Reducing DRAM Refresh Overheads with Refresh-Access Parallelism

05/02/2018
by   K. K. Chang, et al.
0

This article summarizes the idea of "refresh-access parallelism," which was published in HPCA 2014, and examines the work's significance and future potential. The overarching objective of our HPCA 2014 paper is to reduce the significant negative performance impact of DRAM refresh with intelligent memory controller mechanisms. To mitigate the negative performance impact of DRAM refresh, our HPCA 2014 paper proposes two complementary mechanisms, DARP (Dynamic Access Refresh Parallelization) and SARP (Subarray Access Refresh Parallelization). The goal is to address the drawbacks of state-of-the-art per-bank refresh mechanism by building more efficient techniques to parallelize refreshes and accesses within DRAM. First, instead of issuing per-bank refreshes in a round-robin order, as it is done today, DARP issues per-bank refreshes to idle banks in an out-of-order manner. Furthermore, DARP proactively schedules refreshes during intervals when a batch of writes are draining to DRAM. Second, SARP exploits the existence of mostly-independent subarrays within a bank. With minor modifications to DRAM organization, it allows a bank to serve memory accesses to an idle subarray while another subarray is being refreshed. Our extensive evaluations on a wide variety of workloads and systems show that our mechanisms improve system performance (and energy efficiency) compared to three state-of-the-art refresh policies, and their performance bene ts increase as DRAM density increases.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
12/21/2017

Improving DRAM Performance by Parallelizing Refreshes with Accesses

Modern DRAM cells are periodically refreshed to prevent data loss due to...
research
05/04/2018

Exploiting the DRAM Microarchitecture to Increase Memory-Level Parallelism

This paper summarizes the idea of Subarray-Level Parallelism (SALP) in D...
research
01/23/2022

Cuckoo Trie: Exploiting Memory-Level Parallelism for Efficient DRAM Indexing

We present the Cuckoo Trie, a fast, memory-efficient ordered index struc...
research
05/08/2018

Exploiting Row-Level Temporal Locality in DRAM to Reduce the Memory Access Latency

This paper summarizes the idea of ChargeCache, which was published in HP...
research
04/12/2018

On Using Non-Volatile Memory in Apache Lucene

Apache Lucene is a widely popular information retrieval library used to ...
research
02/15/2021

GradPIM: A Practical Processing-in-DRAM Architecture for Gradient Descent

In this paper, we present GradPIM, a processing-in-memory architecture w...
research
04/19/2023

Egalitarian ORAM: Wear-Leveling for ORAM

While non-volatile memories (NVMs) provide several desirable characteris...

Please sign up or login with your details

Forgot password? Click here to reset