By Topic

Efficient Package Pin-Out Planning With System Interconnects Optimization for Package-Board Codesign

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$31 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

2 Author(s)
Ren-Jie Lee ; Dept. of Electron. Eng., Nat. Chiao Tung Univ., Hsinchu, Taiwan ; Hung-Ming Chen

In conventional package design, engineers designate the ball grid array (BGA) pin-out manually, this always postpones the time-to-market (TTM) of products due to the turn-around between package and design houses. Recent papers propose a method of automatically generating the pin-out and taking signal integrity (SI), power delivery integrity (PI), and routability (RA) into account simultaneously by pin-block design and floorplanning, thus dramatically speeding up the developing time. However, this approach ignores the considerations of shorter path length and equilength/length matching in routing printed circuit board (PCB) trace and pin-out assignment for high-speed interface IP designs, such as USB and PCI Express. Since these features are the most important performance metrics during chip-package-board codesign, in this paper we propose the ideas to optimize the system interconnects during package pin-out design. These ideas keep the same minimized package size as aforementioned recent work and ensure that SI, PI, and RA can still be considered with significant reduction in design cost. It is achieved by relaxing the restriction of pin-block side and order on the package, usually specified by package designers. The experimental results on industrial chipset design cases show that the average improvement of our pin-block planner is over 40% when comparing the design cost with the previous work, among which we have one case accommodated over a thousand pins. Our ideas also work for any kind of pin-block or pin-group configurations.

Published in:

Very Large Scale Integration (VLSI) Systems, IEEE Transactions on  (Volume:19 ,  Issue: 5 )