LINQ 中哪种方法更好？

Question

I am using linq in c# and I have a quick question.

I have shown very minimum amount of code here, in real-time there are also some sorting operations. I would like to know from below which approach should I use??

Approach 1

 public class UserDetails
    {
        private dbContext db = new dbContext();
        public List<User> ActiveUser()
        {
          return db.Users.Where(m => m.Active == true).ToList();                   
        }
        public List<User> InActiveUser()
        {
          return db.Users.Where(m => m.Active == false).ToList();                   
        }
       
    }

Approach 2

 public class UserDetails
    {
       List<Defect> allUser = new db.Users.ToList();
        public List<User> ActiveUser()
        {
          return allUser.Where(m => m.Active == true).ToList();                   
        }
        public List<User> InActiveUser()
        {
          return allUser.Where(m => m.Active == false).ToList();                   
        }
       
    }

There are more than 20 methods which are fetching the data, every method is fetching data from same table with different where condition. My question is Should I create dbContext and then use separate query in every method ( Approch 1 ) OR Should I create one List and fetch all data and filter it in method itself using where condition. ( Approch 2 )

Answer 1

It depends on the circumstances. In general, I would prefer approach 1 as you usually can't predict how many users are in your database and you will pull all users into the memory which will cause a big memory overhead. Also for big amounts of users you will take advantage of database optimisations like indexes and query execution plans.

If we are talking about a small amount of users approach 2 might be the more performant one as you reduce the amount of roundtrips to the database. But next to the memory overhead it also has other issues like caching the users and missing database updates.

However, I'd almost always prefer approach 1 as it's good practice to do most of the filtering and sorting work on the database as it is optimized for doing this kind of things. With approach 2 you might get into trouble as your user base grows over time and it will be hard to track down caching or memory issues. Also the difference between one and two database roundtrips is mostly neglegible if you don't have a really bad connection or do it many times.

Answer 2

Never load unnecessary data. In the Approach 2 you load an unnecessary amount of data.

Answer 3

我个人采用第一种方法，因为在第一种方法中，过滤数据是从数据库中获取的，而在第二种方法中，所有数据都被获取，然后过滤是在框架级别而不是数据库级别完成的，所以在第一种方法中我们可以得到性能优势。