Siqi Shen

Statistical Characterization of Business-Critical Workloads Hosted in Cloud Datacenters

Business-critical workloads -- web servers, mail servers, app servers, etc. -- are increasingly hosted in virtualized data enters acting as Infrastructure-as-a-Service clouds (cloud data enters). Understanding how business-critical workloads demand and use resources is key in capacity sizing, in infrastructure operation and testing, and in application performance management. However, relatively little is currently known about these workloads, because the information is complex -- larges-scale, heterogeneous, shared-clusters -- and because datacenter operators remain reluctant to share such information. Moreover, the few operators that have shared data (e.g., Google and several supercomputing centers) have enabled studies in business intelligence (MapReduce), search, and scientific computing (HPC), but not in business-critical workloads. To alleviate this situation, in this work we conduct a comprehensive study of business-critical workloads hosted in cloud data enters. We collect two large-scale and long-term workload traces corresponding to requested and actually used resources in a distributed datacenter servicing business-critical workloads. We perform an in-depth analysis about workload traces. Our study sheds light into the workload of cloud data enters hosting business-critical workloads. The results of this work can be used as a basis to develop efficient resource management mechanisms for data enters. Moreover, the traces we released in this work can be used for workload verification, modelling and for evaluating resource scheduling policies, etc.

An analysis of online match-based games

Online match-based games, such as the online versions of the board game of chess, have already captured a global audience of tens of millions of players. Through a unique combination of characteristics, a relatively short duration-often “coffee-break” minutes instead of hours of continuous gameplay-, weak correlation between matches, and clear emphasis on winning, match-based games may serve a unique segment of the global player population. Although this segment of the gaming population may later become consumers of more sophisticated Massively Multiuser Virtual Environment (MMVE) systems, few previous studies have focused on the characteristics of online match-based games. Complementing them, in this work we collect and analyze information corresponding to 5 online match-based game datasets, totaling over 170 million matches played by a population of 1.3 million unique gamers over 14 years. Our analysis focuses on workload characteristics, win ratio, and player evolution. Studies such as ours may guide the multi-disciplinary field of MMVE design by providing new understanding of player lifetime and behavior. For example, we find a correlation between the interactivity of match-based games and the retention of players over both long and short term, that friendship does not always help to perform better in games, and that in match-based games each player explores tens of different play strategies over time.

Scheduling jobs in the cloud using on-demand and reserved instances

Deploying applications in leased cloud infrastructure is increasingly considered by a variety of business and service integrators. However, the challenge of selecting the leasing strategy -- larger or faster instances? on-demand or reserved instances? etc.-- and to configure the leasing strategy with appropriate scheduling policies is still daunting for the (potential) cloud user. In this work, we investigate leasing strategies and their policies from a brokers perspective. We propose, CoH, a family of Cloud-based, online, Hybrid scheduling policies that minimizes rental cost by making use of both on-demand and reserved instances. We formulate the resource provisioning and job allocation policies as Integer Programming problems. As the policies need to be executed online, we limit the time to explore the optimal solution of the integer program, and compare the obtained solution with various heuristics-based policies; then automatically pick the best one. We show, via simulation and using multiple real-world traces, that the hybrid leasing policy can obtain significantly lower cost than typical heuristics-based policies.

Understanding and recommending play relationships in online social gaming

Online Social Networking (OSN) applications such as Facebooks communication and Zyngas gaming platforms service hundreds of millions of users. To understand and model such relationships, social network graphs are extracted from running OSN applications and subsequently processed using social and complex network analysis tools. In this paper, we focus on the application domain of Online Social Games (OSGs) and deploy a formalism for extracting graphs from large datasets. Our formalism covers notions such as game participation, adversarial relationships, match outcomes, and allows to filter out “weak” links based on one or more threshold values. Using two novel large-scale OSG datasets, we investigate a range of threshold values and their influence on the resulting OSG graph properties. We discuss how an analysis of multiple graphs - obtained through different extraction rules - could be used in an algorithm to improve matchmaking for players.

Socializing by Gaming: Revealing Social Relationships in Multiplayer Online Games

Multiplayer Online Games (MOGs) like Defense of the Ancients and StarCraft II have attracted hundreds of millions of users who communicate, interact, and socialize with each other through gaming. In MOGs, rich social relationships emerge and can be used to improve gaming services such as match recommendation and game population retention, which are important for the user experience and the commercial value of the companies who run these MOGs. In this work, we focus on understanding social relationships in MOGs. We propose a graph model that is able to capture social relationships of a variety of types and strengths. We apply our model to real-world data collected from three MOGs that contain in total over ten years of behavioral history for millions of players and matches. We compare social relationships in MOGs across different game genres and with regular online social networks like Facebook. Taking match recommendation as an example application of our model, we propose SAMRA, a Socially Aware Match Recommendation Algorithm that takes social relationships into account. We show that our model not only improves the precision of traditional link prediction approaches, but also potentially helps players enjoy games to a higher extent.

Modeling Avatar Mobility of Networked Virtual Environments

Movement, one of the most common actions of avatars in virtual worlds, can have an important impact on the performance of networked virtual environments (NVEs). In this work, we propose SAMOVAR, a Statistical Area-based MObility model for VirtuAl enviRonments. SAMOVAR models four mobility characteristics: pause duration, velocity, area popularity, and distinct visited areas using empirical distribution; and then uses a map generation and a traveling procedure to generate movement trajectories of avatars. We show through simulation that the traces generated by our model can produce many mobility characteristics observed in virtual world. Further by comparing to trace-based simulation, the results obtained from SAMOVAR are similar to results obtained with traces from World of Warcraft and Second Life.

RTSenv: An experimental environment for real-time strategy games

Today, Real-Time Strategy (RTS) games entertain tens of millions of players world-wide. This growing population expects new game designs and more scalable games every year. However, few tools and environments exist for game designers and implementers; of these, even fewer are available to researchers and game communities. In this work, we introduce RTSenv, an environment and associated set of tools for RTS games. Our environment can configure and manage the main aspects of RTS games, such as maps, computer-controlled units, and game scenarios. RTSenv leverages multi-cluster systems and reactive fault tolerance mechanisms to perform robust, multi-machine, and multi-instance game experiments. Using our reference implementation of RTSenv in DAS-4, a real multi-cluster system, and Amazon EC2, a commercial cloud, we show that our approach can be used in a variety of scenarios. Our results give evidence that several common assumptions made by researchers about game workloads do not hold in general for RTS games and thus warrant a more detailed investigation.

Toxicity detection in multiplayer online games

Social interactions in multiplayer online games are an essential feature for a growing number of players world-wide. However, this interaction between the players might lead to the emergence of undesired and unintended behavior, particularly if the game is designed to be highly competitive. Communication channels might be abused to harass and verbally assault other players, which negates the very purpose of entertainment games by creating a toxic player-community. By using a novel natural language processing framework, we detect profanity in chat-logs of a popular Multiplayer Online Battle Arena (MOBA) game and develop a method to classify toxic remarks. We show how toxicity is non-trivially linked to game success.

When Game Becomes Life: The Creators and Spectators of Online Game Replays and Live Streaming

Online gaming franchises such as World of Tanks, Defense of the Ancients, and StarCraft have attracted hundreds of millions of users who, apart from playing the game, also socialize with each other through gaming and viewing gamecasts. As a form of User Generated Content (UGC), gamecasts play an important role in user entertainment and gamer education. They deserve the attention of both industrial partners and the academic communities, corresponding to the large amount of revenue involved and the interesting research problems associated with UGC sites and social networks. Although previous work has put much effort into analyzing general UGC sites such as YouTube, relatively little is known about the gamecast sharing sites. In this work, we provide the first comprehensive study of gamecast sharing sites, including commercial streaming-based sites such as Amazon’s Twitch.tv and community-maintained replay-based sites such as WoTreplays. We collect and share a novel dataset on WoTreplays that includes more than 380,000 game replays, shared by more than 60,000 creators with more than 1.9 million gamers. Together with an earlier published dataset on Twitch.tv, we investigate basic characteristics of gamecast sharing sites, and we analyze the activities of their creators and spectators. Among our results, we find that (i) WoTreplays and Twitch.tv are both fast-consumed repositories, with millions of gamecasts being uploaded, viewed, and soon forgotten; (ii) both the gamecasts and the creators exhibit highly skewed popularity, with a significant heavy tail phenomenon; and (iii) the upload and download preferences of creators and spectators are different: while the creators emphasize their individual skills, the spectators appreciate team-wise tactics. Our findings provide important knowledge for infrastructure and service improvement, for example, in the design of proper resource allocation mechanisms that consider future gamecasting and in the tuning of incentive policies that further help player retention.

Characterization of Human Mobility in Networked Virtual Environments

The design and tuning of networked virtual environments (NVEs), such as World of Warcraft (WoW), require understanding the in-NVE mobility characteristics of their citizens. Although many mobility-aware NVE systems already exist, their validation and further development have been hampered by the lack of public datasets and of comparison studies based on multiple datasets. To address these two issues, in this work we collect from WoW mobility traces for over 30,000 virtual citizens, and compare these traces with traces collected from Second Life (SL) where the environment is designed and changed significantly by the citizens themselves. Furthermore, motivated by the existence of numerous studies and models of networked real-world environments (NRE), we systematically compare the characteristics of two NVE and two NRE mobility traces. Our comparative study reveals that long-tail distributions characterize well various mobility characteristics, that the invisible boundary of human movement also appears for NVEs, and that area-visitation shows personal preferences. We also find several differences between NVE and NRE mobility characteristics.