University of Crete (UOC)
Timing-driven physical design EDA algorithms for tackling process variations
Abstract
dc:descriptionMoore’s law states that the total number of transistors of an integrated circuit approximatelydoubles every two years. Maintaining this trend, requires tools able to cope with the everincreasingcomplexity of chip design. Electronic Design Automation (EDA) has so far addressedthis problem by providing automated tools and flows which enabled designers to handle chipsconsisting of more than a few millions transistors.However, the ever shrinking of the size of transistors and interconnects, now poses new obstaclesfor designers and automated EDA flows. Smaller dimension devices, although providingmore speed and less area, pose new challenges. Contemporary Deep-Sub-Micron (DSM) fabricationprocesses suffer from the presence of manufacturing variations, due to unpredictability inthe exact dimensions and characteristics of transistors and wires. These variations now affecthigh-level characteristics of the chips such as their speed and power consumption. Technologyvendors have always provided a number of characterizations for each circuit element atdifferent operating scenaria (operating corners). Nowadays, more corners are needed to accountfor process variations, which adds to the complexion of achieving closure for all cornerssimultaneously.One way to mitigate this phenomenon is to integrate multiple operating scenaria into asingle, unified model, which can then be incorporated into existing flows. Statistical modelsoffer this capability. They can encapsulate each corner into a random distribution, which canreflect the variation of speed and power consumption characteristics of the circuit elements. Inthis case, the delay and power becomes statistical rather than deterministic. Although suchstatistical models exist, their use in EDA flows has not been demonstrated. An alternative approachfor combating variations is to design circuits which include clock-less or asynchronousspeed-independent designs. These, possess the property of adjusting to their operating conditionsinstead of failing for fixed constraints. This approach requires further development ofasynchronous circuits, the implementation of which has not been proven viable in EDA flows.Currently, there is significant lack of EDA tools capable of handling asynchronous circuits,making their use impossible in industrial designs.In this work, we have developed and evaluated placement and post-placement optimizationalgorithms, which aim to tackle the problem of process variations in contemporary EDA flows.We present a novel placement algorithm, SCPlace, which based on a statistical timing modelin its optimization engine, alleviates the need for multi-corner placement. SCPlace is the first large-scale statistical optimization tool appearing in literature targeting placement, which is thecornerstone of physical implementation. SCPlace exploits statistical wire delay bounds, generatedby our novel statistical slack assignment algorithms, which distribute slack according tostatistical distributions. We have also developed a post-placement statistical leakage reductionalgorithm, which is able to perform in-place statistical leakage reduction without negativelyaffecting statistical delay. Our third contribution is CPlace, a fully automated placer for asynchronous,cyclic circuits. CPlace is able to meet both performance and speed independentconstraints.Experimental results indicate that SCPlace compares favourably with state-of-the-art, industrialand academic placers, providing routable designs which achieve superior timing yieldcomputed from the resulting statistical delay distributions. Our statistical leakage reductionflow achieves 20% average leakage reduction, without affecting the statistical delay of the preplacedcircuit. Our results also show that CPlace provides routable placements for asynchronouscircuit and superior placements compared to state-of-the-art industrial and academic placerswhich cannot guarantee speed independent constraints. All three of our flows have been designedwith ease of integration into contemporary EDA flows in mind, through the use of onlyindustry-standard formats and by collaborating with commercial EDA tools.
Degree
thesis:*- Grantor dc:publisher
- University of Crete (UOC)
- Year dc:date
- 2011
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Kounalakis, Evriklis
- Κουναλάκης, Ευρυκλής
Subjects
dc:subject × 20- Ηλεκτρονικός σχεδιασμός
- Μεταβλητότητα
- Χρονισμός
- Κυκλώματα
- Τοποθέτηση
- Ασύγχρονα
- Ρεύμα διαρροής
- Στατιστική
- EDA
- Design automation
- Variability
- Placement
- Timing
- Asynchronous
- Leakage
- Statistical
- Επιστήμες Μηχανικού και Τεχνολογία
- Επιστήμη Ηλεκτρολόγου Μηχανικού, Ηλεκτρονικού Μηχανικού, Μηχανικού Η/Υ
- Engineering and Technology
- Electrical Engineering, Electronic Engineering, Information Engineering
Rights
- Language dc:language
- eng
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/27553