Editorial for the special issue on disruptive computing technologies
- PDF / 411,316 Bytes
- 2 Pages / 595.276 x 790.866 pts Page_size
- 50 Downloads / 285 Views
EDITORIAL
Editorial for the special issue on disruptive computing technologies Yiran Chen1 · Deliang Fan2 · Yanzhi Wang3 · Shigeru Yamashita4 Published online: 8 September 2020 © China Computer Federation (CCF) 2020
The modern computing paradigm is experiencing many revolutions from almost every aspects: the introduction of emerging memories like spintronic memory and resistive memory dramatically as well as quantum computing change the design of memory hierarchy and interface; new applications such as artificial intelligence and deep learning trigger wide adaptation of deep neural network accelerators and neuromorphic computing circuits; new computing models induced by these applications such as in-memory computing also inspires the corresponding circuit- and architecturelevel practices. The goal of this special issue is to present the novel ideas, designs, implementations, and practices of new disruptive computing technologies for resolving the above challenges, by leveraging recent advances in materials, devices, architectures, system designs, and applications. We include seven papers for this special issue (including a regular paper) based on a peer-review procedure, covering various aspects of disruptive computing technologies. The first paper presents a fast hardware simulator architecture for the Walsh–Hadamard transform, which is a core of many quantum algorithms including quantum heuristic algorithms. The authors develop a method to divide the whole computation of the Walsh-Hadamard transform into pieces and process them in a pipelined manner.
* Yiran Chen [email protected] Deliang Fan [email protected] Yanzhi Wang [email protected] Shigeru Yamashita [email protected] 1
Duke University, Durham, USA
2
Arizona State University, Tempe, USA
3
Northeastern University, Boston, USA
4
Ritsumeikan University, Kyoto, Japan
The second paper first presents a quantum multiplier based on Quantum Fourier. Transform (QFT), which is composed of a series of double-controlled phase gates. An optimization is then conducted on the proposed quantum multiplier to reduce the number of qubits in ancillary and reduce the resource cost of quantum multiplier greatly. The third paper proposes an enhanced quantum scheme for generalized novel enhanced quantum image representation. Simulations show that the proposed scheme can greatly improve the security and processing efficiency of image encryption algorithm. The fourth paper develops a flexible precise-time-dependent single-spike neuromorphic computing architecture, namely “FPT-Spike”. “FPT-spike” relies on three hardwarefavorable components: precise ultra-sparse spike temporal encoding, efficient supervised temporal learning and fast asymmetric decoding, to realize flexible spatial–temporal information trade-off for neural network size reduction without scarifying data processing capability. The fifth paper proposes a reconfigurable computing-inmemory architecture for general purpose computing based on STT-MRAM (GCIM). The proposed GCIM could significantly
Data Loading...
