Casual Thoughts On Distributed RPC Service Framework
fujohnwang
2011-09-07
author: fujohnwang
service identity
[artifact version] + FQname + version
service identity
[artifact version] + FQname + version
service lookup
- how many services exist?
- what kind of details operations/services in each service?!
- what’s the signature of each operations?
service extensions
- can I replace the data structure serialization and deserialization?!
- use custom serialization mechanism
- some flag to indicate the mechanism that’s used by service providers
- can I intercept service operations without service implementations’ intrusion?!
- can I replace default service lookup service provider?
- can I hook in custom LB and failover mechanisms in the service users’ side?
- can I hook in audit and monitoring concerns?
- can I tune the server side of the service exposion?
- TcpNoDelay?!
- send buffer or receive buffer size
- timeout
- etc.
- can I expose same service via different transports at the same time?!
service upgrade
- service framework upgrade
- service upgrade
service deployment
- how to ease the large scale service deployment?! even make it automatically?
- template publication node?!
potential points
- are overloaded methods allowed?!
- are multiple languages inter-operations supported?!
- If you want to provide OO-RPC interface for clients to use, what kind of interception mechanisms u want to use? reflection or bytecode gen? Are there some overheads in this point?
开天窗，拉认知，订阅「福报」，即刻拥有自己的全模态人工智能。
