BeginnerQuestion 202 of 97Source PDF page undefined

What is the difference between Core Utilization and Local Placement Density?

From PDVerse PnR Interview Handbook · pdVerse Mentor Guide

Short Answer

Core utilization is a single chip-wide number — total cell area (standard cells plus macros) divided by total core area — and it's typically targeted around 65–75% for modern digital blocks as a static planning goal. Local placement density is a much more local, moving-window measurement: cell area inside one G-cell tile divided by that tile's area — and it can spike to 90–98% in one pocket even while 40% of the die sits completely empty.

Technical Reference DiagramWhat is the difference between Core Utilization and Local Placement Density?
What is the difference between Core Utilization and Local Placement Density?, illustrating the physical design concept.

Technical Explanation

  • Core utilization is a single chip-wide number — total cell area (standard cells plus macros) divided by total core area — and it's typically targeted around 65–75% for modern digital blocks as a static planning goal.
  • Local placement density is a much more local, moving-window measurement: cell area inside one G-cell tile divided by that tile's area — and it can spike to 90–98% in one pocket even while 40% of the die sits completely empty.
  • That gap matters because analytical placers naturally cluster timing-critical registers tightly together, and a chip can look perfectly healthy at the global-utilization level while quietly failing at the local level.
  • High local density is dangerous because it drives pin density past what the local metal layers can supply, causing G-cell overflow and unroutable shorts — the fix is to cap max density explicitly (partial blockages or placement-spacing constraints) so the placer is forced to spread cells more evenly.

Common Mistake

The Trap: Believing a chip with 50% global utilization is guaranteed to route cleanly. If the placer clusters 80% of the cells into a narrow corridor, local density exceeds 100% and routes will fail.

Follow-up Question & Model Response

"How do you limit maximum local placement density in ICC2?"

Candidate Model Response: Use set_app_options -name place.coarse.max_density -value 0.70 to cap local density across all G-cells during global placement.

Practical Example

Reporting Density & Setting Density Limits:

# Synopsys ICC2: Cap local placement density at 70%
set_app_options -name place.coarse.max_density -value 0.70

# Report density distribution map
report_placement -density

Physical Design & Planning Handbook

Dive into 14 comprehensive chapters covering netlist sanity, FinFET grids, macro placement, power grids, CTS, and timing budgeting.