Read "Large Web site Technology architecture: Core Principles and Case studies" chapters v Chapter

Title: Read the "Large Web site Technology architecture: Core Principles and Case analysis," chapters V, combined with the "XXX Demand Collection system", to analyze how to increase the corresponding functions, improve the usability and ease of use of the system, write a page about the blog to explain your point of view

In this lesson, we learned about the usability and usability of the system quality attributes. So what is the quality attribute, the quality attribute is higher than the basic requirement of the system function (i.e., the system ability, the service and the behavior). System quality attributes focus on usability, scalability, performance, security, testability, and ease of use. From the designer aspect, the system mass attribute generally has three questions: (1) The definition provided for the attribute is not operable. (2) Emphasis is usually on which quality attribute a particular aspect belongs to. (3) Each attribute team has developed its own vocabulary.

Today, we will focus on usability and usability learning discussions and combining their methodologies and projects to enhance the usability and ease of use of our actual systems, based on the large Web site Technology architecture: Core principles and case studies.

Let's take a look at usability first. Availability is related to system failures and their associated consequences. A system failure occurs when the system no longer provides the services described in its specification. This type of failure can be observed by the user of the system (person or other system). System availability is the percentage of time that the system is operating normally. System availability is generally defined as: α= average uptime/(average uptime + mean time to repair). There are two well-known examples. April 12, 2011, Amazon Cloud Computing service EC2 failed, its ESB service was unavailable, the failure lasted for several days, and eventually some data failed to recover. This failure has led to the impact of many well-known sites in the United States using Amazon services and has led to large-scale discussions about the security and reliability of using cloud computing. January 12, 2010, Baidu was hacked, its DNS domain name was hijacked, resulting in Baidu's entire station for several hours inaccessible.

Reading These three chapters is mainly about the usability, scalability and extensibility of the site.

High-availability architectures are foolproof. It is almost impossible to ensure that a site is always fully available. We measure the usability of the site through a magical number 9 , using fault points to assess the usability of the site. Usability indicators are important indicators of site architecture design, Web site availability can be seen, touch, and technology, operations, related to the performance of the parties are closely linked. A typical web site design follows the basic layered architecture model i.e. the application layer, the service layer, the data layer. The application layer is responsible for the specific business logic processing, the service layer is responsible for the provision of reusable services, the data layer is responsible for data storage and access. The site's availability architecture is designed to take into account downtime caused by actual hardware failures, as well as downtime due to site upgrade releases. Highly available service policies include tiered management, timeout settings, service demotion (shutting down non-core services), and so on. Highly available data is the most valuable asset, and the means to ensure high availability of data storage are primarily data backup and fail-over conversion mechanisms. Data backup can achieve complete data persistence, the invalidation mechanism is to ensure that the system is available. Ensuring that the site is highly available, foolproof, is a difficult process and requires more effort.

for our system, friendly interface style, simple and convenient operation, reasonable structure layout, good hints are necessary, to improve ease of use is very helpful. The table design needs to provide the user the good visual effect, reduces to the operator to bring the visual fatigue, the operation hint should be easy to understand, uses the accurate appropriate language. Three-level menu implementation algorithm to the time complexity of the standard, display fast, no wait.

the scalability of the site is endless. The so-called scalability of the site, that does not need to change the site's hardware and software design, simply by changing the number of servers deployed to expand or reduce the site's service processing capacity. To achieve the scalability of the Web site, the key technology is how to build a good server cluster. To achieve a good goal, it is required each time the expansion and reduction of the server, the impact on the entire site is minimal. the CAP principle is to choose to harden the availability and scalability of the distributed storage system, and to some extent abandon consistency. the CAP principle is important to the design of scalable distributed systems and improperly caters to a wide variety of needs, which can make design difficult to sustain. Our system has a large amount of statistical data. Our site can be modified at any time, such as the release of new features, it is necessary to close the original application on the server, re-deployment of new applications, the entire process requirements do not affect the user's use. To minimize the impact on users, publishing scripts are typically used to complete a publication. After rigorous testing, software deployment to the server or the problem, the main reason is that the test environment and the online environment is not the same, so when we publish the site, we have to publish the test code to the pre-release machine, verify that the system is not a problem before the official release.

our system is oriented to a wide range of users, through the continuous addition of servers to the cluster to enhance the processing capacity of the entire cluster, this is the scalability of the Web site system architecture, the site to continuously upgrade their size, the evolution of the process is generally gradual. The scalability of the Web site design can be divided into two categories, one is based on the physical separation of functions to achieve scaling, a class is a single function through the cluster to achieve scaling. Where the HTTP request distribution configuration is a load Balancer server, the basic technologies for load Balancing are http redirection,DNS Domain name resolution, reverse proxy, IP, Data link layer load balancing. Scalability and availability, correctness, performance, etc. are coupled together. Scalability is complex, there is no universal, perfect solution and product, but there are many examples in this area for reference. A well-scaled web site designed to always be ahead of the business and ready to handle more access and processing, and when the business needs it, just add servers and simply deploy, and the technical team can easily handle them.

the extensible architecture is changed on demand. Some websites can be released at any time, new features are ready to go online at any time, and some must specify the release date, the reason is dependent on the site's extensibility architecture design. Scalability and scalability are different. In contrast, I found the website I made last semester .... Alas, after learning the MVC model and the three frameworks more feel that they do worse, the so-called ignorance of the more you see yourself, the more widely known the more I feel small is so ....

Read the large web site Technology architecture: Core Principles and Case studies Chapters v Chapter