我可以通过Web.Config连接字符串优化数据库连接吗？

Question

I just want to know is there attribute or property in our connectionstring field so that we can maintain an optimized db connections with our clients. Like in general I use.That contains just required info.

<add name ="ConnectionString" connectionString="Data Source=118.225.221.7;Initial Catalog=mycatalogue;
User Id=dba; password=some_password"/> 

Another Scenario, we can enhance our connection string for MARS and Asynchronous operations too:

MARS
Server=myServerAddress;Database=myDataBase;Trusted_Connection=True;
MultipleActiveResultSets=true;

Asynchronous Processing
Server=myServerAddress;Database=myDataBase;Integrated Security=True;
Asynchronous Processing=True;

From our pages i have 4-5 or sometimes ~10 connections to access data. Take a case when I have 100 users accessing the same page simultaneously assuming the code is executing "SELECT" statement and fetching some random rows for each client. Then 400-500 connections will hit the db at a time, which I think, will definitely degrade the server/application performance.

So I want to know, should I minimize the number of queries, or leave it to the .Net framework intelligence.

Answer 1

Firstly look at optimising your linq, make sure you are not making two queries when all the info can be pulled with one call for instance.

Also look at caching. Here you will have to make the decision on how long the data should be cached for and when it should be invalidated (by someone inserting a record for example) and at what levels you do it. If your page rarely changes and it is not so heavy on insertions/updates you could possible cache at the page level (page level caching).

If that is not an option you can look at caching your data returned from linq using something like httpcontext.current.cache , a quick google search will show you the details, but in essence you cache the data returned from your linq queries, and on subsequent requests you check the cache first to see if the data is there, if it is, use that, if not go to the database and request it again. This limits the hits to your database.