Related papers: Crystalline: Fast and Memory Efficient Wait-Free R…

Persistent AI memory is often reduced to a retrieval problem: store prior interactions as text, embed them, and ask the model to recover relevant context later. This design is useful for thematic recall, but it is mismatched to the kinds of…

LLMs often struggle with memory-constrained deployment on consumer-grade hardware due to their massive parameter sizes. While existing solutions such as model compression and offloading improve deployment feasibility, they often suffer from…

The increasing size of large language models (LLMs) has led to a surge in memory requirements during training, often exceeding the capacity of high-bandwidth memory (HBM). Swap-based memory optimization incurs neither accuracy loss nor…

Modern DRAM modules are often equipped with hardware error correction capabilities, especially for DRAM deployed in large-scale data centers, as process technology scaling has increased the susceptibility of these devices to errors. To…

Emerging Non-Volatile Memories (NVMs) are promising contenders for building future memory systems. On the other side, unlike DRAM systems, NVMs can retain data even after power loss and thus enlarge the attack surface. While data encryption…

Hybrid memory systems, comprised of emerging non-volatile memory (NVM) and DRAM, have been proposed to address the growing memory demand of applications. Emerging NVM technologies, such as phase-change memories (PCM), memristor, and 3D…

Serverless computing is a popular cloud deployment paradigm where developers implement applications as workflows of functions that invoke each other. Cloud providers automatically scale function instances on demand and forward workflow…

Rowhammer is a critical vulnerability in dynamic random access memory (DRAM) that continues to pose a significant threat to various systems. However, we find that conventional load-based attacks are becoming highly ineffective on the most…

Compressed bitmap indexes are used in databases and search engines. Many bitmap compression techniques have been proposed, almost all relying primarily on run-length encoding (RLE). However, on unsorted data, we can get superior performance…

Long-horizon language agents accumulate conversation history far faster than any fixed context window can hold, making memory management critical to both answer accuracy and serving cost. Existing approaches either expand the context window…

To efficiently scale large model (LM) training, researchers transition from data parallelism (DP) to hybrid parallelism (HP) on GPU clusters, which frequently experience hardware and software failures. Existing works introduce in-memory…

New findings in natural language processing (NLP) demonstrate that the strong memorization capability contributes a lot to the success of Large Language Models (LLM). This inspires us to explicitly bring an independent memory mechanism into…

Reusing intermediates in databases to speed-up analytical query processing has been studied in the past. Existing solutions typically require intermediate results of individual operators to be materialized into temporary tables to be…

This dissertation rigorously characterizes many modern commodity DRAM devices and shows that by exploiting DRAM access timing margins within manufacturer-recommended DRAM timing specifications, we can significantly improve system…

Quantization is critical for efficiently deploying large language models (LLMs). Yet conventional methods remain hardware-agnostic, limited to bit-width constraints, and do not account for intrinsic circuit characteristics such as the…

The growing memory footprints of cloud and big data applications mean that data center CPUs can spend significant time waiting for memory. An attractive approach to improving performance in such centralized compute settings is to employ…

Advanced channel reservation is emerging as an important feature of ultra high-speed networks requiring the transfer of large files. Applications include scientific data transfers and database backup. In this paper, we present two new,…

Considering the widespread use of effective capacity in cross-layer design and the extensive existence of renewal service processes in communication networks, this paper thoroughly investigates the effective capacity for renewal processes.…

Fault tolerance overhead of high performance computing (HPC) applications is becoming critical to the efficient utilization of HPC systems at large scale. HPC applications typically tolerate fail-stop failures by checkpointing. Another…

Modern LLM serving now spans multi-stage pipelines including RAG retrieval and KV cache reuse, each with distinct compute, memory, and latency demands. Inference engines expose a large configuration space with no systematic navigation…