Related papers: IOTLB-SC: An Accelerator-Independent Leakage Sourc…

Transformers and LLMs have seen rapid adoption in all domains. Their sizes have exploded to hundreds of billions of parameters and keep increasing. Under these circumstances, the training of transformers is slow and often takes in the order…

The high memory and computation demand of large language models (LLMs) makes them challenging to be deployed on consumer devices due to limited GPU memory. Offloading can mitigate the memory constraint but often suffers from low GPU…

GPUs are critical for compute-intensive applications, yet emerging workloads such as recommender systems, graph analytics, and data analytics often exceed GPU memory capacity. Existing solutions allow GPUs to use CPU DRAM or SSDs as…

The rapid adoption of AI and convenience offered by cloud services have resulted in the growing demands for GPUs in the cloud. Generally, GPUs are physically attached to host servers as PCIe devices. However, the fixed assembly combination…

In recent decades, due to the emerging requirements of computation acceleration, cloud FPGAs have become popular in public clouds. Major cloud service providers, e.g. AWS and Microsoft Azure have provided FPGA computing resources in their…

The vast amount of processing power and memory bandwidth provided by modern Graphics Processing Units (GPUs) make them a platform for data-intensive applications. The database community identified GPUs as effective co-processors for data…

When handling large datasets that exceed the capacity of the main memory, movement of data between main memory and external memory (disk), rather than actual (CPU) computation time, is often the bottleneck in the computation. Since data is…

Meltdown and Spectre exploit microarchitectural changes the CPU makes during transient out-of-order execution. Using side-channel techniques, these attacks enable leaking arbitrary data from memory. As state-of-the-art software mitigations…

Existing GPU spatial sharing systems face a three-way tradeoff: resource utilization, performance isolation, and semantic determinism. Hardware partitioning suffers from hardware under-utilization. Hardware multiplexing fails to avoid…

The discoveries in this paper show that Intelligence Processing Units (IPUs) offer a viable accelerator alternative to GPUs for machine learning (ML) applications within the fields of materials science and battery research. We investigate…

In modern server CPUs, last-level cache (LLC) is a critical hardware resource that exerts significant influence on the performance of the workloads, and how to manage LLC is a key to the performance isolation and QoS in the cloud with…

Fast-evolving machine learning (ML) workloads have increasing requirements for networking. However, host network transport on RDMA NICs is hard to evolve, causing problems for ML workloads. For example, single-path RDMA traffic is prone to…

This paper introduces Out of Hypervisor (OoH), a new research axis close to nested virtualization. Instead of emulating a full virtual hardware inside a VM to support a hypervisor, the OoH principle is to individually expose current…

With the exponentially increasing demand for performance and scalability in cloud applications and systems, data center architectures evolved to integrate heterogeneous computing fabrics that leverage CPUs, GPUs, and FPGAs. FPGAs differ…

Latency-sensitive inference on shared A100 clusters often suffers noisy-neighbor interference on the PCIe fabric, inflating tail latency and SLO violations. We present a fabric-agnostic, VM-deployable host-level controller that combines…

Attacks based on side-channel analysis (SCA) pose a severe security threat to modern computing platforms, further exacerbated on IoT devices by their pervasiveness and handling of private and critical data. Designing SCA-resistant computing…

Non-orthogonal multiple access (NOMA) is an interesting technology that enables massive connectivity as required in future 5G and 6G networks. While purely linear processing already achieves good performance in NOMA systems, in certain…

The continued growth of the computational capability of throughput processors has made throughput processors the platform of choice for a wide variety of high performance computing applications. Graphics Processing Units (GPUs) are a prime…

Pipelining between data loading and computation is a critical tensor program optimization for GPUs. In order to unleash the high performance of latest GPUs, we must perform a synergetic optimization of multi-stage pipelining across the…

Modern microprocessors depend on speculative execution, creating vulnerabilities that enable transient execution attacks. Prior defenses target speculative data leakage but overlook false dependencies from partial address aliasing, where…