Rearchitecting the {TCP} Stack for {I/O-Offloaded} Content Delivery
The recent advancement of high-bandwidth I/O devices enables scalable delivery of online
content. Unfortunately, the traditional programming model for content servers has a tight …
content. Unfortunately, the traditional programming model for content servers has a tight …
{FVM}:{FPGA-assisted} Virtual Device Emulation for Fast, Scalable, and Flexible Storage Virtualization
Emerging big-data workloads with massive I/O processing require fast, scalable, and flexible
storage virtualization support. Hardware-assisted virtualization can achieve reasonable …
storage virtualization support. Hardware-assisted virtualization can achieve reasonable …
Data motion acceleration: Chaining cross-domain multi accelerators
There has been an arms race for devising accelerators for deep learning in recent years.
However, real-world applications are not only neural networks but often span across …
However, real-world applications are not only neural networks but often span across …
BM-Store: A Transparent and High-performance Local Storage Architecture for Bare-metal Clouds Enabling Large-scale Deployment
Bare-metal instances are crucial for high-value, mission-critical applications on the cloud.
Tenants exclusively use these dedicated hardware resources. Local virtualized disks are …
Tenants exclusively use these dedicated hardware resources. Local virtualized disks are …
TrainBox: an extreme-scale neural network training server architecture by systematically balancing operations
Neural network is a major driving force of another golden age of computing; the computer
architects have proposed specialized accelerators (eg, TPU), high-speed interconnects (eg …
architects have proposed specialized accelerators (eg, TPU), high-speed interconnects (eg …
FIDR: A scalable storage system for fine-grain inline data reduction with efficient memory handling
Storage systems play a critical role in modern servers which run highly data-intensive
applications. To satisfy the high performance and capacity demands of such applications …
applications. To satisfy the high performance and capacity demands of such applications …
PREBA: A Hardware/Software Co-Design for Multi-Instance GPU based AI Inference Servers
NVIDIA's Multi-Instance GPU (MIG) is a feature that enables system designers to reconfigure
one large GPU into multiple smaller GPU slices. This work characterizes this emerging GPU …
one large GPU into multiple smaller GPU slices. This work characterizes this emerging GPU …
Smartfvm: A fast, flexible, and scalable hardware-based virtualization for commodity storage devices
A computational storage device incorporating a computation unit inside or near its storage
unit is a highly promising technology to maximize a storage server's performance. However …
unit is a highly promising technology to maximize a storage server's performance. However …
High accuracy positioning for C-V2X
Q Liu, M Song, X Xv, J Qiu - IOP Conference Series: Earth and …, 2021 - iopscience.iop.org
With the rapid development and popularization of 5G and C-V2X, services based on C-V2X
are rapidly expanding. Specially, the positioning accuracy is the most basic requirement in …
are rapidly expanding. Specially, the positioning accuracy is the most basic requirement in …
Accelerating Data Movement at Different Granularities in Datacenters
ST Wang - 2024 - escholarship.org
The dissertation investigates redundant communication between servers for large-scaleweb
and cache requests and redundant data movement between accelerators for compute …
and cache requests and redundant data movement between accelerators for compute …