Logo des Repositoriums

Auflistung nach Autor:in "Reiser, Angelika"

1 - 3 von 3
  • Zeitschriftenartikel
    Bericht vom Herbsttreffen der GI-Fachgruppe Datenbanksysteme
    (Datenbank-Spektrum: Vol. 13, No. 1, 2013) Kemper, Alfons; Mühlbauer, Tobias; Neumann, Thomas; Reiser, Angelika; Rödiger, Wolf
  • Konferenzbeitrag
    HiSim: a highly extensible large-scale P2P network simulator
    (Datenbanksysteme für Business, Technologie und Web (BTW), 2011) Rupprecht, Lukas; Smejkal, Jessica; Reiser, Angelika; Kemper, Alfons
    The popularity of Peer-to-Peer networks is increasing rapidly but developing new protocols for P2P systems is a very complex task as testing and evaluating distributed systems involves high effort. P2P simulators are being developed to tackle this difficulty, to reduce cost and to speed up development. We describe HiSim, a modular and highly scalable P2P network simulator based on the simulation framework PeerSim which we use to simulate HiSbase, a P2P framework for efficient processing of multidimensional data. Because of its modular design, HiSim can easily be extended, e.g., to fit other application domains. Basic design problems, related to query processing, are introduced in general and concretely solved within HiSim. Ad- ditionally, HiSim provides mechanisms to evaluate new protocols using an integrated statistics component. We demonstrate the high scalability by performing simulations of up to 2*104 and 2*107.
  • Konferenzbeitrag
    ScyPer: a hybrid OLTP&OLAP distributed main memory database system for scalable real-time analytics
    (Datenbanksysteme für Business, Technologie und Web (BTW) 2044, 2013) Mühlbauer, Tobias; Rödiger, Wolf; Reiser, Angelika; Kemper, Alfons; Neumann, Thomas
    ScyPer is an abbreviation for Scaled-out HyPer, a version of the HyPer main memory hybrid OLTP&OLAP database system that horizontally scales out on sharednothing commodity hardware. Our demo shows that ScyPer a) achieves a near-linear scale-out of OLAP query throughput with the number of active nodes, b) sustains a constant OLTP throughput, c) is resilient to node failures, and d) offers real-time analytical capabilities through market-leading query response times and periodically forked TX-consistent virtual memory snapshots with sub-second lifetime durations.