Towards traffic minimization for data placement in online social networks

Abstract

With the increasing number of users and a huge scale of data, the service providers of Online Social Networks (OSNs) are facing the problem of how to place users' data to multiple servers. Key-value stores solve the problem based on consistent hashing, and have become a defacto standard. However, random placement manner of hashing cannot preserve social locality, which leads to high intra-data center traffic and unpredictable response time. Many existing works solve the problem by using graph partitioning algorithms. These works have two drawbacks: First, the social graph is constructed with ordinary pairwise graph that cannot fully reflect multi-participant interactions often occurring in OSNs. Second, the underlying network topologies of data center have never been considered. This paper investigates the problem of traffic minimization for OSNs data storage. Motivated by maximally preserving both social locality and distance locality, we formulate the problem as two sub-problems -hypergraph partitioning and partition-to-server mapping, and propose a two-phase data placement (TDP) scheme. Specifically we present two algorithms to solve partition-to-server mapping over two widely used network topologies (i.e., tree and BCube). Evaluations with a large scale Facebook trace show that TDP significantly reduces intra-data center traffic as well as load balancing across servers.

Automated Server Operation Model to Improve the Efficiency of Domain Services

“…The cloud computing environment or the internet data center consists of many servers that process internet users' requests [2]. In addition, as the amount of data is increased due to the increasing number of social network users, service providers are faced with challenges to learn how to deploy user data on many servers [3]. Everyone must have probably experienced the inconvenience of service-use as the internet service tends to display images too long or cause buffering during the streaming play, thus failing to provide normal service.…”

Predicting required licensed spectrum for the future considering big data growth

“…Mobile broadband has become an essential service in the lives of most mobile cellular users, which has contributed to the exponentially increase in mobile data demands . Important factors contributing to this data demand surge are the massive and diverse developments of new portable devices such as smart phones, tablets, laptops, e‐book readers, gaming consoles, and dongles, which has led to the evolution of various mobile applications covering numerous areas of a user's life; such as social and educational applications, and those relating to information, news, science, health, trading, entertainment, Internet‐of‐things (IoTs), and machine‐to‐machine (M2M) applications .…”

Recent advances in big data analysis and application

“…However, random placement manner of hashing cannot preserve social locality, which leads to high intradata center traffic and unpredictable response time. Zhou et al investigates the problem of traffic minimization for OSNs data storage . Motivated by maximally preserving both social locality and distance locality, they formulate the problem as 2 subproblems—hypergraph partitioning and partition‐to‐server mapping—and propose a 2‐phase data placement scheme.…”