Related papers: KRCORE: a microsecond-scale RDMA control plane for…

In this paper, we consider an mmWave-based trainground communication system in the high-speed railway (HSR) scenario, where the computation tasks of users can be partially offloaded to the rail-side base station (BS) or the mobile relays…

We present RDMAbox, a set of low level RDMA optimizations that provide better performance than previous approaches. The optimizations are packaged in easy-to-use kernel and user space libraries for applications and systems in data center.…

Deep Learning Recommendation Models (DLRMs) play a crucial role in delivering personalized content across web applications such as social networking and video streaming. However, with improvements in performance, the parameter size of DLRMs…

Low-level database operators often admit multiple physical implementations ("kernels") that are semantically equivalent but have vastly different performance characteristics depending on the input data distribution. Existing database…

With distributed machine learning being a prominent technique for large-scale machine learning tasks, communication complexity has become a major bottleneck for speeding up training and scaling up machine numbers. In this paper, we propose…

Recent technological advancements have enabled proliferated use of small embedded and IoT devices for collecting, processing, and transferring the security-critical information and user data. This exponential use has acted as a catalyst in…

We experimentally verify a fast control plane with 100 microseconds of configuration time that can support more than 1000 racks, leveraged by a software-defined network controller and an industrial real-time Ethernet standard EtherCAT.

While Mixture-of-Experts (MoE) scales model capacity without proportionally increasing computation, its massive total parameter footprint creates significant storage and memory-access bottlenecks, which hinder efficient end-side deployment…

Serverless computing promises enhanced resource efficiency and lower user costs, yet is burdened by a heavyweight, CPU-bound data plane. Prior efforts exploiting shared memory reduce overhead locally but fall short when scaling across…

Domain-specific accelerators are used in various computing systems ranging from edge devices to data centers. Coarse-grained reconfigurable arrays (CGRAs) represent an architectural midpoint between the flexibility of an FPGA and the…

The integration of dynamic, sparse structures like Mixture-of-Experts (MoE) with parameter-efficient adapters (e.g., LoRA) is a powerful technique for enhancing Large Language Models (LLMs). However, this architectural enhancement comes at…

Reconfigurable computing offers a good balance between flexibility and energy efficiency. When combined with software-programmable devices such as CPUs, it is possible to obtain higher performance by spatially distributing the…

LoRA is a technique that reduces the number of trainable parameters in a neural network by introducing low-rank adapters to linear layers. This technique is used both for fine-tuning and full training of large language models. This paper…

In this article, we consider the problem of relay assisted computation offloading (RACO), in which user A aims to share the results of computational tasks with another user B through wireless exchange over a relay platform equipped with…

In large-scale LLM pre-training systems with 100k+ GPUs, failures become the norm rather than the exception, and restart costs can dominate wall-clock training time. However, existing fault-tolerance mechanisms are largely unprepared for…

AI transport libraries move bytes efficiently, but they commonly assume that buffers are already correctly allocated, placed, shared, registered, and safe under completion and teardown pressure. This paper presents dmaplane, a Linux kernel…

Responding to the "datacenter tax" and "killer microseconds" problems for datacenter applications, diverse solutions including Smart NIC-based ones have been proposed. Nonetheless, they often suffer from high overhead of communications over…

Data-intensive applications in data centers, especially machine learning (ML), have made the network a bottleneck, which in turn has motivated the development of more efficient network protocols and infrastructure. For instance, remote…

Federated cross-modal retrieval faces severe challenges from heterogeneous client data, particularly non-IID semantic distributions and missing modalities. Under such heterogeneity, a single global model is often insufficient to capture…

Serverless platforms essentially face a tradeoff between container startup time and provisioned concurrency (i.e., cached instances), which is further exaggerated by the frequent need for remote container initialization. This paper presents…