Best practices for designing large services on Windows Azure cloud services

Today's post comes from http://www.aliyun.com/zixun/aggregation/16689.html ">jason Roth, the main programming writer." He provides an overview of the new white paper from our customer consulting team, covering best practices for designing large services on Windows Azure. We have recently released a new white paper: Best practices for designing large services on Windows Azure cloud services. This document brings together the design patterns and guidelines based on the actual customer engagement. It combines the best policies and design patterns that consistently demonstrate the success of real-world Windows Azure applications.

The first thing to know about this platform

When you read this white paper, you will note that there are three main sections:

Design concepts
Explore Windows Azure
Best practices
You may be tempted to browse best practices directly, but you should be aware that these best practices derive from the information in the first two sections. Each application is unique. It is important to first understand the Windows Azure platform and basic design principles. This is useful for choosing the right optimizations and for implementing the correct execution.

Good design--worth the effort

Any large application design requires careful thinking, planning, and potentially complex execution. The basic design principle for Windows Azure is expansion. Instead of investing in powerful (and expensive) hardware, the extension strategy responds to increasing demand by adding more machines or service instances.

For each Windows Azure service, many of the best practices involve implementing extensions. For example, in Windows Azure, you cannot magnify a server running on your SQL database. Instead, you must design an application to take advantage of an additional SQL database instance. This involves some type of partitioning strategy for your data.

The challenge, of course, is to choose the right partitioning strategy and successfully coordinate the work between partitions. This article attempts to provide you with a technical understanding of your choices and practical suggestions for your past client programs.

Note that partitioning improves scalability, and SQL databases are just obvious examples. But to maximize the advantages of the platform, other roles and services must be expanded in a similar way. For example, the transaction rate of a storage account is capped, the CPU and memory of the virtual machine are capped, and the maximum limit is achieved through the use of multiple storage accounts and the expansion of the component to the virtual machine setting size service.

While scalability is the driving force behind the design, there are other important design considerations. The document emphasizes the need to plan for telemetry and diagnostic data collection, which is becoming increasingly important as your solutions become more modular and partitioned. Availability and business continuity are the two other major focuses of this article. Extensibility is irrelevant when your service fails or inevitably loses data.

Best Practices and Platform evolution

Windows Azure evolves, improves, and adds new services. In recent releases, new features have been added, such as Windows Azure virtual network and infrastructure as a service (IaaS). These new features provide more choices for large applications. However, this document focuses on version 1.6 and does not include some of the latest features added to the platform.

To understand the reasons for this decision, you must re-examine the objectives of this work. This article will provide successful design guidance in real customer practice. Because these services may take months of planning, testing, and looping, it will take some time before this article updates the latest services and features. However, the design principles in all files still apply, and the same type of thinking can be applied to any new features of Windows Azure.

Continue, we are working on additional files, code samples, and instructions to demonstrate how to perform some of the best practices.

Not a checklist.

Everyone likes to check the table. The idea is: if you can select all the marquee, then you know that you will succeed. With the best practices for designing large services on Windows Azure cloud services, do not view this information as a checklist. Your application is unique. Perhaps, at this moment, your application is "medium-high" in size. It's possible that when you understand the platform and best practices, only a few suggestions are critical to you in a short time. But looking ahead, planning you will need some or all of the possibilities of other design strategies.