Dr. Michael Kuhn

michaelkuhn.jpg

Biography

Michael Kuhn is a postdoctoral researcher in the Scientific Computing group at Universität Hamburg, where he also received his doctoral degree in computer science in 2015. He conducts research in the area of high performance I/O with a special focus on I/O interfaces and data reduction techniques. Other interests of his include file systems and high performance computing in general.

Projects

Theses

Teaching

Sommersemester 2017

Wintersemester 2016/2017

Sommersemester 2016

Wintersemester 2015/2016

Sommersemester 2015

Wintersemester 2014/2015

Sommersemester 2014

Wintersemester 2013/2014

Sommersemester 2013

Wintersemester 2012/2013

Sommersemester 2012

Wintersemester 2011/2012

Sommersemester 2011

Wintersemester 2010/2011

Sommersemester 2010

Wintersemester 2009/2010

Publications

2017

  • Poster: i_SSS – integrated Support System for Sustainability (Jannek Squar, Michael Bock, Olaf Conrad, Christoph Geck, Tobias Kawohl, Michael Kuhn, Lars Landschreiber, Hermann Lenhart, Sandra Wendland, Thomas Ludwig, Jürgen Böhner), Frankfurt, Germany, ISC High Performance 2017, 2017-06-20
    BibTeX URL

2016

  • Analyzing the energy consumption of the storage data path (Pablo Llopis, Manuel F. Dolz, Javier Garcia Blas, Florin Isaila, Mohammad Reza Heidari, Michael Kuhn), In The Journal of Supercomputing, Series: Number 11227, pp. 1–18, (Editors: Hamid Arabnia), Springer US, ISSN: 0920-8542, 2016
    BibTeX DOI

2015

  • ArduPower: A low-cost wattmeter to improve energy efficiency of HPC applications (Manuel F. Dolz, Mohammad Reza Heidari, Michael Kuhn, Thomas Ludwig, Germán Fabregat), In Sixth International Green Computing Conference and Sustainable Computing Conference (IGSC) 2015, pp. 1–8, IEEE, IGSC 2015, Las Vegas, USA, 2015-12
    BibTeX DOI
  • MPI-Checker - Static Analysis for MPI (Alexander Droste, Michael Kuhn, Thomas Ludwig), In Proceedings of the Second Workshop on the LLVM Compiler Infrastructure in HPC, LLVM '15, ACM (New York, USA), SC15, Austin, Texas, USA, ISBN: 978-1-4503-4005-2, 2015-11
    BibTeX URL DOI
  • Analyzing Power Consumption of I/O Operations in HPC Applications (Pablo Llopis, Manuel F. Dolz, Javier García-Blas, Florin Isaila, Jesús Carretero, Mohammad Reza Heidari, Michael Kuhn), In Proceedings of the Second International Workshop on Sustainable Ultrascale Computing Systems (NESUS 2015), pp. 107–116, (Editors: Jesus Carretero, Javier Garcia Blas, Roman Wyrzykowski, Emmanuel Jeannot), Computer Architecture, Communications and Systems Group (ARCOS) (Madrid, Spain), NESUS 2015, Jesus Carretero, Krakow, Poland, ISBN: 978-84-608-2581-4, 2015-10
    BibTeX URL
  • A Best Practice Analysis of HDF5 and NetCDF-4 Using Lustre (Christopher Bartz, Konstantinos Chasapis, Michael Kuhn, Petra Nerge, Thomas Ludwig), In High Performance Computing, Lecture Notes in Computer Science (9137), pp. 274–281, (Editors: Julian Martin Kunkel, Thomas Ludwig), Springer International Publishing (Switzerland), ISC 2015, Frankfurt, Germany, ISBN: 978-3-319-20118-4, ISSN: 0302-9743, 2015-06
    BibTeX DOI
  • Dynamically Adaptable I/O Semantics for High Performance Computing (Michael Kuhn), In High Performance Computing, Lecture Notes in Computer Science (9137), pp. 240–256, (Editors: Julian Martin Kunkel, Thomas Ludwig), Springer International Publishing (Switzerland), ISC 2015, Frankfurt, Germany, ISBN: 978-3-319-20118-4, ISSN: 0302-9743, 2015-06
    BibTeX DOI
  • Big Data Research at DKRZ – Climate Model Data Production Workflow (Michael Lautenschlager, Panagiotis Adamidis, Michael Kuhn), In Big Data and High Performance Computing (Lucio Grandinetti, Gerhard Joubert, Marcel Kunze, Valerio Pascucci), Series: Advances in Parallel Computing, Edition: 26, pp. 133–155, IOS Press, ISBN: 978-1-61499-582-1, 2015
    BibTeX URL DOI
    Abstract: The paper starts with a classification of climate modeling in Big Data and presents research activities in DKRZ's two basic climate modeling workflows, the climate model development and the climate model data production. Research emphasis in climate model development is on code optimization for efficient use of modern and future multi-core high performance computing architectures. Complementary research is related to increase of I/O bandwidth between compute nodes and hard discs as well as efficient use of storage resources. Research emphasis in climate model data production is on optimization of the end-to-end workflow in its different stages starting from climate model calculations over generation and storage of climate data products and ending in long-term archiving, interdisciplinary data utilization research data publication for integration of citable data entities in scientific literature articles.

2014

  • Compression By Default - Reducing Total Cost of Ownership of Storage Systems (Michael Kuhn, Konstantinos Chasapis, Manuel Dolz, Thomas Ludwig), In Supercomputing, Lecture Notes in Computer Science (8488), (Editors: Julian Martin Kunkel, Thomas Ludwig, Hans Werner Meuer), Springer International Publishing (Berlin, Heidelberg), ISC 2014, Leipzig, Germany, ISBN: 978-3-319-07517-4, ISSN: 0302-9743, 2014-06
    BibTeX DOI
  • Exascale Storage Systems – An Analytical Study of Expenses (Julian Kunkel, Michael Kuhn, Thomas Ludwig), In Supercomputing Frontiers and Innovations, Series: Volume 1, Number 1, pp. 116–134, (Editors: Jack Dongarra, Vladimir Voevodin), 2014-06
    BibTeX URL
  • Evaluating Power-Performace Benefits of Data Compression in HPC Storage Servers (Konstantinos Chasapis, Manuel Dolz, Michael Kuhn, Thomas Ludwig), In IARIA Conference, pp. 29–34, (Editors: Steffen Fries, Petre Dini), IARIA XPS Press, ENERGY 2014, Chamonix, France, ISBN: 978-1-61208-332-2, ISSN: 2308-412X, 2014-04-20 – Awards: Best Paper
    BibTeX
    Abstract: Both energy and storage are becoming key issues in high-performance (HPC) systems, especially when thinking about upcoming Exascale systems. The amount of energy consumption and storage capacity needed to solve future problems is growing in a marked curve that the HPC community must face in cost-/energy-efficient ways. In this paper we provide a power-performance evaluation of HPC storage servers that take over tasks other than simply storing the data to disk. We use the Lustre parallel distributed file system with its ZFS back-end, which natively supports compression, to show that data compression can help to alleviate capacity and energy problems. In the first step of our analysis we study different compression algorithms with regards to their CPU and power overhead with a real dataset. Then, we use a modified version of the IOR benchmark to verify our claims for the HPC environment. The results demonstrate that the energy consumption can be reduced by up to 30% in the write phase of the application and 7% for write-intensive applications. At the same time, the required storage capacity can be reduced by approximately 50%. These savings can help in designing more power-efficient and leaner storage systems.
  • Evaluating Lustre's Metadata Server on a Multi-socket Platform (Konstantinos Chasapis, Manuel Dolz, Michael Kuhn, Thomas Ludwig), In Proceedings of the 9th Parallel Data Storage Workshop, PDSW (2014), pp. 13–18, IEEE Press (Piscataway, NJ, USA), SC14, New Orleans, Louisiana, ISBN: 978-1-4799-7025-4, 2014
    BibTeX DOI
    Abstract: With the emergence of multi-core and multi-socket non-uniform memory access (NUMA) platforms in recent years, new software challenges have arisen to use them efficiently. In the field of high performance computing (HPC), parallel programming has always been the key factor to improve applications performance. However, the implications of parallel architectures in the system software has been overlooked until recently. In this work, we examine the implications of such platforms in the performance scalability of the Lustre parallel distributed file system's metadata server (MDS). We run our experiments on a four socket NUMA platform that has 48 cores. We leverage the mdtest benchmark to generate appropriate metadata workloads and include configurations with varying numbers of active cores and mount points. Additionally, we compare Lustre's metadata scalability with the local file systems ext4 and XFS. The results demonstrate that Lustre's metadata performance is limited to a single socket and decreases when more sockets are used. We also observe that the MDS's back-end device is not a limiting factor regarding the performance.

2013

  • A Semantics-Aware I/O Interface for High Performance Computing (Michael Kuhn), In Supercomputing, Lecture Notes in Computer Science (7905), pp. 408–421, (Editors: Julian Martin Kunkel, Thomas Ludwig, Hans Werner Meuer), Springer (Berlin, Heidelberg), ISC 2013, Leipzig, Germany, ISBN: 978-3-642-38749-4, ISSN: 0302-9743, 2013-06
    BibTeX DOI

2012

  • A Study on Data Deduplication in HPC Storage Systems (Dirk Meister, Jürgen Kaiser, Andre Brinkmann, Michael Kuhn, Julian Kunkel, Toni Cortes), In Proceedings of the ACM/IEEE Conference on High Performance Computing (SC), IEEE Computer Society, SC'12, Salt Lake City, USA, 2012-11-10
    BibTeX
  • Simulation-Aided Performance Evaluation of Server-Side Input/Output Optimizations (Michael Kuhn, Julian Kunkel, Thomas Ludwig), In 20th Euromicro International Conference on Parallel, Distributed and Network-Based Processing, pp. 562–566, (Editors: Rainer Stotzka, Michael Schiffers, Yiannis Cotronis), IEEE Computer Society (Los Alamitos, Washington, Tokyo), PDP 2012, Munich Network Management Team, Garching, Germany, ISBN: 978-0-7695-4633-9, ISSN: 1066-6192, 2012
    BibTeX
    Abstract: The performance of parallel distributed file systems suffers from many clients executing a large number of operations in parallel, because the I/O subsystem can be easily overwhelmed by the sheer amount of incoming I/O operations. Many optimizations exist that try to alleviate this problem. Client-side optimizations perform preprocessing to minimize the amount of work the file servers have to do. Server-side optimizations use server-internal knowledge to improve performance. The HDTrace framework contains components to simulate, trace and visualize applications. It is used as a testbed to evaluate optimizations that could later be implemented in real-life projects. This paper compares existing client-side optimizations and newly implemented server-side optimizations and evaluates their usefulness for I/O patterns commonly found in HPC. Server-directed I/O chooses the order of non-contiguous I/O operations and tries to aggregate as many operations as possible to decrease the load on the I/O subsystem and improve overall performance. The results show that server-side optimizations beat client-side optimizations in terms of performance for many use cases. Integrating such optimizations into parallel distributed file systems could alleviate the need for sophisticated client-side optimizations. Due to their additional knowledge of internal workflows server-side optimizations may be better suited to provide high performance in general.
  • Optimizations for Two-Phase Collective I/O (Michael Kuhn, Julian Kunkel, Yuichi Tsujita, Hidetaka Muguruma, Thomas Ludwig), In Applications, Tools and Techniques on the Road to Exascale Computing, Advances in Parallel Computing (22), pp. 455–462, (Editors: Koen De Bosschere, Erik H. D'Hollander, Gerhard R. Joubert, David Padua, Frans Peters), IOS Press (Amsterdam, Berlin, Tokyo, Washington DC), ParCo 2011, University of Ghent, ELIS Department, Ghent, Belgium, ISBN: 978-1-61499-040-6, ISSN: 0927-5452, 2012
    BibTeX
    Abstract: The performance of parallel distributed file systems suffers from many clients executing a large number of operations in parallel, because the I/O subsystem can be easily overwhelmed by the sheer amount of incoming I/O operations. This, in turn, can slow down the whole distributed system. Many optimizations exist that try to alleviate this problem. Client-side optimizations perform preprocessing to minimize the amount of work the file servers have to do. Server-side optimizations use server-internal knowledge to improve performance. This paper provides an overview of existing client-side optimizations and presents new modifications of the Two-Phase protocol. Interleaved Two-Phase is a modification of ROMIO's Two-Phase protocol, which iterates over the file differently to reduce the number of seek operations on disk. Pipelined Two-Phase uses a pipelined scheme which overlaps I/O and communication phases to utilize the network and I/O subsystems concurrently.
  • Evaluating the Influence of File System Interfaces and Semantics on I/O Throughput in High Performance Computing (Christina Janssen, Michael Kuhn, Thomas Ludwig), In Proceedings of the Work in Progress Session, 20th Euromicro International Conference on Parallel, Distributed and Network-Based Processing, SEA-Publications (31), (Editors: Erwin Grosspietsch, Konrad Klöckner), Institute for Systems Engineering and Automation (Johannes Kepler University Linz), PDP 2012, Munich Network Management Team, Garching, Germany, ISBN: 978-3-902457-31-8, 2012
    BibTeX
  • Scientific Computing: Performance and Efficiency in Climate Models (Sandra Schröder, Michael Kuhn, Nathanael Hübbe, Julian Kunkel, Timo Minartz, Petra Nerge, Florens Wasserfall, Thomas Ludwig), In Proceedings of the Work in Progress Session, 20th Euromicro International Conference on Parallel, Distributed and Network-Based Processing, SEA-Publications (31), (Editors: Erwin Grosspietsch, Konrad Klöckner), Institute for Systems Engineering and Automation (Johannes Kepler University Linz), PDP 2012, Munich Network Management Team, Garching, Germany, ISBN: 978-3-902457-31-8, 2012
    BibTeX
  • Tool Environments to Measure Power Consumption and Computational Performance (Timo Minartz, Daniel Molka, Julian Kunkel, Michael Knobloch, Michael Kuhn, Thomas Ludwig), In Handbook of Energy-Aware and Green Computing (Ishfaq Ahmad, Sanjay Ranka), Chapters: 31, pp. 709–743, Chapman and Hall/CRC Press Taylor and Francis Group (6000 Broken Sound Parkway NW, Boca Raton, FL 33487), ISBN: 978-1-4398-5040-4, 2012
    BibTeX

2011

  • Towards an Energy-Aware Scientific I/O Interface – Stretching the ADIOS Interface to Foster Performance Analysis and Energy Awareness (Julian Kunkel, Timo Minartz, Michael Kuhn, Thomas Ludwig), In Computer Science - Research and Development, Series: 1, (Editors: Thomas Ludwig), Springer (Berlin / Heidelberg, Germany), 2011
    BibTeX DOI
    Abstract: Intelligently switching energy saving modes of CPUs, NICs and disks is mandatory to reduce the energy consumption. Hardware and operating system have a limited perspective of future performance demands, thus automatic control is suboptimal. However, it is tedious for a developer to control the hardware by himself. In this paper we propose an extension of an existing I/O interface which on the one hand is easy to use and on the other hand could steer energy saving modes more efficiently. Furthermore, the proposed modifications are beneficial for performance analysis and provide even more information to the I/O library to improve performance. When a user annotates the program with the proposed interface, I/O, communication and computation phases are labeled by the developer. Run-time behavior is then characterized for each phase, this knowledge could be then exploited by the new library.

2010

  • Collecting Energy Consumption of Scientific Data (Julian Kunkel, Olga Mordvinova, Michael Kuhn, Thomas Ludwig), In Computer Science - Research and Development, Series: 3, pp. 1–9, (Editors: Thomas Ludwig), Springer (Berlin / Heidelberg, Germany), ISSN: 1865-2034, 2010
    BibTeX URL DOI
    Abstract: In this paper the data life cycle management is extended by accounting for energy consumption during the life cycle of files. Information about the energy consumption of data not only allows to account for the correct costs of its life cycle, but also provides a feedback to the user and administrator, and improves awareness of the energy consumption of file I/O. Ideas to realize a storage landscape which determines the energy consumption for maintaining and accessing each file are discussed. We propose to add new extended attributes to file metadata which enable to compute the energy consumed during the life cycle of each file.

2009

  • Dynamic file system semantics to enable metadata optimizations in PVFS (Michael Kuhn, Julian Kunkel, Thomas Ludwig), In Concurrency and Computation: Practice and Experience, Series: 21-14, pp. 1775–1788, John Wiley and Sons Ltd. (Chichester, UK), ISSN: 1532-0626, 2009
    BibTeX URL DOI
    Abstract: Modern file systems maintain extensive metadata about stored files. While metadata typically is useful, there are situations when the additional overhead of such a design becomes a problem in terms of performance. This is especially true for parallel and cluster file systems, where every metadata operation is even more expensive due to their architecture. In this paper several changes made to the parallel cluster file system Parallel Virtual File System (PVFS) are presented. The changes target at the optimization of workloads with large numbers of small files. To improve the metadata performance, PVFS was modified such that unnecessary metadata is not managed anymore. Several tests with a large quantity of files were performed to measure the benefits of these changes. The tests have shown that common file system operations can be sped up by a factor of two even with relatively few changes.

2008

  • Directory-Based Metadata Optimizations for Small Files in PVFS (Michael Kuhn, Julian Kunkel, Thomas Ludwig), In Euro-Par '08: Proceedings of the 14th international Euro-Par conference on Parallel Processing, pp. 90–99, Springer-Verlag (Berlin, Heidelberg), Euro-Par-08, University of Las Palmas de Gran Canaria, Las Palmas de Gran Canaria, Spain, ISBN: 978-3-540-85450-0, 2008 – Awards: Best Paper
    BibTeX DOI
    Abstract: Modern file systems maintain extensive metadata about stored files. While this usually is useful, there are situations when the additional overhead of such a design becomes a problem in terms of performance. This is especially true for parallel and cluster file systems, because due to their design every metadata operation is even more expensive. In this paper several changes made to the parallel cluster file system PVFS are presented. The changes are targeted at the optimization of workloads with large numbers of small files. To improve metadata performance, PVFS was modified such that unnecessary metadata is not managed anymore. Several tests with a large quantity of files were done to measure the benefits of these changes. The tests have shown that common file system operations can be sped up by a factor of two even with relatively few changes.

2007

  • Analysis of the MPI-IO Optimization Levels with the PIOViz Jumpshot Enhancement (Thomas Ludwig, Stephan Krempel, Michael Kuhn, Julian Kunkel, Christian Lohse), In Recent Advances in Parallel Virtual Machine and Message Passing Interface, Lecture Notes in Computer Science (4757), pp. 213–222, (Editors: Franck Cappello, Thomas Hérault, Jack Dongarra), Springer (Berlin / Heidelberg, Germany), EuroPVM/MPI-07, Institut national de recherche en informatique et automatique, Paris, France, ISBN: 978-3-540-75415-2, 2007
    BibTeX URL DOI
    Abstract: With MPI-IO we see various alternatives for programming file I/O. The overall program performance depends on many different factors. A new trace analysis environment provides deeper insight into the client/server behavior and visualizes events of both process types. We investigate the influence of making independent vs. collective calls together with access to contiguous and non-contiguous data regions in our MPI-IO program. Combined client and server traces exhibit reasons for observed I/O performance.

Talks

2017

  • JULEA: A Flexible Storage Framework for HPC (Dr. Michael Kuhn), Workshop on Performance and Scalability of Storage Systems (WOPSSS), Frankfurt, Germany, 2017-06-22 Presentation
  • High Performance I/O (Dr. Michael Kuhn), DKRZ Tour “Computational Sciences in Engineering”, Hamburg, Germany, 2017-06-08 Presentation
  • The Case for a Flexible HPC Storage Framework (Dr. Michael Kuhn), Dagstuhl Seminar “Challenges and Opportunities of User-Level File Systems for HPC” (17202), Wadern, Germany, 2017-05-18 Presentation

2016

2015

  • Exploiting Semantical Information for Performance Optimization and Data Reduction (Dr. Michael Kuhn), CluStor 2015, Hamburg, Germany, 2015-07-30
  • A Best Practice Analysis of HDF5 and NetCDF-4 Using Lustre (Dr. Michael Kuhn), ISC High Performance 2015, Frankfurt, Germany, 2015-07-15 Presentation
  • Dynamically Adaptable I/O Semantics for High Performance Computing (Dr. Michael Kuhn), ISC High Performance 2015, Frankfurt, Germany, 2015-07-14 Presentation
  • I/O Semantics for Future Storage Systems (Dr. Michael Kuhn), 14th HLRS/hww Workshop on Scalable Global Parallel File Systems, Stuttgart, Germany, 2015-04-29 Presentation

2014

  • Compression By Default – Reducing Total Cost of Ownership of Storage Systems (Dr. Michael Kuhn), International Supercomputing Conference 2014, Leipzig, Germany, 2014-06-23 Presentation
  • Exascale Storage Systems – An Analytical Study of Expenses (Dr. Michael Kuhn), CluStor 2014, Hamburg, Germany, 2014-06-19

2013

  • A Semantics-Aware I/O Interface for High Performance Computing (Dr. Michael Kuhn), International Supercomputing Conference 2013, Leipzig, Germany, 2013-06-18 Presentation

2012

  • A Semantics-Aware I/O Interface (Dr. Michael Kuhn), HPC Workshop, Leogang, Austria, 2012-02-29
  • Scientific Computing: Performance and Efficiency in Climate Models (Dr. Michael Kuhn), PDP 2012, Munich, Germany, 2012-02-17 Presentation
  • Evaluating the Influence of File System Interfaces and Semantics on I/O Throughput in High Performance Computing (Dr. Michael Kuhn), PDP 2012, Munich, Germany, 2012-02-17 Presentation
  • Simulation-Aided Performance Evaluation of Server-Side Input/Output Optimizations (Dr. Michael Kuhn), PDP 2012, Munich, Germany, 2012-02-16 Presentation

2011

2008

  • Directory-Based Metadata Optimizations for Small Files in PVFS (Dr. Michael Kuhn), Euro-Par 2008, Gran Canaria, Spain, 2008-08-29 Presentation

2006

  • File Systems for Mass Storage of Image Data in Bioinformatics (Dr. Michael Kuhn, Christian Lohse), CluStor 2006, Heidelberg, Germany, 2006-09-21

Supervised Theses

2017

2016

2015

  • Advanced Data Transformation and Reduction Techniques in ADIOS, Tim Alexander Dobert (Bachelor's Thesis), Advisors: Dr. Michael Kuhn, 2015-10-07, BibTeX
  • Static Code Analysis of MPI Schemas in C with LLVM, Alexander Droste (Bachelor's Thesis), Advisors: Dr. Michael Kuhn, 2015-09-25, BibTeX
  • Evaluation of performance and productivity metrics of potential programming languages in the HPC environment, Florian Wilkens (Bachelor's Thesis), Advisors: Dr. Michael Kuhn, Sandra Schröder, 2015-04-28, BibTeX
  • Adaptive Compression for the Zettabyte File System, Florian Ehmke (Master's Thesis), Advisors: Dr. Michael Kuhn, 2015-02-24, Thesis BibTeX

2014

2013

  • Design, Implementation, and Evaluation of a Low-Level Extent-Based Object Store, Sandra Schröder (Master's Thesis), Advisors: Dr. Michael Kuhn, 2013-12-18, Thesis BibTeX
  • Automated File System Correctness and Performance Regression Tests, Anna Fuchs (Bachelor's Thesis), Advisors: Dr. Michael Kuhn, 2013-09-23, Thesis BibTeX URL
  • Evaluating Distributed Database Systems for Use in File Systems, Roman Michel (Bachelor's Thesis), Advisors: Dr. Michael Kuhn, 2013-09-18, BibTeX
  • Evaluation of Different Storage Backends and Technologies for MongoDB, Johann Weging (Bachelor's Thesis), Advisors: Dr. Michael Kuhn, 2013-02-28, BibTeX

2011

  • Evaluation of File Systems and I/O Optimization Techniques in High Performance Computing, Christina Janssen (Bachelor's Thesis), Advisors: Dr. Michael Kuhn, 2011-12-05, Thesis BibTeX URL