Dhj Dick Epema

Performance Analysis of Cloud Computing Services for Many-Tasks Scientific Computing

Cloud computing is an emerging commercial infrastructure paradigm that promises to eliminate the need for maintaining expensive computing facilities by companies and institutes alike. Through the use of virtualization and resource time sharing, clouds serve with a single set of physical resources a large user base with different needs. Thus, clouds have the potential to provide to their owners the benefits of an economy of scale and, at the same time, become an alternative for scientists to clusters, grids, and parallel production environments. However, the current commercial clouds have been built to support web and small database workloads, which are very different from typical scientific computing workloads. Moreover, the use of virtualization and resource time sharing may introduce significant performance penalties for the demanding scientific computing workloads. In this work, we analyze the performance of cloud computing services for scientific computing workloads. We quantify the presence in real scientific computing workloads of Many-Task Computing (MTC) users, that is, of users who employ loosely coupled applications comprising many tasks to achieve their scientific goals. Then, we perform an empirical evaluation of the performance of four commercial cloud computing services including Amazon EC2, which is currently the largest commercial cloud. Last, we compare through trace-based simulation the performance characteristics and cost models of clouds and other scientific computing platforms, for general and MTC-based scientific computing workloads. Our results indicate that the current clouds need an order of magnitude in performance improvement to be useful to the scientific community, and show which improvements should be considered first to address this discrepancy between offer and demand.

TRIBLER: a social-based peer-to-peer system

Most current peer‐to‐peer (P2P) file‐sharing systems treat their users as anonymous, unrelated entities, and completely disregard any social relationships between them. However, social phenomena such as friendship and the existence of communities of users with similar tastes or interests may well be exploited in such systems in order to increase their usability and performance. In this paper we present a novel social‐based P2P file‐sharing paradigm that exploits social phenomena by maintaining social networks and using these in content discovery, content recommendation, and downloading. Based on this paradigms main concepts such as taste buddies and friends, we have designed and implemented the TRIBLER P2P file‐sharing system as a set of extensions to BitTorrent. We present and discuss the design of TRIBLER, and we show evidence that TRIBLER enables fast content discovery and recommendation at a low additional overhead, and a significant improvement in download performance. Copyright

Give-to-Get : free-riding resilient video-on-demand in P2P systems

Centralised solutions for Video-on-Demand (VoD) services, which stream pre-recorded video content to multiple clients who start watching at the moments of their own choosing, are not scalable because of the high bandwidth requirements of the central video servers. Peer-to-peer (P2P) techniques which let the clients distribute the video content among themselves, can be used to alleviate this problem. However, such techniques may introduce the problem of free-riding, with some peers in the P2P network not forwarding the video content to others if there is no incentive to do so. When the P2P network contains too many free-riders, an increasing number of the well-behaving peers may not achieve high enough download speeds to maintain an acceptable service. In this paper we propose Give-to-Get, a P2P VoD algorithm which discourages free-riding by letting peers favour uploading to other peers who have proven to be good uploaders. As a consequence, free-riders are only tolerated as long as there is spare capacity in the system. Our simulations show that even if 20% of the peers are free-riders, Give-to-Get continues to provide good performance to the well-behaving peers. In particular, they show that Give-to-Get performs very well for short videos, which dominate the current VoD traffic on the Internet.

2Fast : Collaborative Downloads in P2P Networks

P2P systems that rely on the voluntary contribution of bandwidth by the individual peers may suffer from free riding. To address this problem, mechanisms enforcing fairness in bandwidth sharing have been designed, usually by limiting the download bandwidth to the available upload bandwidth. As in real environments the latter is much smaller than the former, these mechanisms severely affect the download performance of most peers. In this paper we propose a system called 2Fast, which solves this problem while preserving the fairness of bandwidth sharing. In 2Fast, we form groups of peers that collaborate in downloading a file on behalf of a single group member, which can thus use its full download bandwidth. A peer in our system can use its currently idle bandwidth to help other peers in their ongoing downloads, and get in return help during its own downloads. We assess the performance of 2Fast analytically and experimentally, the latter in both real and simulated environments. We find that in realistic bandwidth limit settings, 2Fast improves the download speed by up to a factor of 3.5 in comparison to state-of-the-art P2P download protocols

Optimizing Peer Relationships in a Super-Peer Network

Super-peer architectures exploit the heterogeneity of nodes in a P2P network by assigning additional responsibilities to higher-capacity nodes. In the design of a super-peer network for file sharing, several issues have to be addressed: how client peers are related to super-peers, how super-peers locate files, how the load is balanced among the super-peers, and how the system deals with node failures. In this paper we introduce a self-organizing super-peer network architecture (SOSPNET) that solves these issues in a fully decentralized manner. SOSPNET maintains a super-peer network topology that reflects the semantic similarity of peers sharing content interests. Super-peers maintain semantic caches of pointers to files which are requested by peers with similar interests. Client peers, on the other hand, dynamically select super-peers offering the best search performance. We show how this simple approach can be employed not only to optimize searching, but also to solve generally difficult problems encountered in P2P architectures such as load balancing and fault tolerance. We evaluate SOSPNET using a model of the semantic structure derived from the 8-month traces of two large file-sharing communities. The obtained results indicate that SOSPNET achieves close-to-optimal file search performance, quickly adjusts to changes in the environment (node joins and leaves), survives even catastrophic node failures, and efficiently distributes the system load taking into account peer capacities.

An Amortized Tit-For-Tat Protocol for Exchanging Bandwidth instead of Content in P2P Networks

Incentives for resource sharing are crucial for the proper operation of P2P networks. The principle of the incentive mechanisms in current content sharing P2P networks such as BitTorrent is to have peers exchange content of mutual interest. As a consequence, a peer can actively participate in the system only if it shares content that is of immediate interest to other peers. In this paper we propose to lift this restriction by using bandwidth rather than content as the resource upon which incentives are based. Bandwidth, in contrast to content, is independent of peer interests and so can be exchanged between any two peers. We present the design of a protocol called amortized tit-for-tat (ATFT) based on the bandwidth-exchange concept. This protocol defines mechanisms for bandwidth exchange corresponding to those in BitTorrent for content exchange, in particular for finding bandwidth borrowers that amortize the bandwidth borrowed in the past with their currently idle bandwidth. In addition to the formally proven incentives for bandwidth contributions, ATFT provides natural solutions to the problems of peer bootstrapping, seeding incentive, peer link asymmetry, and anonymity, which have previously been addressed with much more complex designs. Experiments with a real-world dataset confirm that ATFT is efficient in enforcing bandwidth contributions and results in download performance better than provided by incentive mechanisms based on content exchange.

The Design and Evaluation of a Self-Organizing Superpeer Network

Superpeer architectures exploit the heterogeneity of nodes in a peer-to-peer (P2P) network by assigning additional responsibilities to higher capacity nodes. In the design of a superpeer network for file sharing, several issues have to be addressed: how client peers are related to superpeers, how superpeers locate files, how the load is balanced among the superpeers, and how the system deals with node failures. In this paper, we introduce a self-organizing superpeer network architecture (SOSPNet) that solves these issues in a fully decentralized manner. SOSPNet maintains a superpeer network topology that reflects the semantic similarity of peers sharing content interests. Superpeers maintain semantic caches of pointers to files, which are requested by peers with similar interests. Client peers, on the other hand, dynamically select superpeers offering the best search performance. We show how this simple approach can be employed not only to optimize searching, but also to solve generally difficult problems encountered in P2P architectures such as load balancing and fault tolerance. We evaluate SOSPNet using a model of the semantic structure derived from eight-month traces of two large file-sharing communities. The obtained results indicate that SOSPNet achieves close-to-optimal file search performance, quickly adjusts to changes in the environment (node joins and leaves), survives even catastrophic node failures, and efficiently distributes the system load taking into account superpeer capacities.

The peer-to-peer trace archive: design and comparative trace analysis

Peer-to-Peer (P2P) systems have gained a phenomenal popularity in the past few years; BitTorrent serves daily tens of millions of people and generates an important fraction of the Internet traffic. Measurement data collected from real P2P systems are fundamental for gaining solid knowledge of the usage patterns and the characteristics of these systems, and can improve the modeling, the design, and the evaluation of P2P systems. Although many P2P measurements have been carried out in the last decade [1, 2, 4, 6, 7], few measurement data sets are publicly available, and for these few the data are presented in different formats. This situation hampers the exchange, study, and reuse of existing traces. Furthermore, due to the lack of available datasets, many P2P studies have been based on unrealistic assumptions about the characteristics and usage patterns of P2P systems; as a consequence, many P2P algorithms and methods lack a realistic evaluation. To remedy this situation, we have set to create the P2P Trace Archive (P2PTA), a virtual meeting place that facilitates the collection and exchange of P2P traces.

Decay-usage scheduling in multiprocessors

Decay-usage scheduling is a priority-aging time-sharing scheduling policy capable of dealing with a workload of both interactive and batch jobs by decreasing the priority of a job when it acquires CPU time, and by increasing its priority when it does not use the (a) CPU. In this article we deal with a decay-usage scheduling policy in multiprocessors modeled after widely used systems. The priority of a job consists of a base priority and a time-dependent component based on processor usage. Because t he priorities in our model are time dependent, a queuing-theoretic analysis—for instance, for the mean job response time—seems impossible. Still, it turns out that as a consequence of the scheduling policy, the shares of the available CPU time obtained by jobs converge, and a deterministic analysis for these shares is feasible: We show how for a fixed set of jobs with large processing demands, the steady-state shares can be obtained given the base priorities, and conversely, how to set the base priorities given the required shares. In addition, we analyze the relation between the values of the scheduler parameters and the level of control it can exercise over the steady-state share ratios, and we deal with the rate of convergence. We validate the model by simulations and by measurements of actual systems.

A two-level semantic caching scheme for super-peer networks

Recent measurement studies of file-sharing peer-to-peer networks have demonstrated the presence of semantic proximity between peers and between shared files. In this paper we tackle the problem of exploiting the semantic locality of peer requests to improve the performance of a P2P network by the use of semantic caches. Such caches group together peers with similar interests as well as files with similar request patterns. The resulting two-level caching infrastructure can be integrated in a very natural way with the super-peer concept. The super-peers in our system cache pointers to files recently requested by their client peers. The client peers, on the other hand, keep caches of super-peers that answered most of their requests in the past. As a consequence, peers with similar interests are grouped (clustered) under the same super-peers, and also files with similar request patterns are indexed by the same super-peers. We discuss the design choices and optimizations of the presented model. We also evaluate our system versus the network that uses only one level of semantic caches.