Graphene-based Wireless Agile Interconnects for Massive Heterogeneous Multi-chip Processors

Abadal, Sergi; Guirado, Robert; Taghvaee, Hamidreza; Jain, Akshay; Santana, Elana Pereira de; Bolívar, P. Haring; Elsayed, Mohamed Saeed; Negra, Renato; Wang, Zhenxing; Wang, Kun‐Ta; Lemme, Max C.; Klein, Joshua; Zapater, Marina; Levisse, Alexandre; Atienza, David; Rossi, Davide; Conti, Francesco; Dazzi, Martino; Karunaratne, Geethan; Boybat, Irem; Sebastian, Abu

doi:10.48550/arxiv.2011.04107

Cited by 4 publications

(8 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Firstly, the high mobility of graphene is crucial to enable operation at THz frequencies or even optical frequency [508]. Secondly, graphene can be used to realize antennas with smaller dimensions compared to metals, which is relevant for the high-density integration of such systems [509]. In addition, the Fermi energy in graphene can be controlled by an electric field, which enables the tunability of the antenna oscillation frequency [509].…”

Section: Two-dimensional Materials For Radiofrequency Energy Harvestingmentioning

confidence: 99%

“…Secondly, graphene can be used to realize antennas with smaller dimensions compared to metals, which is relevant for the high-density integration of such systems [509]. In addition, the Fermi energy in graphene can be controlled by an electric field, which enables the tunability of the antenna oscillation frequency [509]. Last but not least, the high mechanical strength and flexibility of two-dimensional materials, combined with a thin-film technology that allows processing them on arbitrary substrates, largely expands the application field of graphene diodes and antennas for wearable electronics compared to canonical semiconductor devices.…”

Section: Two-dimensional Materials For Radiofrequency Energy Harvestingmentioning

confidence: 99%

See 1 more Smart Citation

Roadmap on energy harvesting materials

et al. 2023

View full text Add to dashboard Cite

Ambient energy harvesting has great potential to contribute to sustainable development and address growing environmental challenges. Converting waste energy from energy-intensive processes and systems (e.g., combustion engines and furnaces) is crucial to reducing their environmental impact and achieving net-zero emissions. Compact energy harvesters will also be key to powering the exponentially growing smart devices ecosystem that is part of the Internet of Things, thus enabling futuristic applications that can improve our quality of life (e.g., smart homes, smart cities, smart manufacturing, and smart healthcare). To achieve these goals, innovative materials are needed to efficiently convert ambient energy into electricity through various physical mechanisms, such as the photovoltaic effect, thermoelectricity, piezoelectricity, triboelectricity, and electromagnetic power transfer. By bringing together the perspectives of experts in various types of energy harvesting materials, this Roadmap provides extensive insights into recent advances and present challenges in the field. Additionally, the Roadmap analyzes the key performance metrics of these technologies in relation to their ultimate energy conversion limits. Building on these insights, the Roadmap outlines promising directions for future research to fully harness the potential of energy harvesting materials for green energy anytime, anywhere.

show abstract

Section: Two-dimensional Materials For Radiofrequency Energy Harvestingmentioning

confidence: 99%

Section: Two-dimensional Materials For Radiofrequency Energy Harvestingmentioning

confidence: 99%

Roadmap on energy harvesting materials

et al. 2023

View full text Add to dashboard Cite

show abstract

“…Among the key advantages of WNoCs, one can find a natural support to broadcast communications, reduced latency, and an adaptive network topology [36], [39], [48], [49]. Hence, WNoCs can be especially advantageous if they are used to serve specific communication patterns that are very challenging to tackle using conventional NoCs [46].…”

Section: Wireless Network-on-chipmentioning

confidence: 99%

“…To address the scalability problem of IMC-based HDC architectures, in this paper we propose to use wireless communications technology. Wireless Network-on-Chip (WNoC) have shown promise in alleviating the bottlenecks that traditional NoC and NiP face, especially for collective traffic patterns and large-scale interconnection demands that are common in HDC [35]- [39]. To that end, WNoCs provide native broadcast capabilities.…”

Section: Introductionmentioning

confidence: 99%

Wireless On-Chip Communications for Scalable In-memory Hyperdimensional Computing

Guirado¹,

Rahimi²,

Karunaratne³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Hyperdimensional computing (HDC) is an emerging computing paradigm that represents, manipulates, and communicates data using very long random vectors (aka hypervectors). Among different hardware platforms capable of executing HDC algorithms, in-memory computing (IMC) systems have been recently proved to be one of the most energy-efficient options, due to hypervector manipulations in the memory itself that reduces data movement. Although implementations of HDC on single IMC cores have been made, their parallelization is still unresolved due to the communication challenges that these novel architectures impose and that traditional Networks-on-Chip and Networks-in-Package were not designed for. To cope with this difficulty, we propose the use of wireless on-chip communication technology in unique ways. We are particularly interested in physically distributing a large number of IMC cores performing similarity search across a chip, and maintaining the classification accuracy when each of which is queried with a slightly different version of a bundled hypervector. To achieve it, we introduce a novel over-the-air computing that consists of defining different binary decision regions in the receivers so as to compute the logical majority operation (i.e., bundling, or superposition) required in HDC. It introduces moderate overheads of a single antenna and receiver per IMC core. By doing so, we achieve a joint broadcast distribution and computation with a performance and efficiency unattainable with wired interconnects, which in turn enables massive parallelization of the architecture. It is demonstrated that the proposed approach allows to both bundle at least three hypervectors and scale similarity search to 64 IMC cores seamlessly, while incurring an average bit error ratio of 0.01 without any impact in the accuracy of a generic HDC-based classifier working with 512-bit vectors.

show abstract

“…To address these challenges, for the first time, the opportunities provided by integrated wireless technology [2,5,10,27] for the data distribution in 2.5D DNN accelerators are explored. We show that wireless links can (i) provide higher bandwidth than electrical interposers and (ii) naturally support broadcast.…”

Section: Introductionmentioning

confidence: 99%

Dataflow-Architecture Co-Design for 2.5D DNN Accelerators using Wireless Network-on-Package

Guirado¹,

Kwon²,

Abadal³

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

Deep neural network (DNN) models continue to grow in size and complexity, demanding higher computational power to enable real-time inference. To efficiently deliver such computational demands, hardware accelerators are being developed and deployed across scales. This naturally requires an efficient scale-out mechanism for increasing compute density as required by the application. 2.5D integration over interposer has emerged as a promising solution, but as we show in this work, the limited interposer bandwidth and multiple hops in the Network-on-Package (NoP) can diminish the benefits of the approach. To cope with this challenge, we propose WIENNA, a wireless NoP-based 2.5D DNN accelerator. In WIENNA, the wireless NoP connects an array of DNN accelerator chiplets to the global buffer chiplet, providing highbandwidth multicasting capabilities. Here, we also identify the dataflow style that most efficienty exploits the wireless NoP's high-bandwidth multicasting capability on each layer. With modest area and power overheads, WIENNA achieves 2.2X-5.1X higher throughput and 38.2% lower energy than an interposer-based NoP design.

show abstract

Graphene-based Wireless Agile Interconnects for Massive Heterogeneous Multi-chip Processors

Cited by 4 publications

References 14 publications

Roadmap on energy harvesting materials

Roadmap on energy harvesting materials

Wireless On-Chip Communications for Scalable In-memory Hyperdimensional Computing

Dataflow-Architecture Co-Design for 2.5D DNN Accelerators using Wireless Network-on-Package

Contact Info

Product

Resources

About