Jinjun Xiong
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Jinjun Xiong has a total of 12 co-authors. Here are the top co-authors by number of shared publications:
Lei He
6 shared publicationsXianlong Hong
Dept. of Comput. Sci. & Technol., Tsinghua Univ., Beijing, China
2 shared publications- Yiyu Shi
Missouri University of Science and Technology (formerly University of Missouri, Rolla)
2 shared publications Howard Chen
IBM Research Division, T. J. Watson Research Center, Yorktown Heights, NY, USA
1 shared publicationJ. Ma
Dept. of Electr. & Comput. Eng., Wisconsin Univ., Madison, WI, USA
1 shared publicationJingyu Xu
Dept. of CST, Tsinghua Univ., Beijing, China
1 shared publication
Latest Publications
- Open Access
- 28 February 2011
- Published by IntechOpen
Stochastic Optimization Over Correlated Data Set: A Case Study on VLSI Decoupling Capacitance Budgeting
- 28 February 2011
- Published by IntechOpen
Open AccessStochastic Optimization Over Correlated Data Set: A Case Study on VLSI Decoupling Capacitance Budgeting
- 11 December 2009
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
- Vol. 19 (3) , 508-512
Runtime Resonance Noise Reduction with Current Prediction Enabled Frequency Actuator
Power delivery network (PDN) is a distributed resistance-inductance-capacitance (RLC) network with its dominant resonance frequency in the low-to-middle frequency range. Though high-performance chips' working frequencies are much higher than this resonance frequency in general, chip runtime loading frequency is not. When a chip executes a chunk of instructions repeatedly, the induced current load may have harmonic components close to this resonance frequency, causing excessive power integrity degradation. Existing PDN design solutions are, however, mainly targeted at reducing high-frequency noise and not effective to suppress such resonance noise. In this work, we propose a novel approach to proactively suppress this type of noise. A method based on the high dimension generalized Markov process is developed to predict current load variation. Based on such prediction, a clock frequency actuator design is proposed to proactively select an optimal clock frequency to suppress the resonance. To the best of our knowledge, this is the first in-depth study on proactively reducing instruction loop induced PDN resonance noise at the runtime.
- 11 December 2009
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
- Vol. 19 (3) , 508-512
Runtime Resonance Noise Reduction with Current Prediction Enabled Frequency Actuator
Power delivery network (PDN) is a distributed resistance-inductance-capacitance (RLC) network with its dominant resonance frequency in the low-to-middle frequency range. Though high-performance chips' working frequencies are much higher than this resonance frequency in general, chip runtime loading frequency is not. When a chip executes a chunk of instructions repeatedly, the induced current load may have harmonic components close to this resonance frequency, causing excessive power integrity degradation. Existing PDN design solutions are, however, mainly targeted at reducing high-frequency noise and not effective to suppress such resonance noise. In this work, we propose a novel approach to proactively suppress this type of noise. A method based on the high dimension generalized Markov process is developed to predict current load variation. Based on such prediction, a clock frequency actuator design is proposed to proactively select an optimal clock frequency to suppress the resonance. To the best of our knowledge, this is the first in-depth study on proactively reducing instruction loop induced PDN resonance noise at the runtime.
- 19 March 2007
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
- Vol. 26 (4) , 739-742
Probabilistic Transitive-Closure Ordering and Its Application on Variational Buffer Insertion
We propose a provably transitive-closure ordering rule with theoretical foundations to prune suboptimal design solutions in the presence of process variations. As an example, this probabilistic ordering rule is applied to develop an efficient variational buffering algorithm. Compared to the conventional deterministic approach, variational buffering improves the parametric timing yield by 15.7% on average. This transitive-closure ordering rule may be leveraged to solve other computer-aided-design problems considering process variation effects
- 19 March 2007
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
- Vol. 26 (4) , 739-742
Probabilistic Transitive-Closure Ordering and Its Application on Variational Buffer Insertion
We propose a provably transitive-closure ordering rule with theoretical foundations to prune suboptimal design solutions in the presence of process variations. As an example, this probabilistic ordering rule is applied to develop an efficient variational buffering algorithm. Compared to the conventional deterministic approach, variational buffering improves the parametric timing yield by 15.7% on average. This transitive-closure ordering rule may be leveraged to solve other computer-aided-design problems considering process variation effects
- 30 November 2004
- Published by Institute of Electrical and Electronics Engineers (IEEE)
Performance and RLC crosstalk driven global routing
This work presents a global routing algorithm that minimizes total wire length and satisfies RLC crosstalk constraints specified at sinks. Our algorithm is based on critical network concept and search space traversing technology (SSTT) for global routing synthesis and tabu search for shield insertion and net ordering (SINO) to eliminate noise. The algorithm achieves about 20/spl times/ speedup compared with a recent work using iterative deletion based global routing and simulated annealing based SINO. Furthermore, our algorithm increases the wire length by 4% compared with global routing without crosstalk constraints, achieving a 2.5/spl times/ reduction compared with the aforementioned recent work.
- 30 November 2004
- Published by Institute of Electrical and Electronics Engineers (IEEE)
Performance and RLC crosstalk driven global routing
This work presents a global routing algorithm that minimizes total wire length and satisfies RLC crosstalk constraints specified at sinks. Our algorithm is based on critical network concept and search space traversing technology (SSTT) for global routing synthesis and tabu search for shield insertion and net ordering (SINO) to eliminate noise. The algorithm achieves about 20/spl times/ speedup compared with a recent work using iterative deletion based global routing and simulated annealing based SINO. Furthermore, our algorithm increases the wire length by 4% compared with global routing without crosstalk constraints, achieving a 2.5/spl times/ reduction compared with the aforementioned recent work.
- 30 November 2004
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 5
Shielding area optimization under the solution of interconnect crosstalk
As the technology advances into deep sub-micron era, crosstalk reduction is of paramount importance for signal integrity. Simultaneous shield insertion and net ordering (SINO) has been shown to be effective to reduce both capacitive and inductive coupling. Although shield insertion could reduce crosstalk efficiently, a large scale of unnecessary shields will bring the shielding area problem which is also critical for an efficient SINO algorithm. We propose three novel algorithms using fewer shields to solve SINO problem: namely, net coloring (NC), efficient middle shield insertion (EMSI) and NC+EMSI. Compared to the corresponding algorithms in previous work of L. He and K.M. Lepak, (see inbid.,p55-60, 2000), our algorithms can reduce shielding area largely with short runtime.
- 30 November 2004
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 5
Shielding area optimization under the solution of interconnect crosstalk
As the technology advances into deep sub-micron era, crosstalk reduction is of paramount importance for signal integrity. Simultaneous shield insertion and net ordering (SINO) has been shown to be effective to reduce both capacitive and inductive coupling. Although shield insertion could reduce crosstalk efficiently, a large scale of unnecessary shields will bring the shielding area problem which is also critical for an efficient SINO algorithm. We propose three novel algorithms using fewer shields to solve SINO problem: namely, net coloring (NC), efficient middle shield insertion (EMSI) and NC+EMSI. Compared to the corresponding algorithms in previous work of L. He and K.M. Lepak, (see inbid.,p55-60, 2000), our algorithms can reduce shielding area largely with short runtime.
- 26 June 2003
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 504-509
Post global routing RLC crosstalk budgeting
Existing layout optimization methods often assume a set of interconnects with given RLC crosstalk bounds in a routing region. RLC crosstalk bound partitioning is critical for effectively applying these methods at the full-chip level. In this paper, we develop an optimal crosstalk budgeting scheme based on linear programming (LP) formulation, and apply it to shield insertion and net ordering at the full-chip level. Experiment results show that compared to the best alternative approach, the LP based method reduces the total routing area by up to 7.61% and also uses less runtime. To the best of our knowledge, this is the first in-depth work that studies the RLC crosstalk budgeting problem.
- 26 June 2003
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 504-509
Post global routing RLC crosstalk budgeting
Existing layout optimization methods often assume a set of interconnects with given RLC crosstalk bounds in a routing region. RLC crosstalk bound partitioning is critical for effectively applying these methods at the full-chip level. In this paper, we develop an optimal crosstalk budgeting scheme based on linear programming (LP) formulation, and apply it to shield insertion and net ordering at the full-chip level. Experiment results show that compared to the best alternative approach, the LP based method reduces the total routing area by up to 7.61% and also uses less runtime. To the best of our knowledge, this is the first in-depth work that studies the RLC crosstalk budgeting problem.