THE WORKSHOP ON PERFORMANCE AND SCALABILITY OF STORAGE SYSTEMS

The aim of the first annual Workshop on Performance and Scalability of Storage Systems (WOPSSS), hosted June 23, 2016, at ISC ’16, was to present state-of-the-art research, innovative ideas, and experience that focused on the design and implementation of HPC storage systems in both academic and industrial worlds, with a special interest on performance analysis.
MISSION

STATE-OF-THE-ART RESEARCH & INNOVATION, COMING TOGETHER

The Challenge

Presently, two factors affect the design of new storage architectures. First, the race for Exascale is pushing storage systems to a scale unseen in current practice. Second, new technologies, such Storage Class Memories (SCM), are causing disruptions in the market. In the past, systems could be seen as three well identified components, each with its own order of magnitude: DRAM – nanosecond, network – microsecond, storage – millisecond. Now, this hierarchy and its isolation among components is becoming irrelevant because storage components are filtering down to faster levels (e.g., SCM latency is in nanoseconds, and NVMe storage is in microseconds). Thus, the behavior of data storage and its access is more complex to analyze. Furthermore, Exascale systems will be heterogeneous, harnessing different kinds of nodes – thin and fat compute nodes, I/O delegation nodes, storage node – leading to complex networks and traffic management problems.

These factors will probably lead to significant loss of performance predictability, leaving storage system designers, application developers and the storage community at large in the difficult situation of not being able to detect bottlenecks precisely, evaluate the room for improvement, or estimate the matching of applications with a given storage architecture.

The Solution

Held in conjunction with ISC’16 in Frankfurt, Germany, the intention of WOPSSS was to encourage discussion of these issues through submissions of researchers or practitioners from both academia and the industry.

Topics of interest included (but were not limited to):

  1. Storage systems modeling and analysis tools
  2. Feed-back and empirical evaluation of storage systems
  3. Application I/O characterization
  4. Parallel I/O and storage systems (consistency, caching, replication, reliability and fault recovery overhead)
  5. Network challenges and storage systems (scalability, QoS, partitionability)
  6. Low-latency storage systems usability and analysis (memory-only, Flash, NVRAM, and Storage Class Memory)
  7. File system design
  8. Operating system stack optimization
  9. Cloud and distributed storage

Conference Materials

PRESENTATIONS
Keynote Speaker - Brent Gorda
Eutropia: User-space I/O for US-level Storage Devices
Anastasios Papagiannis, Giorgos Saloustros, Manolis Marazakis and Angelos Bilas
Early Evaluation of the “Infinite Memory Engine” Burst Buffer Solution
Wolfram Schenck, Salem El Sayed, Maciej Foszczynski, Wilhelm Homberg and Dirk Pleiter
IO-Profiling
Bjorn Olausson
Motivation and Implementation of a Dynamic Remote Storage System for I/O demanding HPC Applications
Matthias Neuer, Jürgen Salk, Holger Berger, Erich Focht, Christian Mosch, Karsten Siegmund, Volodymyr Kushnarenko, Stefan Kombrink and Stefan Wesner
Parallel I/O Architecture Modeling Based on File System Counters
Salem El Sayed, Matthias Bolten and Dirk Pleiter
Considering I/O Processing in CloudSim for Performance and Energy Evaluation
Hamza Ouarnoughi, Jalil Boukhobza, Frank Singhoff, Stéphane Rubini and Erwann Kassis
Scaling Spark on Lustre
Nicholas Chaimov, Allen Malony, Costin Iancu and Khaled Ibrahim
ORGANIZATION

STEERING COMMITTEE

The steering committee has a French bias, since this workshop took its roots as the extension of a similar initiative already running in France.

jean-thomas-wopsss

JEAN-THOMAS ACQUAVIVA

Research Engineer, DDN, France

Jean-Thomas Acquaviva successively worked for Intel, the University of Versailles and the French Atomic Commission…

jalil-wopsss

JALIL BOUKHOBZA

Assoc. Prof., Univ. Bretagne Occidentale, France

Jalil Boukhobza received his electrical engineering degree (with honors) from I.N.E.L.E.C (Institut Nationale d’Electricité et…

philippe-wopsss

PHLIPPE DENIEL

Head of Lab; CEA/DIF, France

Philippe Deniel, owns an Engineering Degree from Ecole Centrale Paris. Philippe is a Research Engineer…

massimo-wopsss

MASSIMO LAMANNA

Leader, Data & Storage Services, CERN, Switzerland

Massimo Lamanna received a Ph.D. in High-Energy Physics in 1993 at the Trieste University, Italy….

pedro-wopsss

PEDRO JAVIER GARCÍA

Assoc. Prof., University of Castilla-La Mancha, Spain

Pedro J. Garcia is currently an Assistant Professor at the Computing Systems Department of University…

allen-malony-wopsss

ALLEN D. MALONY

Full Professor, University of Oregon

Allen D. Malony is a Professor in the Department of Computer and Information Science at…


PUBLICITY CHAIRS

WOPSSS-jesus_escuderos

JESUS ESCUDERO-SAHUQUILLO

University of Castilla-La Mancha, Spain

WOPSSS-Kaoutar

KAOUTAR EL MAGHREOUI

IBM T. J. Watson Research Center, USA

WOPSSS-Rekha

REKHA SINGHAL

TCS Innovation Labs, Mumbai, India


PROGRAM COMMITTEE

  • Jean-Thomas Acquaviva, Research Engineer, DDN, France
  • Jalil Boukhobza, Associate professor, Univ. Bretagne Occidentale, France
  • Phlippe Deniel, Head of Storage group, CEA/DIF, France
  • Massimo Lamanna, Leader - Data and Storage Services Group, CERN, Switzerland
  • Pedro Javier García, Associated Professor, University of Castilla-La Mancha, Spain
  • Allen D. Malony, Full Professor, University of Oregon, USA
  • Jesus Escudero-Sahuquillo, University of Castilla-La Mancha, Spain
  • Kaoutar El Maghraoui, IBM T. J. Watson Research Center, USA
  • Rekha Singhal, Tata CS Innovation Labs, India
  • Francisco J. Alfaro, University of Castilla-La Mancha, Spain
  • André Brinkmann, Mainz University, Germany
  • Jason Chun Xue, City University of Hong Kong, Hong Kong
  • Toni Cortes, BSC, Spain
  • Stefano Cozzini,CNR, Italy
  • Liu Duo, Chongqing University, China
  • Kaoutar El Maghraoui, IBM T. J. Watson Research Center, USA
  • Maria E. Gomez,Polytechnic University of Valencia, Spain
  • Pilar González-Férez, University of Murcia, Spain
  • J-Charles Lafoucrière, CEA, France
  • Manolis Marazakis, ICS-FORTH, Greece
  • Franz-Josef Pfreundt, Fraunhofer, Germany
  • Juan Piernas, University of Murcia; Spain
  • Rekha Singhal, Tata CS Innovation Labs, India
  • Pierre Vigneras, ATOS, France
  • Josef Weidendorfer, Tech. Univ. of Munich, Germany
  • Soraya Zertal, University of Versailles, France
SPONSOR

POWERED BY THE HPC STORAGE LEADER

DataDirect Networks (DDN) is the world’s leading big data storage supplier to data-intensive, global organizations. For more than 15 years, DDN has designed, developed, deployed and optimized systems, software and solutions that enable enterprises, service providers, universities and government agencies to generate more value and to accelerate time to insight from their data and information, on premise and in the Cloud. Organizations leverage the power of DDN technology and the deep technical expertise of its team to capture, store, process, analyze, collaborate and distribute data, information and content at largest scale in the most efficient, reliable and cost effective manner. DDN customers include many of the world’s leading financial services firms and banks, healthcare and life science organizations, manufacturing and energy companies, government and research facilities, and web and cloud service providers.

VISIT DDN  CONTACT DDN