当每秒有400-500个请求时，使用“ DatabaseAsync”通过c＃进行Redis异步调用

Question

when I use:

`public call(Func<IDatabase, Task<T>> func){
  var task=func(redisInstance); 
  task.wait();
  return task.Result;
}`

call: call(client => redisInstance.SetMembersAsync(setName))

It works Fine, but when there are many requests 'CPU' usage increases;

but in second case:

`public async call2(Func<IDatabase, Task<T>> func){
   var task=func(redisInstance); 
   return await task.Result;
}`

call: call2(async client => await redisInstance.SetMembersAsync(setName))

CPU is ok but responses take much time;

any Ideas about this case? what is wrong in second method?

Answer 1

There are 2 things I think you could do to improve the wall-clock time speed under load.

1. There are 2 await s, and this is because you have double wrapped the result, so it's now Task<Task<T>> . Change your call to:

    call2(client => redisInstance.SetMembersAsync(setName)) 

2. You could optimize by removing the implicit synchronization of the continuation after awaiting, by using .ConfigureAwait(false) , so change your await to:

    return await task.ConfigureAwait(false);