Acolyte: an in-memory social network query system

WISE 2012 Challenge provides a data set crawled from Sina Weibo, which consists of hundreds of millions of anonymized following relationships and microblogs. In the performance track, 19 typical queries on the data set are given. Most of the queries are to get the top n users or microblogs which have the largest statistical numbers of some features during a specific period. The solution should focus on both throughput and latency in the execution of these queries. In this report, we present Acolyte, which is an in-memory query system that can solve the performance track efficiently.