作者: Bin Shao , Yatao Li , Haixun Wang
DOI:
关键词: Distributed memory 、 Specification language 、 Theoretical computer science 、 Database schema 、 Random access 、 Communications protocol 、 World Wide Web 、 Cloud computing 、 Analytics 、 Computer science 、 Data access
摘要: Computations performed by graph algorithms are data driven, and require a high degree of random access. Despite the great progresses made in disk technology, it still cannot provide level efficient access required computation. On other hand, memory-based approaches usually do not scale due to capacity limit single machines. In this paper, we introduce Trinity, general purpose engine over distributed memory cloud. Through optimized management network communication, Trinity supports fast exploration as well parallel computing. particular, leverages patterns both online offline computation optimize communication for best performance. These enable support query processing analytics on large graphs with just few commodity Furthermore, provides specification language called TSL users declare schema protocols, which brings ease-of-use Our experiments show Trinity’s performance low latency queries throughput web-scale, billion-node graphs.