1. To make this driver as efficient as possible, customizations encapsulate an HTTP1.1 client(agVstCli) with connection pooling.
Comparisons between packaged agVstCli and similar HTTP client tests are available:[Address](https://github.com/SisMaker/httpc_bench)
2. This driver can use connection pooling or simply establish multiple connections in a single process (non-connection pooling mode) for various data operations.
Synchronous and asynchronous operations are supported when using connection pooling,
and you need to save the requestId extra if you want to use asynchronous operations Waiting for the received data to return,
the API encapsulated by the current driver all USES synchronous operation, and can be modified if asynchronous operation is needed.
Only synchronous operations are supported for single-process operations.
In single-process mode, compared with connection pooling mode, data replication between processes can be reduced once.
For operation of large amount of data, database connection can be established separately in data management process instead of connection pooling.
3. The connection pooling mode and connectionless pool mode API interface ensures the identity, does not need to be treated differently,
and is easy to understand and change between connection pooling mode and connectionless pool mode.
Efficient, fast and easy to use.
1. This driver can use connection pooling or simply establish multiple connections in a single process (non-connection
pooling mode) for various data operations. Synchronous and asynchronous operations are supported when using
connection pooling, and you need to save the requestId extra if you want to use asynchronous operations Waiting for
the received data to return, the API encapsulated by the current driver all USES synchronous operation, and can be
modified if asynchronous operation is needed. Only synchronous operations are supported for single-process
operations. In single-process mode, compared with connection pooling mode, data replication between processes can be
reduced once. For operation of large amount of data, database connection can be established separately in data
management process instead of connection pooling.
2. The connection pooling mode and connectionless pool mode API interface ensures the identity, does not need to be
treated differently, and is easy to understand and change between connection pooling mode and connectionless pool
Note: If you build Jiffy on The Windows platform, you will need to set up an additional compilation environment. [See jiffy for details](https://github.com/SisMaker/erlUtils/tree/master/src/docs)
