--> Local Directory Storage --> Cloud

 

 

Diagnostics Data Locations

Event logs / Performance counters/ MS Azure logs/ IIS logs/ IIS failed request logs/ crash dumps/ custom file based logs

 

Caching

Fast

Cheaper

 

In-role Caching: Dedicated cache

Azure Redis Cache:  http://redis.io