相关论文: Re-enabling high-speed caching for LSM-trees

Large Language Models (LLMs) have achieved unprecedented success across various applications, but their substantial memory requirements pose significant challenges to current memory system designs, especially during inference. Our work…

The ever-increasing demand to extract temporal correlations across sequential data and perform context-based learning in this era of big data has led to the development of long short-term memory (LSTM) networks. Furthermore, there is an…

Large Language Models (LLMs) are increasingly deployed in large-scale online services, enabling sophisticated applications. However, the computational overhead of generating key-value (KV) caches in the prefill stage presents a major…

One key requirement for storage clouds is to be able to retrieve data quickly. Recent system measurements have shown that the data retrieving delay in storage clouds is highly variable, which may result in a long latency tail. One crucial…

Large Language Models (LLMs) have become extremely potent instruments with exceptional capacities for comprehending and producing human-like text in a wide range of applications. However, the increasing size and complexity of LLMs present…

To alleviate the memory bandwidth bottleneck in Large Language Model (LLM) inference workloads, weight matrices are stored in memory in quantized and sparsified formats. Hence, before tiles of these matrices can be processed by in-core…

Caching has the potential to be of significant benefit for accessing large language models (LLMs) due to their high latencies which typically range from a small number of seconds to well over a minute. Furthermore, many LLMs charge money…

Locality-sensitive hashing (LSH) based frameworks have been used efficiently to select weight vectors in a dense hidden layer with high cosine similarity to an input, enabling dynamic pruning. While this type of scheme has been shown to…

There emerges a promising trend of using large language models (LLMs) to generate code-like plans for complex inference tasks such as visual reasoning. This paradigm, known as LLM-based planning, provides flexibility in problem solving and…

The next-generation non-volatile memory (NVM) is striding into computer systems as a new tier as it incorporates both DRAM's byte-addressability and disk's persistency. Researchers and practitioners have considered building persistent…

The increase in data storage and power consumption at data-centers has made it imperative to design energy efficient Distributed Storage Systems (DSS). The energy efficiency of DSS is strongly influenced not only by the volume of data,…

As LLMs and foundation models scale, checkpoint/restore has become a critical pattern for training and inference. With 3D parallelism (tensor, pipeline, data), checkpointing involves many processes, each managing numerous tensors of varying…

Over recent years, an increasing amount of compute and data has been poured into training large language models (LLMs), usually by doing one-pass learning on as many tokens as possible randomly selected from large-scale web corpora. While…

KV cache has traditionally been stored in GPU memory to accelerate the decoding phase of large language model (LLM) inference. However, it is increasingly necessary to move KV caches outside GPU devices, to enable cache reuse across…

Language models (LMs) underpin emerging mobile and embedded AI applications like meeting and video summarization and document analysis, which often require processing multiple long-context inputs. Running an LM locally on-device improves…

Distributed tracing serves as a fundamental element in the monitoring of cloud-based and datacenter systems. It provides visibility into the full lifecycle of a request or operation across multiple services, which is essential for…

The Big Data trend is putting strain on modern storage systems, which have to support high-performance I/O accesses for the large quantities of data. With the prevalent Von Neumann computing architecture, this data is constantly moved back…

We propose BS-tree, an in-memory implementation of the B+-tree that adopts the structure of the disk-based index (i.e., a balanced, multiway tree), setting the node size to a memory block that can be processed fast and in parallel using…

With emerging storage-class memory (SCM) nearing commercialization, there is evidence that it will deliver the much-anticipated high density and access latencies within only a few factors of DRAM. Nevertheless, the latency-sensitive nature…

Large language models (LLMs) excel at capturing semantic nuances and therefore show impressive relevance ranking performance in modern recommendation and search systems. However, they suffer from high computational overhead under industrial…