D-RaNGe: Using Commodity DRAM Devices to Generate True Random Numbers with Low Latency and High Throughput

Abstract

We propose a new DRAM-based true random number generator (TRNG) that leverages DRAM cells as an entropy source.The key idea is to intentionally violate the DRAM access timing parameters and use the resulting errors as the source of randomness. Our technique speci cally decreases the DRAM row activation latency (timing parameter t RCD ) below manufacturerrecommended speci cations, to induce read errors, or activation failures, that exhibit true random behavior. We then aggregate the resulting data from multiple cells to obtain a TRNG capable of providing a high throughput of random numbers at low latency.To demonstrate that our TRNG design is viable using commodity DRAM chips, we rigorously characterize the behavior of activation failures in 282 state-of-the-art LPDDR4 devices from three major DRAM manufacturers. We verify our observations using four additional DDR3 DRAM devices from the same manufacturers. Our results show that many cells in each device produce random data that remains robust over both time and temperature variation. We use our observations to develop D-RaNGe, a methodology for extracting true random numbers from commodity DRAM devices with high throughput and low latency by deliberately violating the read access timing parameters. We evaluate the quality of our TRNG using the commonly-used NIST statistical test suite for randomness and nd that D-RaNGe: 1) successfully passes each test, and 2) generates true random numbers with over two orders of magnitude higher throughput than the previous highest-throughput DRAM-based TRNG.

SIMDRAM: a framework for bit-serial SIMD processing using DRAM

“…Using a typical subarray size of 1024 rows [20,67,69,74,85], SIMDRAM splits the row addressing into 1006 D-group rows, 2 C-group rows, and 16 B-group rows.…”

E-BOOT: Preventing Boot-Time Entropy Starvation in Cloud Systems

“…The main three approaches to obtain randomness from DRAM are: 1) Violating standard timing parameters. When the time to refresh memory is reduced below a threshold recommended by the manufacturer, some errors are produced and the value of some random memory cells is changed [26]. 2) Retaining cell charge.…”

“…When a device is turned on, memory cell contents are unpredictable due to interaction between different logic parts, such as precharge, row decoder and column select [76], [77]. The approach that provides good entropy with the lowest latency and highest throughput is ''Violating standard timing parameters'' [26]. It offers more than 5 MiB/s of true random data.…”

Energy-Secure System Architectures (ESSA): A Workshop Report

“…Another item covered was quick destruction of in-memory data (for DRAMs). The security-centric aspects of the presentation focused mainly on the speaker's work published in HPCA 2018, 15 HPCA 2019 16 and arxiv 2019. 17…”