Herman A. Engelbrecht

A Survey of State Persistency in Peer-to-Peer Massively Multiplayer Online Games

Recently, there has been significant research focus on Peer-to-Peer (P2P) Massively Multiplayer Online Games (MMOGs). A number of architectures have been presented in the literature to implement the P2P approach. One aspect that has not received sufficient attention in these architectures is state persistency in P2P MMOGs. This survey presents an overview of the current challenges present in P2P MMOGs, followed by an overview of classic state consistency models used in C/S MMOGs. The survey then classifies the state persistency techniques currently used in P2P MMOGs into super peer storage, overlay storage, hybrid storage, and distance-based storage. Key characteristics, namely scalability, fairness, reliability, responsiveness, and security are then defined. Each state persistency technique is evaluated according to these characteristics and recommendations are then made of possible future areas of research into the different storage types.

Forecasting methods for cloud hosted resources, a comparison

The emergence of cloud management systems, and the adoption of elastic cloud services enable dynamic adjustment of cloud hosted resources and provisioning. In order to effectively provision for dynamic workloads presented on cloud platforms, an accurate forecast of the load on the cloud resources is required. In this paper, we investigate various forecasting methods presented in recent research, identify and adapt evaluation metrics used in literature and compare forecasting methods on prediction performance. We investigate the performance gain of ensemble models when combining three of the best performing models into one model. We find that our 30th order Auto-regression model and Feed-Forward Neural Network method perform the best when evaluated on Googles Cluster dataset and using the provision specific metrics identified. We also show an improvement in forecasting accuracy when evaluating two ensemble models.

An Adaptive Congestion Control and Fairness Scheduling Strategy for Wireless Mesh Networks

Wireless mesh networks (WMNs) are a promising technology for low cost deployments for telemetry networks in rural areas. The popular contention based carrier sense multiple access with collision avoidance (CSMA/CA) technique is widely used in WMN implementations as it does not require time synchronization compared to time division multiple access (TDMA). The IEEE 802.11e standard was introduced to provide data differentiation services to data on a network with data of different priority. With this standard, the enhanced distributed channel access (EDCA) technique for contention based services experiences an fairness problem where high data can starve lower priority data. CSMA/CA was originally developed for single-hop networks. Collisions tend to increase in multi-hop networks as the contention for the medium increases. To address the fairness and performance degradation with an increase in contention in multi-hop network problems, a novel adaptive congestion control and fairness scheduling (CCFS) strategy is proposed in this paper. The proposed strategy is simulated in OMNeT++ and the INETMANET library to ascertain the performance of the strategy. The strategy was compared with EDCA in terms of end-to-end latency, packet loss percentage and Jains fairness index. The proposed adaptive strategy is shown to reduce packet loss in most test cases as well as provide an overall more fair system with data of different priority when compared to EDCA.

Pithos: a state persistency architecture for peer-to-peer massively multiuser virtual environments

An aspect of peer-to-peer (P2P) massively multiuser virtual environments (MMVEs) that has not received significant attention thus far, is how game objects can be stored in a distributed fashion, taking into account the unique aspects of MMVEs. The field is termed P2P MMVE state persistency. This paper explores the different models of state persistency for P2P MMVEs, namely super peer storage, overlay storage, hybrid storage and distance-based storage. All storage types are found to be lacking in some respects and therefore a novel storage system called Pithos is introduced. Pithos provides low latency storage for peers within the same group, while enabling higher latency, but reliable backup to a storage overlay. This is achieved by a two-tiered hybrid architecture making use of player grouping.

A survey of cross-layer protocols for IEEE 802.11 wireless multi-hop mesh networks

Summary There has been an escalation in deployment and research of wireless mesh networks by both the business community and academia in the last few years. Their attractive characteristics include low deployment cost, a low-cost option to extend network coverage and ease of maintenance due to their self-healing properties. Multiple routes exist between the sender and receiver nodes because of the mesh layout that ensures network connectivity even when node or link failures occur. Recent advances among others include routing metrics, optimum routing, security, scheduling, cross-layer designs and physical layer techniques. However, there are still challenges in wireless mesh networks as discussed in this paper that need to be addressed. Cross-layer design allows information from adjacent and non-adjacent layers to be used at a particular layer for performance improvement. This paper presents a survey of cross-layer protocol design approaches applied to the IEEE 802.11 standards for wireless multi-hop mesh networks that have been proposed over the last few years for improved performance. We summarize the current research efforts in cross-layer protocol design using the IEEE 802.11 standard in identifying unsolved issues that are a promising avenue to further research. Copyright

Pithos: Distributed Storage for Massive Multi-User Virtual Environments

There has been significant research effort into peer-to-peer (P2P) massively multi-user virtual environments (MMVEs). A number of architectures have been proposed to implement the P2P approach; however, the development of fully distributed MMVEs has met with a number of challenges. In this work, we address one of the key remaining challenges of state consistency and persistency in P2P MMVEs. Having reviewed state management and persistency architectures currently receiving research attention, we have identified deficiencies such as lack of load balancing, responsiveness, and scalability. To address these deficiencies, we present Pithos—a reliable, responsive, secure, load-balanced, and scalable distributed storage system, suited to P2P MMVEs. Pithos is designed specifically for P2P MMVEs, and we show that it improves the reliability and responsiveness of storage architectures as compared to existing P2P state persistency architectures. Pithos is implemented as an OverSim simulation running on the OMNeT++ network simulation framework. It is evaluated using up to 10,400 peers, with realistic latency profiles, with up to 15.8 million storage and retrieval requests that are generated to store a total of 2.4 million objects. Each peer in Pithos uses a maximum of 1,950Bps bandwidth to achieve 99.98% storage reliability, while the most reliable overlay storage configuration tested only achieved 93.65% reliability, using 2,182Bps bandwidth. Pithos is also more responsive than overlay storage, with an average responsiveness of 0.192s, compared with the average overlay responsiveness of 1.4s when retrieving objects from storage.

Scheduling Strategies to Improve Reliability and Fairness for Priority Based Smart Rural Contention Based Applications Over Low-Cost Wireless Mesh Backbone Networks

Wireless Mesh Networks (WMNs) are viewed as a cheap solution for telemetry networks in rural areas. The main advantages of WMNs are that they allow an easy extension of existing networks to service a wider area by using multi-hop wireless communication and they provide an alternate route when a route becomes faulty. Smart Rural Areas is a new concept for the development of rural areas. It is hypothesized that the Internet of Things (IoT) can help develop rural areas by providing better services resulting in poverty reduction. The widely used carrier sense multiple access with collision avoidance (CSMA/CA) was originally designed for Wireless Local Area Networks (WLANs) consisting of single-hop transmissions. CSMA/CA experiences a rapid decrease in performance when applied to multi-hop distributed networks as an increase in collisions and contention for the medium is experienced. The IEEE 802.11e standard provides data differentiation services for data of different priority levels with enhanced distributed channel access (EDCA) being used in contention based networks. With EDCA, an unfairness problem exists where high priority data can starve lower priority data. To address these problems in low-cost rural smart networks we investigate the performance of six design schemes for wireless backbone networks by assigning different roles to edge and core routers. Simulations were carried out to obtain the results using OMNeT ++ and the INET framework. Simulation results show that hybrid network designs using distributed coordination function (DCF) and EDCA can improve QoS.

Predicting object lifetimes in finite distributed storage systems under churn

Distributed network storage systems regularly make use of object replication to achieve sufficient levels of reliability under network churn. To maintain reliability, a repair mechanism is employed, which replaces destroyed replicas. When designing a distributed storage system, it is of great benefit to design for objects with known lifetimes. That is to say, the expected time that an object will remain available in the storage network under measurable network conditions is known. The paper proposes an embedded continuous time Markov chain to model objects replicated in a finite network under churn, with repair. Object lifetimes are found to be dependant on node departure and arrival rates, initial network size and the average network size if the average network size is comparable to the required number of replicas. The theoretical model results are compared to an OMNeT++ network simulation and found to closely match.

The deep learning Indaba report

Given the focus on deep learning and machine learning, there is a need to address this problem of low participation of Africans in data science and artificial intelligence. The Deep Learning Indaba was thus born to stimulate the participation of Africans within the research and innovation landscape surrounding deep learning and machine learning. This column reports on the Deep Learning Indaba event, which consisted of a 5-day series of introductory lectures on Deep Learning, held from 10-15 September 2017, coupled with tutorial sessions where participants gained practical experience with deep learning software packages. The column also includes interviews with some of the organisers to learn more about the origin and future plans of the Deep Learning Indaba.

Interview column: introduction

The interviews in the SIGMM records aim to provide the community with the insights, visions, and views from outstanding researchers in multimedia. With the interviews we particularly try to find out what makes these researchers outstanding and also to a certain extend what is going on in their mind, what are their visions and what are their thoughts about current topics. Examples from the last issues include interviews with Judith Redi, Klara Nahrstedt, and Wallapak Tavanapong. The interviewers are conducted via Skype or - even better - in person by meeting them at conferences or other community events. We aim to publish three to four interviews a year. If you have suggestions for who to interview, please feel free to contact one of the column editors, which are: