ADAPT'16 keynote (14:00-15:00): "Benchmarks vs The Zombie Apocalypse: A comparison" by Ed Plowman (Director of Performance Analysis Strategy at ARM).

Congratulations to Andrei Lascu, PhD student from Imperial College London, who have received dividiti award - 200 euros cash for the artifacts shared in the open Collective Knowledge format!

Program (Nadir room)

10:15-10:30	ADAPT'16 introduction
10:30-11:00	Iterative compilation on mobile devices Paschalis Mpeis, Pavlos Petoumenos, Hugh Leather University of Edinburgh, UK [ Reddit , ArXiv ]
11:00-11:30	Coffee break
11:30-12:00	Software-Based Fault Recovery via Adaptive Diversity for COTS Multi-Core Processors Andrea Höller, Tobias Rauter, Johannes Iber, Georg Macher, Christian Kreiner Institute for Technical Informatics, Graz University of Technology, Austria [ Reddit , ArXiv ]
12:00-12:30	A lightweight optimization selection method for Sparse Matrix-Vector Multiplication Athena Elafrou, Georgios Goumas, Nectarios Koziris National Technical University of Athens, Greece [ Reddit , ArXiv ]
12:30-13:00	Integrating a large-scale testing campaign in the CK framework Andrei Lascu, Alastair F. Donaldson Imperial College London, UK [ Reddit , ArXiv , Shared experiment workflow]
13:00-14:00	Lunch
14:00-15:00	ADAPT'16 keynote: "Benchmarks vs The Zombie Apocalypse: A comparison" Ed Plowman (Director of Performance Analysis Strategy at ARM) A satirical take on the state of the current benchmark eco system and action you can take to avoid the potential apocalypse!
15:00-15:30	GEMMbench: a framework for reproducible and collaborative benchmarking of matrix multiplication Anton Lokhmotov dividiti, UK [ Reddit , ArXiv , Shared experiment workflow ]
15:30-16:00	Coffee break
16:00-16:30	Autotuning OpenCL Workgroup Size for Stencil Patterns Chris Cummins, Pavlos Petoumenos, Michel Steuwer, Hugh Leather University of Edinburgh, UK [ Reddit , ArXiv ]
16:30-18:00	Public discussion: enabling efficient self-tuning computer systems via collaborative and reproducible computer engineering News about open Collective Knowledge Framework to share benchmarks, data sets, tools and predictive models as reusable components with unified JSON API and quickly prototype and crowdsource experiments (such as customizable and multi-objective OpenCL/CUDA/compiler flag autotuning and run-time adaptation - see our upcoming DATE'16 paper , MULTIPROG'16 article and CK-based interactive report). News about related Artifact Evaluation from CGO/PPoPP'16. Announcement of the dividiti award for ADAPT'16 reproducible and reusable artifact. Public discussion about our open reviewing and publication model and the future of our ADAPT workshop! You may be interested to check our original motivation paper and several other interesting public discussions about our approach at metafilter and slashdot. Looking forward to your participation!

Sponsors and supporters

-

DIVIDITI provided cash prize of 200 euros for the highest-ranked artifact shared in CK format