Differences

This shows you the differences between two versions of the page.

--- io500:bofs:sc19:start [2019/11/15 17:25] – [Agenda] kunkel
+++ io500:bofs:sc19:start [2019/12/20 22:48] – [Agenda] kunkel
@@ Line 42: / Line 42: @@
 We have a series of interactive talks and discussions.
-  * **The Virtual Institute for I/O** -- //Julian Kunkel//
+  * **The Virtual Institute for I/O** -- //Julian Kunkel// \\ [[https://hps.vi4io.org/_media/research/talks/2019/2019-11-19-the_virtual_institute_of_i_o.pdf|Slides]]
-  * **What's new with IO-500** -- //George Markomanolis//
+  * **What's new with IO-500** -- //George Markomanolis// \\ [[https://hps.vi4io.org/_media/research/talks/2019/2019-11-19-io500.pdf|Slides]]
   * **Community lightning talks** (5 minutes each) \\
     * **In-node storage and memory-like I/O** -- //Adrian Jackson (EPCC)// \\ {{ :io500:bofs:sc19:io500-sc19-jackson-innode.pdf |Slides}}
-    * **Demonstrating GPUDirect Storage using the IO500**((Abstract: Accelerators like GPUs are now commonly used in modern HPC systems to relieve computational performance bottlenecks.  As whole workflows are migrated to GPUs, the new bottleneck is IO between storage and GPU memory; the CPU need not be part of the data path.  NVIDIA is introducing GPUDirect Storage to enable GPUs to directly move data into and out of a node's GPUs to storage. To test this capability, the IO500 benchmark suite is being used. This talk will introduce the idea, share some unofficial (or official if done in time) results, the challenges with this technology, and a timeline for when this will be available as a product.)) -- //CJ Newburn, Co-Architect of GPUDirect Storage and Lead for HPC strategy and roadmap for NVIDIA Compute SW and Sven Oehme, Chief Research Officer, DDN//
+    * **Demonstrating GPUDirect Storage using the IO500**((Abstract: Accelerators like GPUs are now commonly used in modern HPC systems to relieve computational performance bottlenecks.  As whole workflows are migrated to GPUs, the new bottleneck is IO between storage and GPU memory; the CPU need not be part of the data path.  NVIDIA is introducing GPUDirect Storage to enable GPUs to directly move data into and out of a node's GPUs to storage. To test this capability, the IO500 benchmark suite is being used. This talk will introduce the idea, share some unofficial (or official if done in time) results, the challenges with this technology, and a timeline for when this will be available as a product.)) -- //CJ Newburn, Co-Architect of GPUDirect Storage and Lead for HPC strategy and roadmap for NVIDIA Compute SW and Sven Oehme, Chief Research Officer, DDN// \\ {{ :io500:bofs:sc19:io500-sc19-newburn.pptx |slides}}
   * **Analysis of the IO-500 data** -- //John Bent//
   * **Announcing Winners** -- //George Markomanolis, John Bent, Julian Kunkel, Jay Lofstead//

Virtual Institute for I/O

User Tools

Site Tools

Differences

Virtual Institute
for
I/O