Database hosting issues, MySQL hosting issues

The 3 integers of only 200w are definitely no problem, and they don't hit without tens of millions of influences.

20w pieces of data are like no data, it's really fine.

You can write your own tool to insert data for testing.

I have a mysql table with 30 million data, and I search the data by index, and I don't find any performance problems when I insert or delete the data.

Depending on your situation, MySQL is up to the task!

All using the MyISAM storage engine, you can limit the size by modifying the MySQL configuration file, and then optimize the corresponding SQL statements accordingly!

I guess not, let's change to oracle, MySQL is just small data, and you do it very much to affect efficiency.

I don't think 10,000 data concurrency is large, and it is no problem to think that oracle database and mysql carry these concurrency.

I think the main problem is that your GPS is constantly changing, because the car is constantly moving, which can affect the performance of the database.

I think, you can use a memory line database, for example, Redis, use this to store GPS information, Redis is memory-based, read and write faster than relational databases (ignoring network factors), you may have to ask how to get GPS storage, you can do a scheduled task, how much time to write Redis data to the database, of course, Redis also supports some algorithms, such as LRU, to set when to synchronize data to the database.

If you use normal database operations, it will definitely become unsuitable when the amount of data becomes larger in the later stage. The first is real-time, the second is high concurrency, and the third is the large amount of data.

About 100 people, look at the condition of the server, if it's very good, it should be possible to regret hundreds of people, and I saw on the Internet that the hall pants can withstand 900 people, I don't know if it's true or not, you can find a stress test tool to test it.

Theoretically, there is no limit to access, but the speed of the file size will be affected, and there should be a special tool to transfer.

I want to ,,, the finch skin next to ,,, core. Change the year of oak...

I haven't used this mode before, how much concurrency does MySQL support?

If it doesn't exceed at the same time, it should be able to fill the number of pickpockets, but I always feel that this is purely a storage process, which is more novel than this.

MySQL can support the concurrency.

Big or small, not by the amount of text on the surface or the frame of the page.

It's not to see who says it's big, it's big, it's small...

Rather, it is the size of the development cycle and technical content.

Let's say you're just trying to add a small box that you think is small.

But this box you need to call the data of the Bank of China, then this small box is very large.

There are programs that already have secondary development to add things.

If you can't do it yourself, the cost of finding someone else to do it is much higher than if someone else does it yourself.

Because people have to re-familiarize themselves with the complete unfamiliar when they are not familiar with most of the content.

Do your own thing and do someone else's business yourself.

This should be understood.

The specific amount of money depends on the specific situation and specific analysis.

Only with details can we know the cost, and only with the cost can we know the profit ratio and the final **.

In the case of not knowing what you are doing in the wild shirt, you can come up with the final ** without knowing the cost.

Pure ... Gao Chun.

I'm glad that the town waiter will answer for you big or small.,It's not about the amount of surface text or page frame.,It's not about who says it's big.,Big and small.。。。 But the size of the development cycle and technical content, for example, you just add a small box for the loss, you think it is very small, but this box you need to call the data of the Bank of China, then this small box is very large, the existing program secondary development, add things, if you can't do it yourself, find someone else to do it, the cost is much higher than others to do your own station, because people are not familiar with most of the content, completely unfamiliar with doing your own thing, and you yourself do other people's business, this should be understood, the specific amount of money depends on the specific situation Specific analysis has details to know the cost, and there is a quiet noise cost to know the profit ratio and the final **.

Hope it helps

Questions. Is it that 5000 members can't work properly? Yes.

You write a script, and the script is mainly a loop, and the things in the loop include:

1. Connect to the database.

2. Sleep in the database for a period of time.

Then put the number of loops to keep increasing. Until an error is reported. Something like can't be connected, at this time, it means that you can't connect anymore.

Of course, the actual number of connections will be much smaller than this. Because the connection of this test didn't do anything after all.

But along the way, you can test the CPU utilization as well as the memory utilization.

Put the 7 programs in a ** root directory, put the data path connection file conn in the root directory, and then merge the tables of each database into a database, this database file is placed in a folder in the root directory, and then modify the conn files of each program so that they all point to the merged database address, so that you can. If there is the same table name, it is more troublesome, in addition to modifying access when merging, the program also needs to be modified, and that kind of project is huge.

You have also been in this situation, I just finished correcting, you said that the 7 programs have a total of multiple tables, before merging a database, it is recommended that you have a detailed table description, to facilitate the operation of merging databases. Because ** is a membership system, you must pay attention to the authority when doing the background! It's not difficult to merge databases, the key is the program!

Then change the database, there is no good way.