The graphical database neo4j has the following advantages

When performance or other problems occur, see performance counters are a good dimension to speculate on the possible causes of the problem, and then reduce the range of issues to be considered, so the periodic collection of counters for each server will make the problem traceable. And the collected data can also be used as baseline, even if there is no problem can prejudge some problems.

In the original Laim has already had the skin-changing function, and in the skin configuration, you can add your own desired skins picture path. These things will not be involved in the next, this article is about the custom skin function, it is true that users upload their own desired skin. and can be replaced at any time. The ritual, the effect shows.

We need to start counting when the user enters the URL or clicks the link, as this will measure the user's wait time. High-end browser navigation timing interface, ordinary browser through the cookie record timestamp way to statistics, need to note that the cookie method can only be counted in the arrival of data jump.

In MySQL, a concern relationship (Big V ID, a fan ID of Big V) is saved as a data, then when the number of users up, focus on the relationship easily broken billion, broken 1 billion, even tens of billions, and in order to ensure the uniqueness of each data, but also need to set up a federated index, MySQL is a bit out of the way. Then somebody has to say: a table. Well, yes, the sub-table does raise some speed at the insertion and reading ends. For example, we can hash to 100 tables based on the ID. It is quick to query which fans a user has, but it is still necessary to traverse the entire table when querying a user who is concerned about which person. Well, we can also construct 100 tables with (ID, one user's ID of interest), so both queries are fast. However, the back of the 100 tables is redundant data, looking at the uncomfortable ... It is also inconvenient to generate a sub-graph (you need to write the SQL check table multiple times).

When I recently searched the web for Python and WMI, I found that most of the articles were stereotyped and basically only used in a very basic way, and did not explain in depth how to use WMI. This article is going to go a step further and let's use Python to play WMI. It also provides a very convenient, locally developed environment. It is not easy to find a detailed and complete tutorial on the Internet, this article combines practical experience to summarize a set of available development and on-line configurations and processes.

To host more users for the server? Improve the responsiveness of your website? Share the pressure on the database server? Just for the dual-machine hot spare and don't want to waste the backup server? The above answers, I think, are not wrong, but they are not entirely correct. The "Read and write separation" is not much of an amazing thing, and does not bring much performance improvement, perhaps more of the role of data security backup it.

From a library to read-write separation, theoretically to the server pressure will bring a performance improvement, but you think carefully, your application server really need this one-fold promotion? Why don't you try it? Using caching systems such as Memcached, Redis, and other distributed caches on the server can be a dozens of times-fold increase in performance. And, in the server hardware is unusually strong and the performance of cheap today, it is completely unnecessary, so, today, I think it is more responsibility for data security and design, but also improve some performance, so it is very good.

In my previous blog post: Netty constructs a distributed Message Queuing (AVATARMQ) design Guide in the architecture, focusing on the main components of AVATARMQ and the existing advantages and disadvantages. Finally, a producer and consumer example is given to demonstrate the basic message routing capabilities of AVATARMQ. The purpose of this paper is to make a simple analysis and explanation of the technical details and principles behind the distributed Message queue, from the angle of development and design, and simply to how to use Netty.

Here the two layer refers to the computer network seven layer model two layer, from the first layer to the 7th layer are the physical layer, Data link layer, network layer, transport layer, Session layer, presentation layer and application layer. Another is the 4-layer (or 5-layer) network model, which is the data link layer, the network layer, the transport layer and the application layer, and if you add the physical layer according to the 5-tier theory. The two layers here refer to the Data link layer.

Many written interviews like to examine the fast line, so that you write a handwriting is not what. I learned this very early, and the process of fast sequencing is very clear. But the recent attempt to hand-write, found that the details of the algorithm is not accurate enough, many places even just an image in the brain, and did not understand its true nature intentions. So today combined with the "Data Structure" (Min), and "Introduction to the algorithm" to explore.

Write react also has a period of time, has been also used Redux management data flow, recently just have time to analyze the source, on the one hand, I hope to have some theoretical understanding of redux, on the other hand, also learn the framework of the way of programming thinking.

Last time did a help the company sister did a reptile, not very delicate, this company project to use, so have done some modification, the function added the URL image collection, download, threading interface URL image download.

Talk about the idea: The Prime Minister gets all the contents of the initial URL at the initial URL to capture the image to the initial URL collection link to put the collected links into the queue to continue to collect pictures, and then continue to collect links, Infinite loop

The graphical database neo4j has the following advantages