PLEXOS Breakout Discussion on Major PLEXOS Enhancements Why - - PowerPoint PPT Presentation

PLEXOS Breakout

Discussion on Major PLEXOS Enhancements

2 2

Why Parallelize?

• Faster, because it will…
• use more of the available CPU
• utilize cloud compute resources
• Simulation speed a function of computer specification:
• number of cores
• amount of memory
• rather than simply clock speed.
• Can utilize cheaper computing resources e.g. many

parallel small Linux-based compute nodes, rather than expensive desktops

3

What can/cannot be parallelized?

Model Phase Steps Samples Solver

Models Batch/Connect Phases LT/MT/ST must run in sequence Steps This is the new feature! Samples Monte Carlo samples and scenarios trees Solvers Use multiple threads

4

How can steps be parallelized?

• New option:

ST Schedule [Step Link Mode] = “Parallel”

• No overall loss of functionality
• Greater use of MT Schedule solution:
• Storage start points
• Storage targets
• Constraint decomposition
• Minor loss of fidelity e.g. initial generator positions,

constraint “roll over”

• Huge potential speed up
• Very cloud friendly and cost effective

5

Step Link Mode = “Link” (default mode)

MT Schedule Step 1 Step 2 Step 3 Step 4 Step N

Reoptimize

ST Schedule

Setup Solve Setup Solve Setup Solve Setup Solve Setup Solve

Target 1 Target 2 Target 3 Target 4 Target N

6

Step Link Mode = “Break”

MT Schedule Step 1 Step 2 Step 3 Step 4 Step N

Target 1 Target 2 Target 3 Target 4 Target N Reoptimize

ST Schedule

Setup Solve Setup Solve Setup Solve Setup Solve Setup Solve

Start Point 2 Start Point 3 Start Point 4 Start Point N

7

Step Link Mode = “Parallel”

• Same MT/ST

coordination approach as “Break” links mode, but

• steps execute

asynchronously.

• Slower steps do not

hold up faster ones

• Solution written as it

becomes available

Setup Solve Setup Setup Setup Solve Solve Solve

Available CPU resources distributed between multiple steps Can utilize remote compute nodes for unlimited parallelization

8

Maximizing Performance

Limit Filtering

• New option

Transmission [Limit Filtering]:

• Use MT Schedule

to estimate flows

• Filter out

constraints + contingencies unlikely to bind

Transmission Iterations

• New method!
• Line, Transformer,

Interface, Contingency solved using primal feasibility test (decomposition)

• Work done at

solver layer not in business logic

Unserved & Dump Energy

• Can ‘bloat’ the

model making it very slow to solve, so…

• New method
• Initially keep these

terms out of the formulation, and

• do dual feasibility

test

• determines when

variables needed

9

Benchmarks

Ngrid Pan-European Model - 5 Years, Hourly Resolution, Weekly Optimization Step

Time Memory (MB) Memory (MB)/CPU Objective Sequential (8.2) 03:24:45 14,447.56 300.99

• 3.17E+11

Parallel (8.2) 00:21:10 44,239.10 921.65

• 3.17E+11

Change 9.7 × faster 3.1 × more

• 0.0012%

Fortum Pan-European Model - 36 Years, Hourly Resolution, Weekly Optimization Step

Time Memory (MB) Memory (MB)/CPU Objective Sequential (8.2) 68:57:26 27,924.85 581.77 3.39E+15 Parallel (8.2) 06:01:11 173,829.92 3,621.46 3.39E+15

Change 11.5 × faster 6.2 × more

• 0.0016%

PLEXOS World - 1 Year, Hourly Resolution, Daily Optimization Step

Time Memory (MB) Memory (MB)/CPU Objective Sequential (8.2) 31:23:52 31,021.96 646.29 2.29E+13 Parallel (8.2) 02:56:51 116,826.34 2,433.88 2.28E+13

Change 10.7 × faster 3.8 × more

• 0.7253%

Eastern Interconnect (MMWG) - 1 Year, Hourly Resolution, Daily Optimization Step

Time Memory (MB) Memory (MB)/CPU Objective Sequential (7.5) 11:34:18 39,368.74 820.18 1.50E+10 Parallel (8.2) 02:08:18 89,787.43 1,870.57 1.55E+10

Change 5.4 × faster 2.3 × more

3.1606%

Eastern Interconnect (Full) - 1 Year, Hourly Resolution, Daily Optimization Step

Time Memory (MB) Memory (MB)/CPU Objective Sequential (8.2) 152:53:03 36,888.50 768.51 5.32E+10 Parallel (8.2) 08:12:45 174,939.11 3,644.56 5.32E+10

Change 18.6 × faster 4.7 × more

• 0.0001%

10

Getting Started with PLEXOS 8.2

Transmission [Limit Filtering] (for Nodal customers) Reduce ST Schedule runtime ST Schedule [Step Link Mode] = “Break” Are you happy with the results compared to “Link” mode? ST Schedule [Step Link Mode] = “Parallel” Same results as “Break” but much faster Compute Hardware Correct balance of CPU and memory for your case Cloud compute environment Practically unlimited parallelization