Ultra-Efficient Processing In-Memory for Data Intensive Applications
Mohsen Imani,Saransh Gupta,Tajana Rosing +2 more
- pp 6
TLDR
This paper proposes an ultra-efficient approximate processing in-memory architecture, called APIM, which exploits the analog characteristics of non-volatile memories to support addition and multiplication inside the crossbar memory, while storing the data.Abstract:
Recent years have witnessed a rapid growth in the domain of Internet of Things (IoT). This network of billions of devices generates and exchanges huge amount of data. The limited cache capacity and memory bandwidth make transferring and processing such data on traditional CPUs and GPUs highly inefficient, both in terms of energy consumption and delay. However, many IoT applications are statistical at heart and can accept a part of inaccuracy in their computation. This enables the designers to reduce complexity of processing by approximating the results for a desired accuracy. In this paper, we propose an ultra-efficient approximate processing in-memory architecture, called APIM, which exploits the analog characteristics of non-volatile memories to support addition and multiplication inside the crossbar memory, while storing the data. The proposed design eliminates the overhead involved in transferring data to processor by virtually bringing the processor inside memory. APIM dynamically configures the precision of computation for each application in order to tune the level of accuracy during runtime. Our experimental evaluation running six general OpenCL applications shows that the proposed design achieves up to 20× performance improvement and provides 480× improvement in energy-delay product, ensuring acceptable quality of service. In exact mode, it achieves 28× energy savings and 4.8× speed up compared to the state-of-the-art GPU cores.read more
Citations
More filters
Proceedings ArticleDOI
FloatPIM: in-memory acceleration of deep neural network training with high precision
TL;DR: FloatPIM is proposed, a fully-digital scalable PIM architecture that accelerates CNN in both training and testing phases and natively supports floating-point representation, thus enabling accurate CNN training.
Proceedings ArticleDOI
FELIX: fast and energy-efficient logic in memory
TL;DR: This paper proposes an in-memory implementation of fast and energy-efficient logic (FELIX) which combines the functionality of PIM with memories and is the first PIM logic to enable the single cycle NOR, NOT, NAND, minority, and OR directly in crossbar memory.
Journal ArticleDOI
SIMPLER MAGIC: Synthesis and Mapping of In-Memory Logic Executed in a Single Row to Improve Throughput
Rotem Ben-Hur,Ronny Ronen,Ameer Haj-Ali,Debjyoti Bhattacharjee,Adi Eliahu,Natan Peled,Shahar Kvatinsky +6 more
TL;DR: A novel automatic framework for efficient implementation of arbitrary combinational logic functions within a memristive memory using synthesis and in-memory mapping of logic execution in a single row (SIMPLER), a tool that optimizes the execution of in- memory logic operations in terms of throughput and area.
Proceedings ArticleDOI
Efficient Algorithms for In-Memory Fixed Point Multiplication Using MAGIC
TL;DR: The algorithms proposed in this paper not only improve the latency as compared to previously proposed algorithms by 1.8× on average, but their significantly better area efficiency now makes it possible to perform numerous fixed point multiplications simultaneously within memristive memory arrays.
Journal ArticleDOI
SearcHD: A Memory-Centric Hyperdimensional Computing With Stochastic Training
Mohsen Imani,Xunzhao Yin,John Messerly,Saransh Gupta,Michael Niemier,Xiaobo Sharon Hu,Tajana Rosing +6 more
TL;DR: SearcHD is proposed, a fully binarized HD computing algorithm with a fully binary training which generates multiple binary hypervectors for each class and uses the analog characteristic of nonvolatile memories to perform all encoding, training, and inference computations in memory.
References
More filters
Journal ArticleDOI
Internet of Things (IoT): A vision, architectural elements, and future directions
TL;DR: In this article, the authors present a cloud centric vision for worldwide implementation of Internet of Things (IoT) and present a Cloud implementation using Aneka, which is based on interaction of private and public Clouds, and conclude their IoT vision by expanding on the need for convergence of WSN, the Internet and distributed computing directed at technological research community.
Journal ArticleDOI
‘Memristive’ switches enable ‘stateful’ logic operations via material implication
Julien Borghetti,Gregory S. Snider,Philip J. Kuekes,Jianhua Yang,Duncan Stewart,Duncan Stewart,R. Stanley Williams +6 more
TL;DR: Bipolar voltage-actuated switches, a family of nonlinear dynamical memory devices, can execute material implication (IMP), which is a fundamental Boolean logic operation on two variables p and q such that pIMPq is equivalent to (NOTp)ORq.
Proceedings ArticleDOI
Approximate computing: An emerging paradigm for energy-efficient design
Jie Han,Michael Orshansky +1 more
TL;DR: This paper reviews recent progress in the area, including design of approximate arithmetic blocks, pertinent error and quality measures, and algorithm-level techniques for approximate computing.
Journal ArticleDOI
Low-Power Digital Signal Processing Using Approximate Adders
TL;DR: This paper proposes logic complexity reduction at the transistor level as an alternative approach to take advantage of the relaxation of numerical accuracy, and demonstrates the utility of these approximate adders in two digital signal processing architectures with specific quality constraints.
Journal ArticleDOI
MAGIC—Memristor-Aided Logic
Shahar Kvatinsky,Dmitry Belousov,Slavik Liman,Guy Satat,Nimrod Wald,Eby G. Friedman,Avinoam Kolodny,Uri Weiser +7 more
TL;DR: In this brief, a memristor-only logic family, i.e., memristar-aided logic (MAGIC), is presented, and in each MAGIC logic gate, memristors serve as an input with previously stored data, and an additional Memristor serves as an output.