Related papers: CLR-DRAM: A Low-Cost DRAM Architecture Enabling Dy…

Even with generational improvements in DRAM technology, memory access latency still remains the major bottleneck for application accelerators, primarily due to limitations in memory interface IPs which cannot fully account for variations in…

Current day processors employ multi-level cache hierarchy with one or two levels of private caches and a shared last-level cache (LLC). An efficient cache replacement policy at LLC is essential for reducing the off-chip memory transfer as…

There is an extremely high demand for a high speed, low power, low leakage, and low noise Static Random-Access Memory (SRAM) for high performance cache memories. The energy efficiency of SRAM is of paramount importance in both high…

DRAM Main memory is a performance bottleneck for many applications due to the high access latency. In-DRAM caches work to mitigate this latency by augmenting regular-latency DRAM with small-but-fast regions of DRAM that serve as a cache for…

The memory controller is in charge of managing DRAM maintenance operations (e.g., refresh, RowHammer protection, memory scrubbing) to reliably operate modern DRAM chips. Implementing new maintenance operations often necessitates…

Computing-In-Memory (CIM) offers a potential solution to the memory wall issue and can achieve high energy efficiency by minimizing data movement, making it a promising architecture for edge AI devices. Lightweight models like MobileNet and…

Memory controller scheduling is crucial in multicore processors, where DRAM bandwidth is shared. Since increased number of requests from multiple cores of processors becomes a source of bottleneck, scheduling the requests efficiently is…

As conventional technology scaling approaches physical and power limitations, modern computing systems increasingly face performance bottlenecks arising from memory latency, energy consumption, scalability constraints, and data movement…

The performance and efficiency of distributed machine learning (ML) depends significantly on how long it takes for nodes to exchange state changes. Overly-aggressive attempts to reduce communication often sacrifice final model accuracy and…

Storage-class memory (SCM) combines the benefits of a solid-state memory, such as high-performance and robustness, with the archival capabilities and low cost of conventional hard-disk magnetic storage. Among candidate solid-state…

Efficient parallelism is necessary for achieving low-latency, high-throughput inference with large language models (LLMs). Tensor parallelism (TP) is the state-of-the-art method for reducing LLM response latency, however GPU communications…

On-device learning allows AI models to adapt to user data, thereby enhancing service quality on edge platforms. However, training AI on resource-limited devices poses significant challenges due to the demanding computing workload and the…

The ever-increasing computation complexity of fastgrowing Deep Neural Networks (DNNs) has requested new computing paradigms to overcome the memory wall in conventional Von Neumann computing architectures. The emerging Computing-In-Memory…

In recent years, researchers have explored use of non-volatile devices such as STT-RAM (spin torque transfer RAM) for designing on-chip caches, since they provide high density and consume low leakage power. A common limitation of all…

Memory caches are being aggressively used in today's data-parallel frameworks such as Spark, Tez and Storm. By caching input and intermediate data in memory, compute tasks can witness speedup by orders of magnitude. To maximize the chance…

Gain Cell memory (GCRAM) offers higher density and lower power than SRAM, making it a promising candidate for on-chip memory in domain-specific accelerators. To support workloads with varying traffic and lifetime metrics, GCRAM also offers…

Next-generation networks aim to provide performance guarantees to real-time interactive services that require timely and cost-efficient packet delivery. In this context, the goal is to reliably deliver packets with strict deadlines imposed…

We propose a class of interleavers for a novel deep neural network (DNN) architecture that uses algorithmically pre-determined, structured sparsity to significantly lower memory and computational requirements, and speed up training. The…

As conventional memory technologies are challenged by their technological physical limits, emerging technologies driven by novel materials are becoming an attractive option for future memory architectures. Among these technologies,…

SRAM-based compute-in-memory (CIM) offers high computational density and energy efficiency for deep neural network (DNN) accelerators, but its limited capacity causes on/off-chip data movement overhead for large DNN models. Existing CIM…