1 / 24

SimPL : An Effective Placement Algorithm

SimPL : An Effective Placement Algorithm. Myung-Chul Kim, Dong-Jin Lee and Igor L. Markov Dept. of EECS, University of Michigan. Global Placement: Motivation. Interconnect lagging in performance while transistors continue scaling

sofia
Télécharger la présentation

SimPL : An Effective Placement Algorithm

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. SimPL: An Effective Placement Algorithm Myung-Chul Kim, Dong-Jin Lee and Igor L. Markov Dept. of EECS, University of Michigan ICCAD 2010, Myung-Chul Kim, University of Michigan

  2. Global Placement: Motivation Interconnect lagging in performance while transistors continue scaling Circuit delay, power dissipation and areadominated by interconnect Routing quality highly controlled by placement Circuit size and complexity rapidly increasing Scalable placement algorithm is critical Simplicity, integration with other optimizations IR drop Coupling RC delay Unloaded ICCAD 2010, Myung-Chul Kim, University of Michigan

  3. Placement Formulation Objective: Minimize estimated wirelength (half-perimeter wirelength) Subject to constraints: Legality: Row-based placement with no overlaps Routability: Limiting localinterconnect congestion forsuccessful routing Timing: Meeting performancetarget of a design ICCAD 2010, Myung-Chul Kim, University of Michigan

  4. Prior Work Ideal Placer Fastruntimewithout sacrificing solution quality Simplicity, integration with other optimization Ideal placer Speed mFAR, Kraftwerk2, FastPlace3 Quadratic and force-directed mPL6, APlace2, NTUPlace3 Non-convex optimization Solution Quality ICCAD 2010, Myung-Chul Kim, University of Michigan

  5. Key features of SimPL Flat quadratic placement Primal dual optimization Closing the gap between upper and lower bounds Upper-Bound Solution byLook-ahead Legalization Wirelength Final Solution Final Legal Solution Lower-Bound SolutionbyLinear System Solver Initial WL Opt. Iteration

  6. Placement Instance Converge Common Analytical Placement Flow Initial WLOptimization GlobalPlacement no yes Legalization and Detailed Placement ICCAD 2010, Myung-Chul Kim, University of Michigan

  7. Placement Instance WLConverge Converge SimPL Flow Initial WLOptimization no B2B GraphBuilding Linear System Solver yes Look-aheadLegalization(Upper-Bound) GlobalPlacement Pseudonet Insertion no yes B2B GraphBuilding Legalization and Detailed Placement Linear System Solver (Lower-Bound) B2B net model[P. Spindler, et al, “Kraftwerk2 - A Fast Force-Directed Quadratic Placement Approach Using an Accurate Net Model,” TCAD 2008] We delegate final legalization and detailed placement to FastPlace-DP [M. Pan, et al, “An Efficient and Effective Detailed Placement Algorithm”, ICCAD2005]

  8. SimPL: Look-ahead Legalization Purpose: Produces almost-legal placement (Upper-Bound) while preserving the relative cell ordering given by linear system solver (Lower-Bound) Identify target region Find overflow bin b Create a minimal wide enough bin cluster B around b Perform geometric top-down partitioning Find cell area median (Cc) and whitespace median (CB) Assign cells (Cc) to corresponding partitions (CB) Non-linear scaling Form stripe regions Move cells across stripe regions in-order based on whitespace

  9. SimPL: Look-ahead Legalization (1) Performing geometric top-down partitioning Cell-area median (Cc) whitespacemedian (CB) Overfilled bin B1 Bin cluster (B) B0 ICCAD 2010, Myung-Chul Kim, University of Michigan

  10. SimPL: Look-ahead Legalization (2) Cell-area median (Cc) whitespacemedian (CB) B0 ICCAD 2010, Myung-Chul Kim, University of Michigan

  11. SimPL: Look-ahead Legalization (2) CB CB CB 4 4 7 7 3 3 1 1 5 5 8 8 6 6 2 2 Per-stripeLinear Scaling CellOrdering Uniform cutlines Obstacle borders ICCAD 2010, Myung-Chul Kim, University of Michigan

  12. SimPL: Look-ahead Legalization (3) Example (adaptec1) Look-ahead legalization stops when target regions become small enough

  13. SimPL: Using legal locations as anchors Purpose: Gradually perturb the linear system to generate lower-bound solutions with less overlap Anchors and Pseudonets Look-ahead locations used as fixed, zero-area anchors Anchors and original cells connected with 2-pin pseudonets Pseudonet weights grow linearly with iterations ICCAD 2010, Myung-Chul Kim, University of Michigan

  14. Next illustration: Tug-of-war between low-wirelength and legalized placements ICCAD 2010, Myung-Chul Kim, University of Michigan

  15. SimPL Iterations on Adaptec1 (1) Iteration=0 (Init WL Opt.) Iteration=1 (Upper Bound) Iteration=2 (Lower Bound) Iteration=3 (Upper Bound)

  16. SimPL Iterations on Adaptec1 (2) Iteration=10 (Lower Bound) Iteration=11 (Upper Bound) Iteration=11 (Upper Bound) Iteration=20 (Lower Bound) Iteration=20 (Lower Bound) Iteration=21 (Upper Bound) Iteration=21 (Upper Bound)

  17. SimPL Iterations on Adaptec1 (3) Iteration=30 (Lower Bound) Iteration=31 (Upper Bound) Iteration=40 (Lower Bound) Iteration=41 (Upper Bound)

  18. Convergence of SimPL Legal solution is formed between two bounds ICCAD 2010, Myung-Chul Kim, University of Michigan

  19. Empirical Results: ISPD05 Benchmarks Experimental setup Single threaded runs on a 3.2GHz Intel core i7 Quad CPU Q660 Linux workstation HPWL is computed by GSRC Bookshelf Evaluator < 5000 lines of code in C++, including CG-based solver for sparse linear systems with Jacobi preconditioner Improvements after ICCAD submission ICCAD 2010, Myung-Chul Kim, University of Michigan

  20. Empirical Results: Scalability Study Take an existing design (ISPD 2005) and split each movable cell into two cells of smaller size Each connection to the original cell is inherited byone of two split cells, which are connected by a 2-pin net Not in ICCAD paper

  21. Parallelism in Conjugate Gradient Solver Runtime bottleneck in SimPL: Conjugate gradient linear system solver Coarse-grain row partitioning Implemented using OpenMP3.0 compiler intrinsic SSE2 (Streaming SIMD Extensions) instructions Process 4 multiple data with a single instruction Marginal runtime improvement in SpMxV Reducing memory bandwidth demand of SpMxV CSR (Compressed Sparse Row) format Y. Saad, “Iterative Methods for Sparse Linear Systems,” SIAM 2003 ICCAD 2010, Myung-Chul Kim, University of Michigan

  22. On-going Research Integration with physical synthesis Look-ahead placement offers opportunity for early estimation of circuit parameters Timing look-ahead Congestion look-ahead Power-density look-ahead Improving the speed and quality of physical synthesis Parallel look-ahead legalization Run independently in separate sub-regions ICCAD 2010, Myung-Chul Kim, University of Michigan

  23. Conclusions New flat quadratic placement algorithm: SimPL Novel primal-dual approach Amenable to integration with physical synthesis Self-contained, compact implementation Fastest among available academic placers Highly competitive solution quality ICCAD 2010, Myung-Chul Kim, University of Michigan

  24. Questions and Answers Thank you! Time for Questions ICCAD 2010, Myung-Chul Kim, University of Michigan

More Related