Richard Dorrance

Richard Dorrance

Auto-generated profile, based on publicly available information

Richard Dorrance

Intel Labs, Intel Corporation, Hillsboro, OR, USA

Research Focus & Metrics

Academic Career Length

16Years (2011-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
6 January 2023 in IEEE Journal of Solid-State Circuits
Verified
Journal Article
Closed access
HWHechen Wang
Hechen Wang
RLRenzhi Liu
Renzhi Liu
RDRichard Dorrance
Richard Dorrance
DDDeepak Dasalukunte
Deepak Dasalukunte
DLDan Lake
Dan Lake
BCBrent Carlton
Brent Carlton
Compute-in-memory (CiM) is one promising solution to address the memory bottleneck existing in traditional computing architectures. However, the tradeoff between energy efficiency and computing precision plagues most CiM implementations, and the low precision imposes a major limitation on CiM’s ability to support practical computational workloads. In this article, a static random access memory (SRAM)-based analog CiM macro is presented with the Intel 22FFL process. By introducing a 1-to-2 ratioed capacitor ladder (C-2C)-based charge domain computing scheme, the proposed CiM prototype chip demonstrates a maximum of 2k multiply-accumulate (MAC) operations in one clock cycle and achieves 32.2-TOPS/W peak power efficiency with 8-bit precision in both input activation and weight while ensuring accurate on-chip matrix–vector multiplications (MVMs) with a computation error less than 0.5%. A 4.0-TOPS/mm $^{2}$ peak area efficiency is attained by adopting a local weight multiplexing scheme with a 9T SRAM cell, which improves the memory density and reduces the need to refresh the weight stored in the SRAM array. A variety of analog impairment factors, including parasitics, mismatch, and noise, were analyzed to guarantee a sufficiently high multibit linearity. The proposed passive analog computing mechanism ensures the computation accuracy over process–voltage–temperature (PVT), with the measured MVM error deviation of less than 1% over PVT variations.

1 December 2012
Verified
Conference Paper
Closed access
JAJ. G. Alzate
J. G. Alzate
PKPedram Khalili
Pedram Khalili
PUP. Upadhyaya
P. Upadhyaya
SCS. S. Cherepov
S. S. Cherepov
JZJ. Zhu
J. Zhu
KWK. L. Wang
K. L. Wang
We demonstrate voltage-induced (non-STT) switching of nanoscale, high resistance voltage-controlled magnetic tunnel junctions (VMTJs) with pulses down to 10 ns. We show ~10x reduction in switching energies (compared to STT) with leakage currents <; 105 A/cm2. Switching dynamics, from quasi-static to the nanosecond regime, are studied in detail. Finally, a strategy for eliminating the need for external magnetic-fields, where switching is performed by set/reset voltages of different amplitudes but same polarity, is proposed and verified experimentally.

26 February 2014
Verified
Conference Paper
Open access
PDF file
RDRichard Dorrance
Richard Dorrance
FRFengbo Ren
Fengbo Ren
DMDejan Marković
Dejan Marković
Sparse Matrix-Vector Multiplication (SpMxV) is a widely used mathematical operation in many high-performance scientific and engineering applications. In recent years, tuned software libraries for multi-core microprocessors (CPUs) and graphics processing units (GPUs) have become the status quo for computing SpMxV. However, the computational throughput of these libraries for sparse matrices tends to be significantly lower than that of dense matrices, mostly due to the fact that the compression formats required to efficiently store sparse matrices mismatches traditional computing architectures. This paper describes an FPGA-based SpMxV kernel that is scalable to efficiently utilize the available memory bandwidth and computing resources. Benchmarking on a Virtex-5 SX95T FPGA demonstrates an average computational efficiency of 91.85%. The kernel achieves a peak computational efficiency of 99.8%, a >50x improvement over two Intel Core i7 processors (i7-2600 and i7-4770) and showing a >300x improvement over two NVIDA GPUs (GTX 660 and GTX Titan), when running the MKL and cuSPARSE sparse-BLAS libraries, respectively. In addition, the SpMxV FPGA kernel is able to achieve higher performance than its CPU and GPU counterparts, while using only 64 single-precision processing elements, with an overall 38-50x improvement in energy efficiency.

26 January 2012 in IEEE Transactions on Electron Devices
Verified
Journal Article
Closed access
RDRichard Dorrance
Richard Dorrance
FRFengbo Ren
Fengbo Ren
YTYuta Toriyama
Yuta Toriyama
AHAmr Amin Hafez
Amr Amin Hafez
CYChih-Kong Ken Yang
Chih-Kong Ken Yang
DMDejan Markovic
Dejan Markovic
We present a design-space feasibility region, as a function of magnetic tunnel junction (MTJ) characteristics and target memory specifications, to explore the design margin of a one-transistor-one-magnetic-tunnel-junction (1T-1MTJ) memory cell for spin-transfer torque random access memories (STT-RAMs). Data from measured devices are used to model the statistical variation of an MTJ's critical switching current and resistance. The sensitivity of the design space to different design parameters is also analyzed for the scaling of both the MTJ and the underlying transistor technology. A design flow, using a sensitivity-based analysis and an MTJ switching model based on the Landau-Lifshitz-Gilbert equation, is proposed to optimize design margins for gigabit-scale memories. Design points for improved yield, density, and memory performance are extracted from MTJ-compatible complementary metal-oxide-semiconductor (CMOS) technologies for 90-, 65-, 45-, and 32-nm processes. Predictive technology models are used to explore the future scalability of STT-RAMs in upcoming 22- and 16-nm technology nodes. Our analysis shows that, to achieve Flash-like densities ( <; 6F2) in advanced CMOS technologies, aggressive scaling of the critical switching current density will be required.

3 May 2013 in IEEE Electron Device Letters
Verified
Journal Article
Closed access
RDRichard Dorrance
Richard Dorrance
JAJuan G. Alzate
Juan G. Alzate
SCSergiy S. Cherepov
Sergiy S. Cherepov
PUPramey Upadhyaya
Pramey Upadhyaya
IKIlya N. Krivorotov
Ilya N. Krivorotov
DMDejan Marković
Dejan Marković
This letter presents a diode-magnetic tunnel junction (MTJ) magnetic random access memory cell in a 65-nm complimentary metal-oxide-semiconductor compatible process. A voltage-controlled magnetic anisotropy switching mechanism, in addition to STT, allows for a unipolar set/reset write scheme, where voltage pulses of the same polarity, but different amplitudes, are used to switch the MTJs. A small crossbar array is constructed from 65-nm MTJs fabricated on a silicon wafer, with switching voltages ~ 1 V and thermal stability greater than 10 years, with discrete germanium diodes as access devices to allow for read/write operations. The crossbar architecture can be extended to multiple layers to create a 3-D stackable, nonvolatile memory with a sub-1F2 effective cell size.

24 December 2019 in IEEE Journal of Solid-State Circuits
Verified
Journal Article
Closed access
RLRenzhi Liu
Renzhi Liu
ATAsma Beevi K. T.
Asma Beevi K. T.
RDRichard Dorrance
Richard Dorrance
DDDeepak Dasalukunte
Deepak Dasalukunte
VKVinod Kristem
Vinod Kristem
BCBrent R. Carlton
Brent R. Carlton
This article presents an 802.11ba-based wake-up radio (WUR) receiver that is fully integrated within an 802.11a/b/g/n/ac Wi-Fi transceiver. Low-cost integration is achieved with the WUR only occupying 0.13 mm² while not adding any additional overhead to the Wi-Fi platform. The WUR consumes 667 μW from 1.1-V Wi-Fi system supply. When measured from internal 0.6- and 0.9-V supplies, the receiver consumes 495 μW. The receiver has a measured sensitivity of -92.6 dBm and can tolerate -40.5-dBm Wi-Fi adjacent channel blocker with 3-dB receiver de-sensitization. The WUR and Wi-Fi system co-operation is also demonstrated. The WUR receiver can operate when the Wi-Fi system is in sleep mode and can turn on Wi-Fi radio upon receiving 802.11ba-based wake-up packet.