.png){kind=icon}
返回
Remote Method Delegation: a Platform for Grid Computing
.png){kind=icon}
PRE
AI
DOI：10.1007/s10723-020-09525-z.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
While many cluster and grid computing frameworks are available, the task of building secure distributed systems or implementing distributed algorithms continue to be a challenging task due to the inherent distributed nature of such systems with multiple failure modes and security issues. In this paper, we present the design and development of remote method delegation (RMD), which is a secure lightweight grid computing platform with load balancing and code migration. RMD is focused on improving the usability issues that plague related industry solutions. The platform is implemented on the JVM (Java Virtual Machine) and supports the Java and Kotlin programming languages, however, the platform should theoretically work with other JVM languages. RMD was designed to simplify the implementation of distributed algorithms by providing a Kotlin DSL (domain specific language) that allows the programmer to define jobs within dedicated code blocks. Users from around the world can donate their own computing resources by hosting their own job server. RMD is secured by placing all untrusted code within a sandbox environment that prevents potentially malicious actions from taking place. To demonstrate the feasibility of the proposed model, a proof of concept implementation has been constructed with real examples demonstrating the usefulness of the proposed solution.
Keyword:
Java virtual machine
Message passing interface
Load balancing
Code migration
Cluster computing
Grid computing
Remote procedure calls
Remote method delegation
High performance computing
Parallel computing
Domain specific language
Java
Kotlin
期刊
IF:
2.9
论文数:
761
被引数:
1.2K
机构
暂无机构信息
引用论文
.png){kind=icon}
查看更多内容