Theme
Government agencies and companies are looking for a new generation of tools able to efficiently solve large scale combinatorial problems for applications ranging from big data science to national security. This BoF aims at gathering the community of people interested at frameworks and workflows for large scale graph analytics, surveying the current approaches, identifying new challenges and opportunities, and laying a path towards future interoperable infrastructures. The BoF will discuss GraphBLAS as a key component of such toolkits and bring together the GraphBLAS community. We will report: (1) the essential design patterns using GraphBLAS, and (2) requirements for GraphBLAS 2.0.
Please, join us for the first edition of this BOF that will feature two - raucous - panels that will:
1. Identify critical applications needs (for DOE and DOD), opportunities, issues, and challenges in creating interoperable HPC Graph Toolkits able to address current and future graph and combinatorial workloads (including property graphs and integrated workflows with machine learning)
2. distill current design patterns, advantages, limits, issues of GraphBLAS and requirements and needs for GraphBLAS 2.0
| Time | Event |
|---|---|
| 5.15 - 5.20 | State of the Union (Welcome and Introduction) Aydin Buluç (LBNL), Antonino Tumeo (PNNL) |
| 5.20 - 5.30 | GraphBLAS: Status and Outlook José Moreira (IBM) |
| 5.30 - 5.40 | HPC Graph Toolkits (or lack thereof!) Mahantesh Halappanavar (PNNL) |
| 5.40 - 6.20 | Panel 1: Message can't solve your graph problems Moderator: John Feo (PNNL/NIAC) Panelists: David Bader (Georgia Tech), Joe Eaton (NVIDIA), Fabrizio Petrini (Intel), Alex Pothen (Purdue) |
| 6.20 - 7.00 | Panel 2: We have the GraphBLAS, so we're done since they can do everything. Moderator: John Gilbert (UCSB) Panelists: Andrew Lumsdaine (PNNL/NIAC), Kamesh Madduri (PSU), Tim Mattson (Intel), Michael M Wolf (SNL) |