An overview of processing-in-memory circuits for artificial intelligence and machine learning
Artificial intelligence (AI) and machine learning (ML) are revolutionizing many fields of study,
such as visual recognition, natural language processing, autonomous vehicles, and …
such as visual recognition, natural language processing, autonomous vehicles, and …
A review of near-memory computing architectures: Opportunities and challenges
The conventional approach of moving stored data to the CPU for computation has become a
major performance bottleneck for emerging scale-out data-intensive applications due to their …
major performance bottleneck for emerging scale-out data-intensive applications due to their …
A modern primer on processing in memory
Modern computing systems are overwhelmingly designed to move data to computation. This
design choice goes directly against at least three key trends in computing that cause …
design choice goes directly against at least three key trends in computing that cause …
Ambit: In-memory accelerator for bulk bitwise operations using commodity DRAM technology
Many important applications trigger bulk bitwise operations, ie, bitwise operations on large
bit vectors. In fact, recent works design techniques that exploit fast bulk bitwise operations to …
bit vectors. In fact, recent works design techniques that exploit fast bulk bitwise operations to …
Benchmarking a new paradigm: Experimental analysis and characterization of a real processing-in-memory system
Many modern workloads, such as neural networks, databases, and graph processing, are
fundamentally memory-bound. For such workloads, the data movement between main …
fundamentally memory-bound. For such workloads, the data movement between main …
Google workloads for consumer devices: Mitigating data movement bottlenecks
We are experiencing an explosive growth in the number of consumer devices, including
smartphones, tablets, web-based computers such as Chromebooks, and wearable devices …
smartphones, tablets, web-based computers such as Chromebooks, and wearable devices …
SIMDRAM: A framework for bit-serial SIMD processing using DRAM
Processing-using-DRAM has been proposed for a limited set of basic operations (ie, logic
operations, addition). However, in order to enable full adoption of processing-using-DRAM …
operations, addition). However, in order to enable full adoption of processing-using-DRAM …
Processing data where it makes sense: Enabling in-memory computation
Today's systems are overwhelmingly designed to move data to computation. This design
choice goes directly against at least three key trends in systems that cause performance …
choice goes directly against at least three key trends in systems that cause performance …
Rowhammer: A retrospective
This retrospective paper describes the RowHammer problem in dynamic random access
memory (DRAM), which was initially introduced by Kim et al. at the ISCA 2014 Conference …
memory (DRAM), which was initially introduced by Kim et al. at the ISCA 2014 Conference …
Recnmp: Accelerating personalized recommendation with near-memory processing
Personalized recommendation systems leverage deep learning models and account for the
majority of data center AI cycles. Their performance is dominated by memory-bound sparse …
majority of data center AI cycles. Their performance is dominated by memory-bound sparse …