2005 International Conference on Integrated Circuit Design and Technology, 2005. ICICDT 2005. 2005
DOI: 10.1109/icicdt.2005.1502575
|View full text |Cite
|
Sign up to set email alerts
|

Clock distribution on a dual-core, multi-threaded Itanium/spl reg/ family microprocessor

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1

Citation Types

0
3
0

Year Published

2007
2007
2023
2023

Publication Types

Select...
4
2
1

Relationship

0
7

Authors

Journals

citations
Cited by 7 publications
(3 citation statements)
references
References 1 publication
0
3
0
Order By: Relevance
“…Comparable solutions offer similar or worse levels of skew reduction, sub-10 ps in a 90-nm technology [64], 70 ps in a 0.18-m technology [2] and 15 ps in a 0.25-m technology [65]. [66] demonstrates a skew reduction scheme capable of reducing skew to within 10% of the clock period, versus under 4% here for the same feature size.…”
Section: Simulation Results For Single Clock Systemmentioning
confidence: 98%
“…Comparable solutions offer similar or worse levels of skew reduction, sub-10 ps in a 90-nm technology [64], 70 ps in a 0.18-m technology [2] and 15 ps in a 0.25-m technology [65]. [66] demonstrates a skew reduction scheme capable of reducing skew to within 10% of the clock period, versus under 4% here for the same feature size.…”
Section: Simulation Results For Single Clock Systemmentioning
confidence: 98%
“…The total clock skew ranges from 3.9 ps to 5.5 ps with an overall power consumption of 62.82 mW, or 4.188 mW per tap. Comparable solutions offer similar or worse levels of skew reduction: sub-10 ps for [16], 70 ps for [5], 28 ps for [17] and 15 ps for [18]; [9] is capable of reducing skew to within 10% of the clock period, versus under 4% here; [1] achieves 3 ps skew resolution by modifying an H-tree and requiring a duplicate co-located return path for all leaves. PLL-based distributions typically consume hundreds of mW [17].…”
Section: Simulation Resultsmentioning
confidence: 99%
“…In post-silicon, the aim is to derive the best-known recipe based on critical-path delays that are measured on actual silicon. Prior works have used post-silicon tunable clock-tree synthesis to fix timing violation in microprocessor clock design [4], [14]. Nagaraj et al [14] describe a method that uses transition-ATPG patterns to measure delays on an actual tester on any given chip, and then use clock tuning to optimize the frequency of that particular piece of silicon before shipping it to the customer.…”
Section: Introductionmentioning
confidence: 99%