Workshop - Heterogeneous and Unconventional Cluster Architectures and Applications (HUCAA) 2018

FYI: HUCAA'18 workshop has been cancelled

The seventh edition of the HUCAA workshop ( is going to take place in conjunction with the 47th International Conference on Parallel Processing (ICPP,, August 13-16, 2018, Eugene, OR. The workshop is scheduled for Aug 13, 2018.


The workshop on Heterogeneous and Unconventional Cluster Architectures and Applications gears to gather recent work on heterogeneous and unconventional cluster architectures and applications, which might have an impact on future mainstream cluster architectures. This includes any cluster architecture that is not based on the usual commodity components and therefore makes use of some special hard- or software elements, or that is used for special and unconventional applications. In particular we call for GPUs and other accelerators (Intel MIC/Xeon Phi, FPGA) used at cluster level. Even though accelerators are already used pervasively, we still see many unconventional and even disruptive uses of them.

Other examples of unconventional cluster architectures and applications include virtualization, in-memory storage, hard- and software interactions, run-times, databases, and device-to-device communication. We are in particular encouraging work on disruptive approaches, which may show inferior performance today but can already point out their  performance potential. The broad scope of the workshop facilitates submissions on unconventional uses of hardware or software, gearing to gather ideas that are coming to life now and not limiting them except for their context: clusters. Also, these proposals may rather be reflective of a broader industry trend.

We are seeking new proposals presented from a holistic perspective. Also, proposals may rather be reflective of a broader industry trend. In this regard, one of the aims of the workshop is anticipating the evolution of clusters. Instead of just presenting new work carried out in the traditional cluster areas usually addressed in other conferences and workshops, we are thinking on creating the right atmosphere for a discussion of opportunities in cluster computing. In this regard, contributions would not only be accepted according to their technical merits but also according to their contribution to this discussion.


For the 2018 edition, HUCAA is explicitly calling for contributions in the areas of:

  • Beyond performance: portability, power, programmability
  • Accelerators and their use in clusters
  • Comprehensive power & energy optimizations
  • Deep learning in clusters

Further topics of interest include any heterogeneous or unconventional cluster architecture or application. Examples include, but are not limited to:

  • Clustered GPUs, FPGAs or other accelerators
  • Runtimes, resource management and scheduling for heterogeneous cluster architectures
  • Communication methods for distributed or clustered accelerators
  • Energy-aware data movement techniques
  • Energy efficiency at the cluster or node level
  • New industry and technology trends and their potential impact
  • High-performance, data-intensive, and power-aware computing
  • Application-specific cluster and datacenter architectures
  • Emerging programming paradigms for parallel heterogeneous computing
  • Software cluster-level virtualization for consolidation purposes
  • Hardware techniques for resource aggregation
  • New uses of GPUs, FPGAs, and other specialized hardware
In the case of questions, please contact the workshop co-chairs.


In 2018, HUCAA is again calling for contributions of a special issue of Wiley's CCPE. For the 2018 workshop edition, the workshop notification dates are aligned with an associated special issue journal submission deadline. Selected workshop contributions are invited to submit extended versions of their paper to the special issue. For more details about this special issue, please refer to:

Furthermore, this special issue is also open to any other contribution that advances science in the areas listed as topics of interest. In case of questions, please contact the SI guest editors.


  • EXTENDED paper submission: April 29, 2018 April 1, 2018
  • Notification of acceptance: May 22, 2018
  • Camera-ready paper: May 31, 2018
  • Workshop: August 13, 2018


Submissions may not exceed 8 pages in PDF format including figures and references, and must be formatted in the ACM format found at: Submitted papers must be original work that has not appeared in and is not under consideration for another conference or journal. Work in progress is welcome, but first results should be made available as a proof of concept. Submissions only consisting of a proposal will be rejected. 



  • Jose Duato, Technical University of Valencia, Spain
  • Sudhakar Yalamanchili, Georgia Tech, USA
  • Ulrich Brüning, U. Heidelberg, Germany


  • Stephan Diestelhorst, ARM, UK
  • Boris Düdder, University of Copenhagen, Denmark
  • Jesús Escudero Sahuquillo, U. Castilla-La Mancha, Spain
  • Simon Gawlok, Heidelberg University, Germany
  • Benjamin Klenk, NVIDIA Research, US
  • Heiner Litz, UC Santa Cruz, CA, US
  • Richard Membarth, DFKI, Germany
  • Lena Oden, FZ Jülich, Germany
  • Dirk Pleiter, Research Center Jülich, Germany
  • Antonio Robles, Universitat Politecnica de Valenia, Spain
  • Sven-Bodo Scholz, Heriot-Watt University, UK
  • Robert Strzodka, Heidelberg University, Germany
  • Shuaiwen Leon Song, PNNL, US
  • Prudence W.H. Wong, U. of Liverpool, UK
  • Huanhuan Xiong, U. College Cork, Ireland
  • Jeff Young, Georgia Tech, GA, US


Contact us using either one or preferable both email addresses:

  • Federico Silla, Technical University of Valencia: fsilla {at}
  • Holger Fröning, University of Heidelberg: holger.froening {at}