In-DRAM Bulk Bitwise Execution Engine

Abstract

Many applications heavily use bitwise operations on large bitvectors as part of their computation. In existing systems, performing such bulk bitwise operations requires the processor to transfer a large amount of data on the memory channel, thereby consuming high latency, memory bandwidth, and energy. In this paper, we describe Ambit, a recently-proposed mechanism to perform bulk bitwise operations completely inside main memory. Ambit exploits the internal organization and analog operation of DRAM-based memory to achieve low cost, high performance, and low energy. Ambit exposes a new bulk bitwise execution model to the host processor. Evaluations show that Ambit significantly improves the performance of several applications that use bulk bitwise operations, including databases.

arXiv commentarXiv admin note: substantial text overlap with arXiv:1605.06483, arXiv:1610.09603, arXiv:1611.09988

References in corpus (3)

• Nanopore Sequencing Technology and Tools for Genome Assembly: Computational Analysis of the Current State, Bottlenecks and Future Directions
• Shouji: A Fast and Efficient Pre-Alignment Filter for Sequence Alignment
• Simple DRAM and Virtual Memory Abstractions to Enable Highly Efficient Memory Systems

Cited by in corpus (14)

• BlockHammer: Preventing RowHammer at Low Cost by Blacklisting Rapidly-Accessed DRAM Rows
• A Deeper Look into RowHammer`s Sensitivities: Experimental Analysis of Real DRAM Chips and Implications on Future Attacks and Defenses
• pLUTo: Enabling Massively Parallel Computation in DRAM via Lookup Tables
• Benchmarking a New Paradigm: An Experimental Analysis of a Real Processing-in-Memory Architecture
• CLR-DRAM: A Low-Cost DRAM Architecture Enabling Dynamic Capacity-Latency Trade-Off
• Understanding the Interactions of Workloads and DRAM Types: A Comprehensive Experimental Study
• FIGARO: Improving System Performance via Fine-Grained In-DRAM Data Relocation and Caching
• A Workload and Programming Ease Driven Perspective of Processing-in-Memory
• SMART: Investigating the Impact of Threshold Voltage Suppression in an In-SRAM Multiplication/Accumulation Accelerator for Accuracy Improvement in 65 nm CMOS Technology
• SIMDRAM: An End-to-End Framework for Bit-Serial SIMD Computing in DRAM
• Triangle Counting Accelerations: From Algorithm to In-Memory Computing Architecture
• EDEN: Enabling Energy-Efficient, High-Performance Deep Neural Network Inference Using Approximate DRAM
• Data-Centric and Data-Aware Frameworks for Fundamentally Efficient Data Handling in Modern Computing Systems
• TCIM: Triangle Counting Acceleration With Processing-In-MRAM Architecture