The document discusses the challenges of exascale computing, including issues related to energy, resilience, and programmability. It proposes providing 1000 researchers with their own petascale systems instead of developing a single exascale system, arguing that this approach would enhance productivity while addressing power and resilience concerns. The author suggests that many applications can effectively utilize petascale computing without the complexities of exascale systems.

1. Operated by Los Alamos National Security, LLC for the U.S. Department of Energy’s NNSA
U N C L A S S I F I E D
An Easy Path to Exascale
Scott Pakin
Applied Computer Science Group
Los Alamos National Laboratory
25 April 2012
Slide 1

2. Operated by Los Alamos National Security, LLC for the U.S. Department of Energy’s NNSA
U N C L A S S I F I E D
Disclaimer
 The opinions expressed herein definitely do not reflect the positions of
Los Alamos National Laboratory, the National Nuclear Security
Administration, or the United States Department of Energy
Slide 2
 Please don’t cut off my funding

3. Operated by Los Alamos National Security, LLC for the U.S. Department of Energy’s NNSA
U N C L A S S I F I E D
Background
 Many of us have been to lots of
exascale meetings
 If these meetings forbade the
use of the words
• energy
• resilience
• programmability
• co-design
 the meetings would be a lot
shorter
Slide 3
 (but perhaps more fun)

4. Operated by Los Alamos National Security, LLC for the U.S. Department of Energy’s NNSA
U N C L A S S I F I E D
Exascale Refrains
 We don’t know how to reduce
system power
 We don’t have a good way to
tolerate frequent faults
 As component heterogeneity
increases, programmability
decreases
Slide 4
Exascale is
hard!

5. Operated by Los Alamos National Security, LLC for the U.S. Department of Energy’s NNSA
U N C L A S S I F I E D
Solution
 How can we deal with all of the
problems of exascale?
Slide 5

6. Operated by Los Alamos National Security, LLC for the U.S. Department of Energy’s NNSA
U N C L A S S I F I E D
My Proposal
Slide 6
 What if instead of building an exascale system, we gave each of 1000
researchers his/her own, private, petascale system?
exascale
petascale

7. Operated by Los Alamos National Security, LLC for the U.S. Department of Energy’s NNSA
U N C L A S S I F I E D
My Proposal
 Power problem solved
 Resiliency problem solved
 Programmability problem solved
 Increased productivity
 Lower cost
Slide 7
 What if instead of building an exascale system, we gave each of 1000
researchers his/her own, private, petascale system?
 Different locations → distribute
load across the power grid
 Fewer components → faults
occur less frequently
 Less complex hardware → can
stick with MPI+X (where X=∅)
 Dedicated systems → no long
queues waiting for job to start
 More aligned with vendor
roadmaps → no cajoling needed

8. Operated by Los Alamos National Security, LLC for the U.S. Department of Energy’s NNSA
U N C L A S S I F I E D
What About Applications that Require Exascale?
 Approaching a singularity where we don’t have any applications
 (except perhaps LINPACK)
Slide 8
Numberofsuitableapps.
Sequential
Small
SMP
Large
NUMA
Distributed
memory
Heterogeneous
architecture
Exascale
(power, resiliency)
Zettascale
(quantum
decoherence?)
Yottascale
(inter-D-brane
interactions on
11-dimensional
M-theoretic
superstrings?)

9. Operated by Los Alamos National Security, LLC for the U.S. Department of Energy’s NNSA
U N C L A S S I F I E D
We May be Facing a Petascale Performance Asymptote
 Parallelism
• By Amdahl’s Law, if your application is less than 99.9% parallelized, you’ll see at
best only petascale performance on an exascale system
 Resilience
• Performing the same work three times gives you 1/3 of the remaining performance
 Power
• To stay within the power budget, some components may have to be turned off or
downclocked, lowering performance even further
1E+0
1E+2
1E+4
1E+6
1E+8
1E+10
1E+12
1E+14
1E+16
1E+3 1E+6 1E+9 1E+12 1E+15 1E+18 1E+21 1E+24
Theoretical peak performance
Application
performance
Mega Giga Tera Peta Exa Zetta Yotta

10. Operated by Los Alamos National Security, LLC for the U.S. Department of Energy’s NNSA
U N C L A S S I F I E D
Summary
 Exascale computing poses numerous challenges
• Energy, resilience, programmability
 Relatively few applications both require exascale and can be
programmed to exploit complex exascale hardware
 Many applications can take advantage of petascale
• Limited by access to machines
• (Long queue if you want the machine for yourself)
 Proposition: Buy 1000 research teams their own petascale computer
• Increases scientists’ productivity
• Why put ourselves through the pain of an exascale before its time?
• I’ll wait until I can borrow an exascale cell phone from my grandkids
 It’s not like there isn’t still exciting research to be done at petascale
Slide 10