TTC 2018 SOLUTION PRESENTATION A JastAdd- and ILP-based Solution to - - PowerPoint PPT Presentation

▶
ttc 2018 solution presentation
SMART_READER_LITE
LIVE PREVIEW

TTC 2018 SOLUTION PRESENTATION A JastAdd- and ILP-based Solution to - - PowerPoint PPT Presentation

Jul 28, 2023 45 likes •199 views

, TTC 2018 SOLUTION PRESENTATION A JastAdd- and ILP-based Solution to the Software-Selection and Hardware-Mapping-Problem Sebastian Gtz, Johannes Mey, Rene Schne and Uwe Amann TTC 2018 Solution Presentation Slide 1 of 15 , Idea Step 1

slide-1
SLIDE 1

,

TTC 2018 SOLUTION PRESENTATION

A JastAdd- and ILP-based Solution to the Software-Selection and Hardware-Mapping-Problem Sebastian Götz, Johannes Mey, Rene Schöne and Uwe Aßmann

TTC 2018 Solution Presentation Slide 1 of 15

slide-2
SLIDE 2

,

Idea

Step 1

  • Transform problem into Integer Linear Program (ILP)

Step 2

  • Solve ILP

TTC 2018 Solution Presentation Slide 2 of 15

slide-3
SLIDE 3

,

Idea

TTC 2018 Solution Presentation Slide 3 of 15

slide-4
SLIDE 4

,

Step 1: Problem → ILP

  • Utilize reference attribute grammars

– Transform model with higher order attributes:

  • 1. Extend grammar to store ILP as subtree
  • 2. Transform problem into ILP using nonterminal attributes
  • 3. Use subtree to load ILP into solver
  • 4. Profit

TTC 2018 Solution Presentation Slide 4 of 15

slide-5
SLIDE 5

,

Step 2: ILP → Solution

  • Enumerate all mappings
  • Assing one binary variable to choose an implementation for a request

deploying it on a resource Name = request#implA0#res2

TTC 2018 Solution Presentation Slide 5 of 15

slide-6
SLIDE 6

,

Validity and Optimization

Architectural constraints

  • Max. 1 Impl/Component and max. 1 Impl/Resource

Request constraints

  • Request target component chosen fulfilling requirements

Negotiation constraints

  • Deploy required components
  • Fulfill component requirements

Optimize generation

  • Exclude mappings with invalid hardware constraints

TTC 2018 Solution Presentation Slide 6 of 15

slide-7
SLIDE 7

,

ILP intermediate model (Model)

TTC 2018 Solution Presentation Slide 7 of 15

slide-8
SLIDE 8

,

ILP intermediate model (Text form)

Minimize 123 req0#implA1#r1 + 125 req0#implA1#r2 Subject To req0#implA0+ req0#implA1 <= 1 Bounds 0 <= req0#implA1#r1 <= 1 Generals req0#implA1#r1 End

TTC 2018 Solution Presentation Slide 8 of 15

slide-9
SLIDE 9

,

Backup

TTC 2018 Solution Presentation Slide 9 of 15

slide-10
SLIDE 10

,

Grammar Hardware

TTC 2018 Solution Presentation Slide 10 of 15

slide-11
SLIDE 11

,

Grammar Expression

TTC 2018 Solution Presentation Slide 11 of 15

slide-12
SLIDE 12

,

Grammar Software

TTC 2018 Solution Presentation Slide 12 of 15

slide-13
SLIDE 13

,

Grammar General

TTC 2018 Solution Presentation Slide 13 of 15

slide-14
SLIDE 14

,

Grammar Solution

TTC 2018 Solution Presentation Slide 14 of 15

slide-15
SLIDE 15

,

Measurement results

= valid and in time = valid, but timeout = invalid = optimal (if known from ILP solver) Scenario ACO EMFeR ILP (direct/ext) Simple 0 trivial 6 194 24 / 21 1 1 small 8 / 212 37 / 40 6 2 small-hw 11 240 44 / 61 8 3 small-sw 451 7min52s 377 / 572 15min 4 medium 1min33 / 8min22s 8min28s / 15min 5 medium-hw 4min48s 11min15s 15min / 15min 6 medium-sw 15min 11min15s 15min 15min

TTC 2018 Solution Presentation Slide 15 of 15