Parameterized Workload Adaptation for Fork-Join Tasks with Dynamic Workloads and Deadlines

 
Parameterized Workload Adaptation for Fork-Join
Tasks with Dynamic Workloads and Deadlines
 
RTCSA 2023
Niigata, Japan
 
This research was supported in part by NSF grants CSR-1814739, CNS-1763503, CNS-2229290 (CPS),
and NASA grant 80NSSC21K1741.
 
1
 
Marion Sudvarg
, Jeremy Buhler,
Roger Chamberlain, Chris Gill
Computer Science and Engineering
Washington University in St. Louis
 
James Buckley
Department of Physics
Washington University in St. Louis
 
Wenlei Chen
School of Physics and Astronomy
University of Minnesota
 
In dynamic environments, task 
workloads
 and
deadlines 
may not be known prior to job release
May result in system overload – jobs unable to
meet deadlines
Adjust computational parameters to provide an
imprecise result 
prior to the deadline
Anytime workloads
Select from a discrete set of computational modes
2
RTCSA 2023 – Marion Sudvarg
 
But … these semantics may not fully capture the dimensions
over which a task’s workload can adapt to meet its deadline
 
Computations may have multiple parameterized degrees of
freedom (categorical, discrete numeric, continuous numeric)
 
Given an allocation of resources to a task …
 
How do we adjust its computational workload to minimize
degradation of result utility
 
While remaining within the dynamic constraints of
schedulability?
3
RTCSA 2023 – Marion Sudvarg
4
RTCSA 2023 – Marion Sudvarg
 
1.
Identify parameterized degrees of
freedom over which computational
workload may be compressed
2.
Characterize loss over parameter space
3.
Characterize worst-case response times
4.
Reduce parameter space to Pareto-
optimal surface
5.
Online: search for best set of
parameters satisfying constraints
How do we adapt to minimize loss
within schedulability constraints?
 
5
 
RTCSA 2023 – Marion Sudvarg
 
The Advanced Particle-astrophysics Telescope (APT)
 
Motivating
Example
Real-Time Gamma-Ray Burst (GRB) Localization
 
GRBs are rapid, immense bursts of energy
 
We are developing a satellite telescope to
detect and localize GRBs in 
real-time
6
RTCSA 2023 – Marion Sudvarg
Image courtesy NASA, ESA and M. Kornmesser
 (https://www.nasa.gov/feature/goddard/2019/hubble-studies-gamma-ray-burst-with-highest-energy-ever-seen)
 
Will enable secondary instruments to immediately point at the burst
 
Algorithms will run on embedded hardware onboard the satellite
 
 
How can we 
adapt
 and 
compress
 the computational pipeline to maximize localization
accuracy even for bright transients while guaranteeing short deadlines?
7
RTCSA 2023 – Marion Sudvarg
Challenge: Dynamic Workloads and Deadlines
8
RTCSA 2023 – Marion Sudvarg
Adapt to Minimize Loss
within Schedulability Constraints
 
1.
Identify parameterized degrees of freedom over which
computational workload may be compressed
Number of gamma-
rays selected for
reconstruction
Technique
Sample
Size
Number of
Iterations
9
RTCSA 2023 – Marion Sudvarg
Adapt to Minimize Loss
within Schedulability Constraints
 
2.
Characterize loss (due to compression) over parameter space
 
Expected 68% confidence in GRB
localization error
10
RTCSA 2023 – Marion Sudvarg
Adapt to Minimize Loss
within Schedulability Constraints
 
3.
Characterize worst-
case response times
(WCRT)
 
Highly Parallelizable
Fork-Join
Number of gamma-
rays selected for
reconstruction
Technique
Sample
Size
Number of
Iterations
Linear In
Linear In
With
constants
defined by
Quadratic In
Multiplied By
11
RTCSA 2023 – Marion Sudvarg
Adapt to Minimize Loss
within Schedulability Constraints
 
4.
Reduce parameter space to Pareto-optimal subset
 
Sort parameter space by WCRT
 
Reduce to subset of candidate parameter space where 
loss
decreases as 
WCRT
 increases
2657
initial candidate
parameter sets
~80
parameter sets in
Pareto-optimal subset
12
RTCSA 2023 – Marion Sudvarg
Adapt to Minimize Loss
within Schedulability Constraints
 
5.
Online: search for best set of parameters satisfying the
dynamic constraints
 
Binary search over Pareto-optimal parameter subspace to find
best parameter set with WCRT not exceeding deadline
 
Interpolate or extrapolate over parameter + WCRT surface to
find Pareto-optimal parameters with WCRT equal to deadline
Results and Conclusions
 
Offline parameter and loss characterization allows for fast
online workload adaptation for GRB localization!
This approach extends to other dynamic fork-join
workloads with loss functions that are difficult to
characterize in closed form.
Slack reclamation and OpenMP implementation tricks
discussed in paper.
13
RTCSA 2023 – Marion Sudvarg
Simulated 4 known short,
bright GRBs
Worst-case response time
on flight computer:
200ms – 1.3s
Imposed 33ms deadline:
Localized to within 1
degree!
 
Thank You!
 
14
 
RTCSA 2023 – Marion Sudvarg
 
Any Questions?
Methods for Utilization Compression
 
1.
Increase task period: U=C/T
Straightforward when D=T (implicit-deadline)
What if D is fixed (constrained-deadline)?
 
2.
Decrease task workload
How can we adapt task computation while minimizing the
impact on result quality?
 
Pareto-Optimization
15
RTCSA 2023 – Marion Sudvarg
2. Characterize Loss over Parameter Space
16
RTCSA 2023 – Marion Sudvarg
 
 
APT: expected 68% confidence in
GRB localization error
 
Loss may not be easily fit to
closed-form function of
compressible parameters
 
Use union of half spaces,
constrained by parameter bounds
 
Localization Accuracy By Events Reconstructed
 
17
 
RTCSA 2023 – Marion Sudvarg
 
Approximation Accuracy vs Computation
 
18
 
RTCSA 2023 – Marion Sudvarg
3. Characterize Worst-Case Response Times
 
Determine subtask WCETs as function of input parameters
 
Use these to upper-bound task response time given
allocated cores at each tested point in parameter space
 
 
 
We can be less pessimistic!
19
RTCSA 2023 – Marion Sudvarg
 
Event Reconstruction Response Times
 
20
 
RTCSA 2023 – Marion Sudvarg
 
Approximation Response Times
 
21
 
RTCSA 2023 – Marion Sudvarg
 
ApproxCircles
 
FibSpiral
 
Refinement Response Times
 
22
 
RTCSA 2023 – Marion Sudvarg
Improved Processor Assignment
 
If C, L, and D all integers,
Heuristics based on DAG structure (critical path and
workload rooted from each node) often optimal
Express processor assignment as subgraph isomorphism,
often feasible to solve with Glasgow Subgraph Solver
23
RTCSA 2023 – Marion Sudvarg
Find minimum core
assignment for given deadline
Upper bound response time
for given core assignment
 
Dual
M. Sudvarg and C. Gill. “Analysis of Federated Scheduling for Integer-Valued Workloads.” RTNS 2022.
4. Online Pareto-Optimal Compression
 
Sort parameter space by response time
 
Produce Pareto-optimal candidate set – loss monotonically
decreasing as function of response time
APT: reduces state space cardinality from 2657 to ~80
 
Given a dynamic workload and deadline online, find minimum
candidate state under constraints
Linear or quasilinear on size of reduced state space
APT: <250
μ
s on RPi3
 
Apply parameters to computation before execution begins
24
RTCSA 2023 – Marion Sudvarg
 
Slack Reclamation
 
25
 
RTCSA 2023 – Marion Sudvarg
 
Overheads
 
26
 
RTCSA 2023 – Marion Sudvarg
 
Comparisons of
Interpolation, Extrapolation, and Slack Reclamation
 
27
 
RTCSA 2023 – Marion Sudvarg
 
APT Compressed Localization Results: RPi3
 
28
 
RTCSA 2023 – Marion Sudvarg
 
APT Compressed Localization Results: RPi4
 
29
 
RTCSA 2023 – Marion Sudvarg
 
APT Compressed Localization Results: Atom
 
30
 
RTCSA 2023 – Marion Sudvarg
Results and Conclusions
Offline parameter and loss characterization allows for fast online workload
adaptation for GRB localization!
This approach extends to other dynamic fork-join workloads with loss functions
that are difficult to characterize in closed form
Other tricks discussed in paper, e.g., reducing pessimism through slack
reclamation
Future work: federated scheduling of general DAG tasks.
Response times can be characterized given core allocation, but interpolation/extrapolation
may be difficult.
What if core allocations are not fixed a priori?
31
RTCSA 2023 – Marion Sudvarg
Simulated 4 known short,
bright GRBs
Worst-case response time
on flight computer:
200ms – 1.3s
Imposed 33ms deadline:
Localized to within 1
degree!
Slide Note
Embed
Share

In dynamic environments, tasks may face unknown workloads and deadlines, leading to system overload and missed deadlines. This research focuses on adapting task workloads to provide results before the deadline by adjusting computational parameters. It explores how to minimize degradation of result utility while meeting dynamic constraints of schedulability, utilizing multiple degrees of freedom in workload adaptation. Motivated by the Advanced Particle-astrophysics Telescope project, the study aims to optimize computational workload adjustments for real-time applications like Gamma-Ray Burst localization.

  • Workload Adaptation
  • Dynamic Environments
  • Real-Time Systems
  • Computational Parameters
  • Task Deadlines

Uploaded on Oct 01, 2024 | 0 Views


Download Presentation

Please find below an Image/Link to download the presentation.

The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author. Download presentation by click this link. If you encounter any issues during the download, it is possible that the publisher has removed the file from their server.

E N D

Presentation Transcript


  1. Parameterized Workload Adaptation for Fork-Join Tasks with Dynamic Workloads and Deadlines James Buckley Department of Physics Washington University in St. Louis Wenlei Chen Marion Sudvarg, Jeremy Buhler, Roger Chamberlain, Chris Gill Computer Science and Engineering Washington University in St. Louis School of Physics and Astronomy University of Minnesota RTCSA 2023 Niigata, Japan This research was supported in part by NSF grants CSR-1814739, CNS-1763503, CNS-2229290 (CPS), and NASA grant 80NSSC21K1741. 1

  2. In dynamic environments, task workloads and deadlines may not be known prior to job release May result in system overload jobs unable to meet deadlines Adjust computational parameters to provide an imprecise result prior to the deadline Anytime workloads Select from a discrete set of computational modes RTCSA 2023 Marion Sudvarg 2

  3. But these semantics may not fully capture the dimensions over which a task s workload can adapt to meet its deadline Computations may have multiple parameterized degrees of freedom (categorical, discrete numeric, continuous numeric) Given an allocation of resources to a task How do we adjust its computational workload to minimize degradation of result utility While remaining within the dynamic constraints of schedulability? RTCSA 2023 Marion Sudvarg 3

  4. How do we adapt to minimize loss within schedulability constraints? 1. Identify parameterized degrees of freedom over which computational workload may be compressed 2. Characterize loss over parameter space 3. Characterize worst-case response times 4. Reduce parameter space to Pareto- optimal surface 5. Online: search for best set of parameters satisfying constraints RTCSA 2023 Marion Sudvarg 4

  5. Motivating Example The Advanced Particle-astrophysics Telescope (APT) RTCSA 2023 Marion Sudvarg 5

  6. Real-Time Gamma-Ray Burst (GRB) Localization GRBs are rapid, immense bursts of energy We are developing a satellite telescope to detect and localize GRBs in real-time Will enable secondary instruments to immediately point at the burst Algorithms will run on embedded hardware onboard the satellite Image courtesy NASA, ESA and M. Kornmesser (https://www.nasa.gov/feature/goddard/2019/hubble-studies-gamma-ray-burst-with-highest-energy-ever-seen) RTCSA 2023 Marion Sudvarg 6

  7. Challenge: Dynamic Workloads and Deadlines Every GRB is unique! Workload depends on Deadline may depend on Brightness The number of gamma rays entering the detector The duration of the burst (103 106 incident gamma rays) Spectral energy distributions Availability of follow-up instruments Their physical interactions with the detector Initial burst durations (10 milliseconds 20 minutes) Not known prior to burst arrival! How can we adapt and compress the computational pipeline to maximize localization accuracy even for bright transients while guaranteeing short deadlines? RTCSA 2023 Marion Sudvarg 7

  8. Adapt to Minimize Loss within Schedulability Constraints 1. Identify parameterized degrees of freedom over which computational workload may be compressed Event Source Iterative Refinement Reconstruction Approximation Sample Size Number of gamma- rays selected for reconstruction Number of Iterations Technique RTCSA 2023 Marion Sudvarg 8

  9. Adapt to Minimize Loss within Schedulability Constraints 2. Characterize loss (due to compression) over parameter space Expected 68% confidence in GRB localization error RTCSA 2023 Marion Sudvarg 9

  10. Adapt to Minimize Loss within Schedulability Constraints 3. Characterize worst- case response times (WCRT) Linear In Quadratic In Multiplied By Linear In With constants defined by Highly Parallelizable Fork-Join Number of Iterations Sample Size Number of gamma- rays selected for reconstruction Technique RTCSA 2023 Marion Sudvarg 10

  11. Adapt to Minimize Loss within Schedulability Constraints 4. Reduce parameter space to Pareto-optimal subset Sort parameter space by WCRT Reduce to subset of candidate parameter space where loss decreases as WCRT increases 2657 ~80 initial candidate parameter sets parameter sets in Pareto-optimal subset RTCSA 2023 Marion Sudvarg 11

  12. Adapt to Minimize Loss within Schedulability Constraints 5. Online: search for best set of parameters satisfying the dynamic constraints Binary search over Pareto-optimal parameter subspace to find best parameter set with WCRT not exceeding deadline Interpolate or extrapolate over parameter + WCRT surface to find Pareto-optimal parameters with WCRT equal to deadline RTCSA 2023 Marion Sudvarg 12

  13. Results and Conclusions Imposed 33ms deadline: Localized to within 1 degree! Worst-case response time on flight computer: 200ms 1.3s Simulated 4 known short, bright GRBs Offline parameter and loss characterization allows for fast online workload adaptation for GRB localization! This approach extends to other dynamic fork-join workloads with loss functions that are difficult to characterize in closed form. Slack reclamation and OpenMP implementation tricks discussed in paper. RTCSA 2023 Marion Sudvarg 13

  14. Thank You! Any Questions? RTCSA 2023 Marion Sudvarg 14

  15. Methods for Utilization Compression 1. Increase task period: U=C/T Straightforward when D=T (implicit-deadline) What if D is fixed (constrained-deadline)? 2. Decrease task workload How can we adapt task computation while minimizing the impact on result quality? Pareto-Optimization RTCSA 2023 Marion Sudvarg 15

  16. 2. Characterize Loss over Parameter Space APT: expected 68% confidence in GRB localization error Loss may not be easily fit to closed-form function of compressible parameters Use union of half spaces, constrained by parameter bounds RTCSA 2023 Marion Sudvarg 16

  17. Localization Accuracy By Events Reconstructed RTCSA 2023 Marion Sudvarg 17

  18. Approximation Accuracy vs Computation RTCSA 2023 Marion Sudvarg 18

  19. 3. Characterize Worst-Case Response Times Determine subtask WCETs as function of input parameters Use these to upper-bound task response time given allocated cores at each tested point in parameter space We can be less pessimistic! RTCSA 2023 Marion Sudvarg 19

  20. Event Reconstruction Response Times RTCSA 2023 Marion Sudvarg 20

  21. Approximation Response Times ApproxCircles FibSpiral RTCSA 2023 Marion Sudvarg 21

  22. Refinement Response Times RTCSA 2023 Marion Sudvarg 22

  23. Improved Processor Assignment If C, L, and D all integers, Heuristics based on DAG structure (critical path and workload rooted from each node) often optimal Express processor assignment as subgraph isomorphism, often feasible to solve with Glasgow Subgraph Solver Dual Find minimum core assignment for given deadline Upper bound response time for given core assignment M. Sudvarg and C. Gill. Analysis of Federated Scheduling for Integer-Valued Workloads. RTNS 2022. RTCSA 2023 Marion Sudvarg 23

  24. 4. Online Pareto-Optimal Compression Sort parameter space by response time Produce Pareto-optimal candidate set loss monotonically decreasing as function of response time APT: reduces state space cardinality from 2657 to ~80 Given a dynamic workload and deadline online, find minimum candidate state under constraints Linear or quasilinear on size of reduced state space APT: <250 s on RPi3 Apply parameters to computation before execution begins RTCSA 2023 Marion Sudvarg 24

  25. Slack Reclamation Iterative Refinement Approx Source Compress Reconstruct Produce Result Slack? No Yes Refinement Iteration Reconstruct RTCSA 2023 Marion Sudvarg 25

  26. Overheads RTCSA 2023 Marion Sudvarg 26

  27. Comparisons of Interpolation, Extrapolation, and Slack Reclamation RTCSA 2023 Marion Sudvarg 27

  28. APT Compressed Localization Results: RPi3 RTCSA 2023 Marion Sudvarg 28

  29. APT Compressed Localization Results: RPi4 RTCSA 2023 Marion Sudvarg 29

  30. APT Compressed Localization Results: Atom RTCSA 2023 Marion Sudvarg 30

  31. Results and Conclusions Imposed 33ms deadline: Localized to within 1 degree! Worst-case response time on flight computer: 200ms 1.3s Simulated 4 known short, bright GRBs Offline parameter and loss characterization allows for fast online workload adaptation for GRB localization! This approach extends to other dynamic fork-join workloads with loss functions that are difficult to characterize in closed form Other tricks discussed in paper, e.g., reducing pessimism through slack reclamation Future work: federated scheduling of general DAG tasks. Response times can be characterized given core allocation, but interpolation/extrapolation may be difficult. What if core allocations are not fixed a priori? RTCSA 2023 Marion Sudvarg 31

Related


More Related Content

giItT1WQy@!-/#giItT1WQy@!-/#giItT1WQy@!-/#giItT1WQy@!-/#giItT1WQy@!-/#giItT1WQy@!-/#giItT1WQy@!-/#giItT1WQy@!-/#giItT1WQy@!-/#