TY - CHAP A1 - Enke, Harry A1 - Fiedler, Norman A1 - Fischer, Thomas A1 - Gnadt, Timo A1 - Ketzan, Erik A1 - Ludwig, Jens A1 - Rathmann, Torsten A1 - Stöckle, Gabriel A1 - Schintke, Florian ED - Enke, Harry ED - Ludwig, Jens T1 - Leitfaden zum Forschungsdaten-Management T2 - Leitfaden zum Forschungsdaten-Management Y1 - 2013 PB - Verlag Werner Hülsbusch, Glückstadt ER - TY - JOUR A1 - Enke, Harry A1 - Partl, Adrian A1 - Reinefeld, Alexander A1 - Schintke, Florian T1 - Handling Big Data in Astronomy and Astrophysics JF - Datenbank-Spektrum Y1 - 2012 UR - http://dx.doi.org/10.1007/s13222-012-0099-1 U6 - https://doi.org/10.1007/s13222-012-0099-1 VL - 12 IS - 3 SP - 173 EP - 181 PB - Springer-Verlag ER - TY - JOUR A1 - Schintke, Florian T1 - XtreemFS & Scalaris JF - Science & Technology Y1 - 2013 IS - 6 SP - 54 EP - 55 PB - Pan European Networks ER - TY - JOUR A1 - Seibert, Felix A1 - Peters, Mathias A1 - Schintke, Florian T1 - Improving I/O Performance Through Colocating Interrelated Input Data and Near-Optimal Load Balancing JF - Proceedings of the IPDPSW; Fourth IEEE International Workshop on High-Performance Big Data, Deep Learning, and Cloud Computing (HPBDC) N2 - Most distributed file systems assign new files to storage servers randomly. While working well in some situations, this does not help to optimize the input performance for most MapReduce computations’ data access patterns. In this work, we consider an access pattern where input files are partitioned into groups of heterogeneous size. Each group is accessed by exactly one process. We design and implement a data placement strategy that places these file groups together on the same storage server. This colocation approach is combined with near-optimal storage load balancing. To do so, we use a classical scheduling approximation algorithm to solve the NP hard group assignment problem. We argue that local processing is not only beneficial because of reduced network traffic, but especially because it imposes an even resource schedule. Our experiments, based on the parallel processing of remote sensing images, reveal an enormous reduction of network traffic and up to 39 % faster input read times. Further, simulations show that our approximate assignments limit storage server imbalances to less than 5 % above the theoretical minimum, in contrast to more than 85 % with random assignment. Y1 - 2018 U6 - https://doi.org/10.1109/IPDPSW.2018.00081 N1 - Best Paper Award VL - 2018 SP - 448 EP - 457 ER - TY - GEN A1 - Scheffler, Daniel A1 - Sips, Mike A1 - Behling, Robert A1 - Dransch, Doris A1 - Eggert, Daniel A1 - Fajerski, Jan A1 - Freytag, Johann-Christoph A1 - Griffiths, Patrick A1 - Hollstein, André A1 - Hostert, Patrick A1 - Köthur, Patrick A1 - Peters, Mathias A1 - Pflugmacher, Dirk A1 - Rabe, Andreas A1 - Reinefeld, Alexander A1 - Schintke, Florian A1 - Segel, Karl T1 - GeoMultiSens – Scalable Multisensoral Analysis of Satellite Remote Sensing Data T2 - ESA Living Planet Symposium, EO Open Science Posters Y1 - 2016 ER - TY - CHAP A1 - Scheffler, Daniel A1 - Sips, Mike A1 - Behling, Robert A1 - Dransch, Doris A1 - Eggert, Daniel A1 - Fajerski, Jan A1 - Freytag, Johann-Christoph A1 - Griffiths, Patrick A1 - Hollstein, André A1 - Hostert, Patrick A1 - Köthur, Patrick A1 - Peters, Mathias A1 - Pflugmacher, Dirk A1 - Rabe, Andreas A1 - Reinefeld, Alexander A1 - Schintke, Florian A1 - Segel, Karl T1 - Geomultisens – a common automatic processing and analysis system for multi-sensor satellite data T2 - Advancing Horizons for Land Cover Services Entering the Big Data Era, Second joint Workshop of the EARSeL Special Interest Group on Land Use & Land Cover and the NASA LCLUC Program Y1 - 2016 SP - 18 EP - 19 ER - TY - JOUR A1 - Salem, Farouk A1 - Schintke, Florian A1 - Schütt, Thorsten A1 - Reinefeld, Alexander T1 - Improving the throughput of a scalable FLESnet using the Data-Flow Scheduler JF - CBM Progress Report 2018 Y1 - 2019 SN - 978-3-9815227-6-1 U6 - https://doi.org/10.15120/GSI-2019-01018 SP - 149 EP - 150 ER - TY - CHAP A1 - Gholami, Masoud A1 - Schintke, Florian T1 - Multilevel Checkpoint/Restart for Large Computational Jobs on Distributed Computing Resources T2 - 2019 IEEE 38th Symposium on Reliable Distributed Systems (SRDS) Y1 - 2019 U6 - https://doi.org/10.1109/SRDS47363.2019.00025 SP - 143 EP - 152 ER - TY - CHAP A1 - Hartung, Marc A1 - Schintke, Florian A1 - Schütt, Thorsten T1 - Pinpoint Data Races via Testing and Classification T2 - 2019 IEEE International Symposium on Software Reliability Engineering Workshops (ISSREW); 3rd International Workshop on Software Faults (IWSF 2019) Y1 - 2019 U6 - https://doi.org/10.1109/ISSREW.2019.00100 SP - 386 EP - 393 ER - TY - JOUR A1 - Salem, Farouk A1 - Schintke, Florian A1 - Schütt, Thorsten A1 - Reinefeld, Alexander T1 - Scheduling data streams for low latency and high throughput on a Cray XC40 using Libfabric JF - Concurrency and Computation Practice and Experience Y1 - 2020 U6 - https://doi.org/10.1002/cpe.5563 VL - 32 IS - 20 SP - 1 EP - 14 ER - TY - JOUR A1 - Skrzypczak, Jan A1 - Schintke, Florian A1 - Schütt, Thorsten T1 - RMWPaxos: Fault-Tolerant In-Place Consensus Sequences JF - IEEE Transactions on Parallel and Distributed Systems Y1 - 2020 U6 - https://doi.org/10.1109/TPDS.2020.2981891 SN - 1045-9219 VL - 31 IS - 10 SP - 2392 EP - 2405 ER - TY - JOUR A1 - Skrzypczak, Jan A1 - Schintke, Florian T1 - Towards Log-Less, Fine-Granular State Machine Replication JF - Datenbank Spektrum Y1 - 2020 U6 - https://doi.org/10.1007/s13222-020-00358-4 VL - 20 IS - 3 SP - 231 EP - 241 ER - TY - CHAP A1 - Allen, Gabrielle A1 - Dramlitsch, Thomas A1 - Goodale, Tom A1 - Lanfermann, Gerd A1 - Radke, Thomas A1 - Seidel, Edward A1 - Kielmann, Thilo A1 - Verstoep, Kees A1 - Balaton, Zoltán A1 - Kacsuk, Péter A1 - Szalai, Ferenc A1 - Gehring, Jörn A1 - Keller, Axel A1 - Streit, Achim A1 - Matyska, Ludek A1 - Ruda, Miroslav A1 - Krenek, Ales A1 - Knipp, Harald A1 - Merzky, André A1 - Reinefeld, Alexander A1 - Schintke, Florian A1 - Ludwiczak, Bogdan A1 - Nabrzyski, Jarek A1 - Pukacki, Juliusz A1 - Kersken, Hans-Peter A1 - Aloisio, Giovanni A1 - Cafaro, Massimo A1 - Ziegler, Wolfgang A1 - Russell, Michael T1 - Early experiences with the EGrid testbed T2 - Proceedings First IEEE/ACM International Symposium on Cluster Computing and the Grid Y1 - 2001 U6 - https://doi.org/10.1109/CCGRID.2001.923185 SP - 130 EP - 137 ER - TY - CHAP A1 - Miller, Barton P. A1 - Labarta, Jesús A1 - Schintke, Florian A1 - Simon, Jens T1 - Performance Evaluation, Analysis and Optimization T2 - Euro-Par 2002 Parallel Processing Y1 - 2002 U6 - https://doi.org/10.1007/3-540-45706-2_15 VL - 2400 SP - 131 EP - 131 ER - TY - CHAP A1 - Iamnitchi, Adriana A1 - Trunfio, Paolo A1 - Ledlie, Jonathan A1 - Schintke, Florian T1 - Peer-to-Peer Computing T2 - Euro-Par 2010 - Parallel Processing Y1 - 2010 U6 - https://doi.org/10.1007/978-3-642-15277-1_42 VL - 6271 SP - 444 EP - 445 ER - TY - CHAP A1 - Schintke, Florian A1 - Simon, Jens A1 - Reinefeld, Alexander T1 - A Cache Simulator for Shared Memory Systems T2 - Computational Science - ICCS 2001 Y1 - 2001 U6 - https://doi.org/10.1007/3-540-45718-6_62 VL - 2074 SP - 569 EP - 578 ER - TY - JOUR A1 - Schütt, Thorsten A1 - Schintke, Florian A1 - Skrzypczak, Jan T1 - Transactions on Red-black and AVL trees in NVRAM JF - arXiv Y1 - 2020 ER - TY - CHAP A1 - Bärring, Olof A1 - Lopez, Maite Barroso A1 - Cancio, German A1 - Röblitz, Thomas A1 - Schintke, Florian A1 - et al., T1 - Towards automation of computing fabrics using tools from the fabric management workpackage of the EU DataGrid project T2 - Computing in High Energy and Nuclear Physics (CHEP03) Y1 - 2003 SP - 1 EP - 8 ER - TY - JOUR A1 - Salem, Farouk A1 - Schintke, Florian A1 - Reinefeld, Alexander T1 - Handling Compute-Node Failures in FLESnet JF - CBM Progress Report 2019 Y1 - 2020 SN - 978-3-9815227-8-5 U6 - https://doi.org/10.15120/GSI-2020-00904 SP - 167 EP - 168 ER - TY - CHAP A1 - Skrzypczak, Jan A1 - Schintke, Florian A1 - Schütt, Thorsten T1 - Linearizable State Machine Replication of State-Based CRDTs without Logs T2 - Proceedings of the 2019 ACM Symposium on Principles of Distributed Computing, PODC 2019 N2 - General solutions of state machine replication have to ensure that all replicas apply the same commands in the same order, even in the presence of failures. Such strict ordering incurs high synchronization costs due to the use of distributed consensus or a leader. This paper presents a protocol for linearizable state machine replication of conflict-free replicated data types (CRDTs) that neither requires consensus nor a leader. By leveraging the properties of state-based CRDTs—in particular the monotonic growth of a join semilattice—synchronization overhead is greatly reduced. In addition, updates just need a single round trip and modify the state ‘in-place’ without the need for a log. Furthermore, the message size overhead for coordination consists of a single counter per message. While reads in the presence of concurrent updates are not wait-free without a coordinator, we show that more than 97 % of reads can be handled in one or two round trips under highly concurrent accesses. Our protocol achieves high throughput without auxiliary processes such as command log management or leader election. It is well suited for all practical scenarios that need linearizable access on CRDT data on a fine-granular scale. Y1 - 2019 U6 - https://doi.org/10.1145/3293611.3331568 SP - 455 EP - 457 ER - TY - CHAP A1 - Hartung, Marc A1 - Schintke, Florian T1 - Learned Clause Minimization in Parallel SAT Solvers T2 - Pragmatics of SAT 2019 N2 - Learned clauses minimization (LCM) let to performance improvements of modern SAT solvers especially in solving hard SAT instances. Despite the success of LCM approaches in sequential solvers, they are not widely incorporated in parallel SAT solvers. In this paper we explore the potential of LCM for parallel SAT solvers by defining multiple LCM approaches based on clause vivification, comparing their runtime in different SAT solvers and discussing reasons for performance gains and losses. Results show that LCM only boosts performance of parallel SAT solvers on a fraction of SAT instances. More commonly applying LCM decreases performance. Only certain LCM approaches are able to improve the overall performance of parallel SAT solvers. Y1 - 2019 UR - http://www.pragmaticsofsat.org/2019/ SP - 1 EP - 11 ER - TY - CHAP A1 - Salem, Farouk A1 - Schütt, Thorsten A1 - Schintke, Florian A1 - Reinefeld, Alexander T1 - Scheduling Data Streams for Low Latency and High Throughput on a Cray XC40 Using Libfabric T2 - CUG Conference Proceedings N2 - Achieving efficient many-to-many communication on a given network topology is a challenging task when many data streams from different sources have to be scattered concurrently to many destinations with low variance in arrival times. In such scenarios, it is critical to saturate but not to congest the bisectional bandwidth of the network topology in order to achieve a good aggregate throughput. When there are many concurrent point-to-point connections, the communication pattern needs to be dynamically scheduled in a fine-grained manner to avoid network congestion (links, switches), overload in the node’s incoming links, and receive buffer overflow. Motivated by the use case of the Compressed Baryonic Matter experiment (CBM), we study the performance and variance of such communication patterns on a Cray XC40 with different routing schemes and scheduling approaches. We present a distributed Data Flow Scheduler (DFS) that reduces the variance of arrival times from all sources at least 30 times and increases the achieved aggregate bandwidth by up to 50%. Y1 - 2019 ER - TY - JOUR A1 - Skrzypczak, Jan A1 - Schintke, Florian A1 - Schütt, Thorsten T1 - Linearizable State Machine Replication of State-Based CRDTs without Logs JF - arXiv N2 - General solutions of state machine replication have to ensure that all replicas apply the same commands in the same order, even in the presence of failures. Such strict ordering incurs high synchronization costs caused by distributed consensus or by the use of a leader. This paper presents a protocol for linearizable state machine replication of conflict-free replicated data types (CRDTs) that neither requires consensus nor a leader. By leveraging the properties of state-based CRDTs - in particular the monotonic growth of a join semilattice - synchronization overhead is greatly reduced. In addition, updates just need a single round trip and modify the state `in-place' without the need for a log. Furthermore, the message size overhead for coordination consists of a single counter per message. While reads in the presence of concurrent updates are not wait-free without a coordinator, we show that more than 97% of reads can be handled in one or two round trips under highly concurrent accesses. Our protocol achieves high throughput without auxiliary processes like command log management or leader election. It is well suited for all practical scenarios that need linearizable access on CRDT data on a fine-granular scale. Y1 - 2019 ER -