Routing and Algorithms For VLSI design.pptx

634 views 43 slides Apr 01, 2024
Slide 1
Slide 1 of 43
Slide 1
1
Slide 2
2
Slide 3
3
Slide 4
4
Slide 5
5
Slide 6
6
Slide 7
7
Slide 8
8
Slide 9
9
Slide 10
10
Slide 11
11
Slide 12
12
Slide 13
13
Slide 14
14
Slide 15
15
Slide 16
16
Slide 17
17
Slide 18
18
Slide 19
19
Slide 20
20
Slide 21
21
Slide 22
22
Slide 23
23
Slide 24
24
Slide 25
25
Slide 26
26
Slide 27
27
Slide 28
28
Slide 29
29
Slide 30
30
Slide 31
31
Slide 32
32
Slide 33
33
Slide 34
34
Slide 35
35
Slide 36
36
Slide 37
37
Slide 38
38
Slide 39
39
Slide 40
40
Slide 41
41
Slide 42
42
Slide 43
43

About This Presentation

Routing algorithms of VLSI Design


Slide Content

Routing and Algorithms By Prateek Tripathi VLSI Physical Design: From Graph Partitioning to Timing Closure; Andrew B. Kahng, Jens Lienig

Basic Idea Global Routing: Global routes assign nets to particular metal layers and global routing cells . in an approximate manner. Detailed Routing: For detailed routing, the router decides the actual physical interconnections of nets by allocating wires on each metal layer and vias for switching between metal layers. VLSI Physical Design: From Graph Partitioning to Timing Closure; Andrew B. Kahng, Jens Lienig

VLSI Physical Design: From Graph Partitioning to Timing Closure; Andrew B. Kahng, Jens Lienig

Some points to remember Vertical and horizontal tracks are laid out on different layers The junctions are connected by a “via”. There can be multiple layers of tracks depending upon the complexity of the routing problem. As a conventional routing method we label the two or more pins with the same name to specify the program that a connection is to be form between them. VLSI Physical Design: From Graph Partitioning to Timing Closure; Andrew B. Kahng, Jens Lienig

Optimization goal of Global Routing Seeks to determine whether a given placement is routable Seeks to determine a coarse routing for all nets within available routing regions. VLSI Physical Design: From Graph Partitioning to Timing Closure; Andrew B. Kahng, Jens Lienig

Grid Graph Model Illustration VLSI Physical Design: From Graph Partitioning to Timing Closure; Andrew B. Kahng, Jens Lienig

Channel connectivity graph VLSI Physical Design: From Graph Partitioning to Timing Closure; Andrew B. Kahng, Jens Lienig For horizontal channel, draw a line parallel to x-axis along the block edges which are parallel to x-axis, unless stopped by a block. 2. For vertical channel, draw a line parallel to y-axis along the block edges which are parallel to y-axis., unless stopped by a block.

Switch-box VLSI Physical Design: From Graph Partitioning to Timing Closure; Andrew B. Kahng, Jens Lienig

Single net routing(rectilinear routing) then the tree is a rectilinear minimum spanning tree(RMST) VLSI Physical Design: From Graph Partitioning to Timing Closure; Andrew B. Kahng, Jens Lienig

Further points The total edge length LRSMT of the RSMT is at least half the perimeter of the minimum bounding box of the net: LRSMT >= LMBB/2 VLSI Physical Design: From Graph Partitioning to Timing Closure; Andrew B. Kahng, Jens Lienig

VLSI Physical Design: From Graph Partitioning to Timing Closure; Andrew B. Kahng, Jens Lienig

Hanan grid VLSI Physical Design: From Graph Partitioning to Timing Closure; Andrew B. Kahng, Jens Lienig For each points draw a lines perpendicular to y and x axis which passes through the points Mark all the intersections of those lines as Hannan points

A sequential steiner tree heuristic VLSI Physical Design: From Graph Partitioning to Timing Closure; Andrew B. Kahng, Jens Lienig T: Tree

VLSI Physical Design: From Graph Partitioning to Timing Closure; Andrew B. Kahng, Jens Lienig

Steps to global routing in connectivity graph 1. Define the routing regions 2. Define the connectivity graph 3. Determine the net order( can be prioritised based on no. of pins, criticality, size of bounding box ) 4. Assigning tracks for all pin connections( for each pin a horizontal and a vertical track are reserved ) 5. Global routing of all nets VLSI Physical Design: From Graph Partitioning to Timing Closure; Andrew B. Kahng, Jens Lienig

Step 1 Step 2

Illustration of assigning capacities to the nodes VLSI Physical Design: From Graph Partitioning to Timing Closure; Andrew B. Kahng, Jens Lienig

Final graph for previous slide VLSI Physical Design: From Graph Partitioning to Timing Closure; Andrew B. Kahng, Jens Lienig

Now coming to global routing(step 5) VLSI Physical Design: From Graph Partitioning to Timing Closure; Andrew B. Kahng, Jens Lienig

Soln.

Try yourself

Routing by Integer Linear Programming A linear program of a set of constraints and an optional objective function. Objective function is maximised or minimised Constraints and objective must be linear. Constraints form a system of linear equation and inequalities An ILP is a linear program where every variable assume integer value. VLSI Physical Design: From Graph Partitioning to Timing Closure; Andrew B. Kahng, Jens Lienig

Routing by Integer Linear Programming VLSI Physical Design: From Graph Partitioning to Timing Closure; Andrew B. Kahng, Jens Lienig

VLSI Physical Design: From Graph Partitioning to Timing Closure; Andrew B. Kahng , Jens Lienig

Horizontal Constraint Graph Draw all the horizontal Connections, each on a Single line

VLSI Physical Design: From Graph Partitioning to Timing Closure; Andrew B. Kahng, Jens Lienig

These graphs are used to detect conflict between routing paths and minimum number of routing paths VLSI Physical Design: From Graph Partitioning to Timing Closure; Andrew B. Kahng, Jens Lienig

Vertical constraint graph VLSI Physical Design: From Graph Partitioning to Timing Closure; Andrew B. Kahng, Jens Lienig

Vertical constraint graph VLSI Physical Design: From Graph Partitioning to Timing Closure; Andrew B. Kahng, Jens Lienig

Vertical constraint graph VLSI Physical Design: From Graph Partitioning to Timing Closure; Andrew B. Kahng , Jens Lienig

CYCLE CONFLICT VLSI Physical Design: From Graph Partitioning to Timing Closure; Andrew B. Kahng, Jens Lienig

Left edge algorithm VLSI Physical Design: From Graph Partitioning to Timing Closure; Andrew B. Kahng, Jens Lienig

Left edge algorithm Soln. VLSI Physical Design: From Graph Partitioning to Timing Closure; Andrew B. Kahng , Jens Lienig

Soln continued From horizontal constraint graph we figured out that we need at least 5 tracks VLSI Physical Design: From Graph Partitioning to Timing Closure; Andrew B. Kahng, Jens Lienig

Final soln. VLSI Physical Design: From Graph Partitioning to Timing Closure; Andrew B. Kahng, Jens Lienig

Dogleg algorithm VLSI Physical Design: From Graph Partitioning to Timing Closure; Andrew B. Kahng, Jens Lienig

Dogleg Algorithm After splitting of the net, it follows the left edge algorithm Example: Solve the given figure with dogleg algorithm VLSI Physical Design: From Graph Partitioning to Timing Closure; Andrew B. Kahng , Jens Lienig

Solution VLSI Physical Design: From Graph Partitioning to Timing Closure; Andrew B. Kahng, Jens Lienig

END