English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

This paper makes a case for accelerating lattice-based post quantumcryptography (PQC) with memristor based crossbars, and shows that theseinherently error-tolerant algorithms are a good fit for noisy analog MACoperations in crossbars. We compare different NIST round-3 lattice-basedcandidates for PQC, and identify that SABER is not only a front-runner whenexecuting on traditional systems, but it is also amenable to acceleration withcrossbars. SABER is a module-LWR based approach, which performs modularpolynomial multiplications with rounding. We map the polynomial multiplicationsin SABER on crossbars and show that analog dot-products can yield a$1.7-32.5\times$ performance and energy efficiency improvement, compared torecent hardware proposals. This initial design combines the innovations inmultiple state-of-the-art works -- the algorithm in SABER and the memristiveacceleration principles proposed in ISAAC (for deep neural networkacceleration). We then identify the bottlenecks in this initial design andintroduce several additional techniques to improve its efficiency. Thesetechniques are synergistic and especially benefit from SABER's power-of-twomodulo operation. First, we show that some of the software techniques used inSABER, that are effective on CPU platforms, are unhelpful in crossbar-basedaccelerators. Relying on simpler algorithms further improves our efficienciesby $1.3-3.6\times$. Second, we exploit the nature of SABER's computations tostagger the operations in crossbars and share a few variable precision ADCs,resulting in up to $1.8\times$ higher efficiency. Third, to further reduce ADCpressure, we propose a simple analog Shift-and-Add technique, which results ina $1.3-6.3\times$ increase in the efficiency. Overall, our designs achieve$3-15\times$ higher efficiency over initial design, and $3-51\times$ higherthan prior work.

XCRYPT: Accelerating Lattice Based Cryptography with Memristor Crossbar Arrays