Simple DRAM and Virtual Memory Abstractions to Enable Highly Efficient Memory Systems

TL;DR

This paper introduces new DRAM and virtual memory techniques, including page overlays, in-DRAM bulk operations, and efficient gather-scatter mechanisms, to significantly improve memory system efficiency and performance.

Contribution

It presents novel memory management and DRAM techniques that enhance efficiency for bulk data operations and multi-granularity memory management.

Findings

Order-of-magnitude improvement in bulk data copy and initialization.

Near-ideal bandwidth for gather-scatter operations with GS-DRAM.

Enhanced dirty block tracking for better coherence and scheduling.

Abstract

In most modern systems, the memory subsystem is managed and accessed at multiple different granularities at various resources. We observe that such multi-granularity management results in significant inefficiency in the memory subsystem. Specifically, we observe that 1) page-granularity virtual memory unnecessarily triggers large memory operations, and 2) existing cache-line granularity memory interface is inefficient for performing bulk data operations and operations that exhibit poor spatial locality. To address these problems, we present a series of techniques in this thesis. First, we propose page overlays, a framework augments the existing virtual memory framework with the ability to track a new version of a subset of cache lines within each virtual page. We show that this extension is powerful by demonstrating its benefits on a number of applications. Second, we show that DRAM can be used to perform more complex operations than just store data. We propose RowClone, a mechanism to perform bulk data copy and initialization completely inside DRAM, and Buddy RAM, a mechanism to perform bulk bitwise operations using DRAM. Both these techniques achieve an order-of-magnitude improvement in the efficiency of the respective operations. Third, we propose Gather-Scatter DRAM, a technique that exploits DRAM organization to effectively gather/scatter values with a power-of-2 strided access patterns. For these access patterns, GS-DRAM achieves near-ideal bandwidth and cache utilization, without increasing the latency of fetching data from memory. Finally, we propose the Dirty-Block Index, a new way of tracking dirty blocks. In addition to improving the efficiency of bulk data coherence, DBI has several applications including high-performance memory scheduling, efficient cache lookup bypassing, and enabling heterogeneous ECC.