Ioannis A. Papistas

Ioannis A. Papistas

Auto-generated profile, based on publicly available information

Ioannis A. Papistas

Advanced Processor Technologies Group, School of Computer Science, University of Manchester, Manchester, U.K.

Research Focus & Metrics

Academic Career Length

12Years (2015-Present)

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

Representative Publications

View All Publications
31 October 2022 in IEEE Journal of Solid-State Circuits
Verified
Journal Article
Closed access
PHPouya Houshmand
Pouya Houshmand
GSGiuseppe M. Sarda
Giuseppe M. Sarda
VJVikram Jain
Vikram Jain
KUKodai Ueyoshi
Kodai Ueyoshi
IPIoannis A. Papistas
Ioannis A. Papistas
MVMarian Verhelst
Marian Verhelst
DIgital-ANAlog (DIANA), a heterogeneous multi-core accelerator, combines a reduced instruction set computer -five (RISC-V) host processor with an analog in-memory computing (AIMC) artificial intelligence (AI) accelerator and a digital reconfigurable deep neural network (DNN) accelerator in a single system-on-chip (SoC) to support a wide variety of neural network (NN) workloads. AIMC cores can bring extreme computational parallelism and efficiency at the expense of accuracy and dataflow flexibility. Digital AI co-processors, on the other hand, guarantee accuracy through deterministic compute, but cannot achieve the same computational density and efficiency. DIANA exploits this fundamental tradeoff by integrating both types of cores in a shared and optimized memory system, to enable seamless execution of the workloads on the parallel cores. The system’s performance benefits further from pipelined parallel execution across both accelerator cores and enhanced AIMC spatial unrolling techniques, leading to drastically reduced execution latency and reduced memory footprints. The design has been implemented in a 22-nm technology and achieves peak efficiencies of 600 TOP/s/W for the AIMC core (I/W/O: 7/1.5/6 bit) and 14 TOP/s/W (I/W/O: 8/8/8 bit) for the digital accelerator, respectively. End-to-end performance evaluation of CIFAR-10 and ImageNet classification workloads is carried out on the chip, reporting 7.02 and 5.56 TOP/s/W, respectively, at the system level.

5 October 2020 in IEEE Transactions on Electron Devices
Verified
Journal Article
Closed access
DSD. Saito
D. Saito
JDJ. Doevenspeck
J. Doevenspeck
SCS. Cosemans
S. Cosemans
HOH. Oh
H. Oh
MPM. Perumkunnil
M. Perumkunnil
MNM. H. Na
M. H. Na
We propose, for the first time, an indium gallium zinc oxide (IGZO)-based 2T1C compute cell (IGZO-cell) for analog in-memory computing. To assess the impact of an IGZO-cell-based array including the periphery on power and accuracy, a PyTorch framework was developed to analytically modeled analog components. The results are reported for a ResNet20 network on the Canadian Institute For Advanced Research-10 (CIFAR-10) benchmark. The state-of-the-art energy efficiency of 15 peta operations per second (POPS)/W including the periphery is achieved by using our proposed IGZO-cell with CMOS compatibility. Finally, it is shown that, with a properly trained neural network model, there is no degradation of test accuracy with 10% device to device variability for the IGZO devices.

1 June 2018 in Integration
Verified
Journal Article
Closed access
IPIoannis A. Papistas
Ioannis A. Papistas
VPVasilis F. Pavlidis
Vasilis F. Pavlidis

9 April 2019 in IEEE Transactions on Circuits and Systems II: Express Briefs
Verified
Journal Article
Closed access
IPIoannis A. Papistas
Ioannis A. Papistas
Vasilis F. PavlidisVasilis F. Pavlidis
Vasilis F. Pavlidis
DVDimitrios Velenis
Dimitrios Velenis
The manufacturing cost of contactless 3-D circuits with inductive links is explored for the first time. The high-volume manufacturing (HVM) cost of contactless 3-D ICs is modeled by considering the on-chip inductor area and other important parameters, such as the coupling efficiency, which depend upon specific steps of the fabrication process. Inductive links based on current mode logic (CML) transceivers are simulated across a number of integration scenarios in a commercial 65-nm CMOS technology process, thereby evaluating the cost of these manufacturing options for performance and power objectives. Assuming a 10 Gb/s data rate, the power and area of the inductive link are determined for wafer-to-wafer and die-to-wafer integration and communication distances in the range of 5 $\mu \text{m}$ (face-to-face) to 120 $\mu \text{m}$ (face-to-back). An HVM cost estimation for each investigated inductive link scheme is provided for these integration approaches. Additional processing steps attributed to 3-D integration increase the aggregate cost by 2.5%–4.5% for a case study of a three-tier stack, compared to a conventional 2-D process. Moreover, the yield is explored for each integration approach considering the inductor area for contactless 3-D ICs.

1 August 2015
Verified
Conference Paper
Closed access
IPIoannis A. Papistas
Ioannis A. Papistas
VPVasilis F. Pavlidis
Vasilis F. Pavlidis
Three-dimensional (3-D) integration is a promising technology that can mitigate the deleterious effects of the increased interconnect length in modern ICs by vertically stacking dies. Through silicon vias (TSVs) and AC coupling have been proposed as communication interfaces between multiple stacked dies. This paper investigates these communication schemes for 3-D systems where the link bandwidth is treated as a constraint rather than an objective. A frequency dependent RC model for the TSV and the redistribution layer (RDL) is employed. A first order delay analysis between the two schemes shows comparable performance, but a better area efficiency for the TSV. Considering, however, a multiplexing scheme shows that TSVs with a pitch lower than 20 μm exhibit better bandwidth-to-area efficiency without multiplexing, while the inductive link demonstrates higher efficiency in comparison to TSVs with a pitch larger than 20 μm for a 12:1 multiplexing ratio.

1 May 2017
Verified
Conference Paper
Closed access
IPIoannis A. Papistas
Ioannis A. Papistas
VPVasilis F. Pavlidis
Vasilis F. Pavlidis
A heterogeneous contactless transceiver circuit is designed to provide half duplex communication for a 3-D system considering specific bonding constraints. The system is composed of two tiers and is integrated face-to-back to support fluidic sensing. Communication between the tiers is achieved through inductive links. Each tier is considered to be fabricated in a different technology node to enable low manufacturing cost and benefit from the advantages each technology offers. Both the uplink and downlink transceivers achieve data rates that reach 1 Gbps with non-return-to-zero data encoding. Energy efficiency is the primary objective, with the uplink dissipating 4.93 mW and the downlink 10.53 mW. A 5.2 × power reduction is achieved when using heterogeneous technologies, compared to a state-of-the-art 0.35 βm transceiver, while the dissipated energy is decreased by 34% as compared to a state-of-the-art 65 nm transceiver.