Author:
Wang Xingwei,Zhao Hong,Zhu Jiakeng
Abstract
RPC (Remote Procedure Call) is very popular in distributed application design. However, most existing RPC systems have some defects on parallelism and reliability and transparency, they can only support point-to-point and(or) broadcast communication. We try to bring the process group concept into RPC systems and present a new kind of communication cooperation mechanism supporting the cooperation among distributed entities in distributed systems. We call it GRPC(Group Remote Procedure Call) mechanism. According to different application background, the related GRPC mechanism are named
lookup, function-convergence
and
update
GRPC. The prototype systems of these three specific classes of GRPC mechanisms have been implemented in the local area network which consists of SUN 4 workstations. The performance evaluation on these prototype systems has been made. Practice has shown: the GRPC mechanism can not only improve the parallelism, reliability and transparency of RPC systems, but also enable RPC systems to support group communication.
Publisher
Association for Computing Machinery (ACM)
Reference10 articles.
1. Implementing remote procedure calls
2. [
4
] Luping Ling Samuel T. Channon and Gerald W. Neufeld Process Group and Group Communications: Classifications and Requirements COMPUTER. Feb. 1990. p. 55-66. 10.1109/2.44904 [4] Luping Ling Samuel T. Channon and Gerald W. Neufeld Process Group and Group Communications: Classifications and Requirements COMPUTER . Feb. 1990. p. 55-66. 10.1109/2.44904
3. [
5
] Sape J. Mullender. Distributed Systems Addison-Wesley Publications 1989. [5] Sape J. Mullender. Distributed Systems Addison-Wesley Publications 1989.
Cited by
27 articles.
订阅此论文施引文献
订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献
1. An efficient cloud-integrated distributed deep neural network framework for IoT malware classification;Future Generation Computer Systems;2024-08
2. DUST: Resource-Aware Telemetry Offloading with A Distributed Hardware-Agnostic Approach;2024 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW);2024-05-27
3. FedRDMA;Proceedings of the 4th Workshop on Machine Learning and Systems;2024-04-22
4. The Next 700 ML-Enabled Compiler Optimizations;Proceedings of the 33rd ACM SIGPLAN International Conference on Compiler Construction;2024-02-17
5. Benchmarking Federated Learning Frameworks for Medical Imaging Tasks;Lecture Notes in Computer Science;2024