Related papers: ARMS: Adaptive and Robust Memory Tiering System
The escalating scale of Large Language Models (LLMs) necessitates efficient adaptation techniques. Model merging has gained prominence for its efficiency and controllability. However, existing merging techniques typically serve as post-hoc…
In this article, we present a novel approach for block-structured adaptive mesh refinement (AMR) that is suitable for extreme-scale parallelism. All data structures are designed such that the size of the meta data in each distributed…
Many existing data mining algorithms use feature values directly in their model, making them sensitive to units/scales used to measure/represent data. Pre-processing of data based on rank transformation has been suggested as a potential…
This work investigates the problem of instance-level image retrieval re-ranking with the constraint of memory efficiency, ultimately aiming to limit memory usage to 1KB per image. Departing from the prevalent focus on performance…
Compute-in-memory (CiM) is a promising approach to improving the computing speed and energy efficiency in dataintensive applications. Beyond existing CiM techniques of bitwise logic-in-memory operations and dot product operations, this…
MEMS storage devices are new non-volatile secondary storages that have outstanding advantages over magnetic disks. MEMS storage devices, however, are much different from magnetic disks in the structure and access characteristics. They have…
Reverse time migration (RTM) is an algorithm widely used in the oil and gas industry to process seismic data. It is a computationally intensive task that suits well in parallel computers. Methods such as RTM can be parallelized in shared…
Selecting appropriate values for the configurable parameters of Database Management Systems (DBMS) to improve performance is a significant challenge. Recent machine learning (ML)-based tuning systems have shown strong potential, but their…
Aligning general-purpose large language models (LLMs) to downstream tasks often incurs significant training adjustment costs. Prior research has explored various avenues to enhance alignment efficiency, primarily through minimal-data…
Tiered memory, built upon a combination of fast memory and slow memory, provides a cost-effective solution to meet ever-increasing requirements from emerging applications for large memory capacity. Reducing the size of fast memory is…
Atomic multicast is a communication primitive used in dependable systems to ensure consistent ordering of messages delivered to a set of replica groups. This primitive enables critical services to integrate replication and sharding (i.e.,…
Model merging has emerged as a cost-efficient approximation to multitask learning. Among merging strategies, task arithmetic is notable for its simplicity and effectiveness. In this work, we provide a theoretical motivation for task vectors…
Inference serving is of great importance in deploying machine learning models in real-world applications, ensuring efficient processing and quick responses to inference requests. However, managing resources in these systems poses…
As deep learning models continue to increase in size, the memory requirements for training have surged. While high-level techniques like offloading, recomputation, and compression can alleviate memory pressure, they also introduce…
The memory capacity in edge devices is often limited due to constraints on cost, size, and power. Consequently, memory competition leads to inevitable page swapping in memory-constrained mixed-criticality edge devices, causing slow storage…
Current Adaptive Mesh Refinement (AMR) simulations require algorithms that are highly parallelized and manage memory efficiently. As compute engines grow larger, AMR simulations will require algorithms that achieve new levels of efficient…
Processing-in-Memory (PIM) architectures offer promising solutions for efficiently handling AI applications in energy-constrained edge environments. While traditional PIM designs enhance performance and energy efficiency by reducing data…
Thermal-Aware Scheduling (TAS) provides methods to manage the thermal dissipation of a computing chip during task execution. These methods aim to avoid issues such as accelerated aging of the device, premature failure and degraded chip…
Deploying large language model (LLM) inference at scale requires jointly selecting base models, provisioning heterogeneous GPUs, configuring parallelism, and distributing workloads under tight latency, accuracy, and budget constraints.…
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, scalability and energy bottlenecks: (1) data access from memory…