Oobleck enables resilient distributed training of large DNN models with
...
Compute and memory are tightly coupled within each server in traditional...
Deep learning has experienced significant growth in recent years, result...
Cross-device federated learning (FL) has been well-studied from algorith...
Machine learning (ML) models can leak information about users, and
diffe...
Federated learning (FL) is an emerging machine learning (ML) paradigm th...
Existing DNN serving solutions can provide tight latency SLOs while
main...
Training deep neural networks (DNNs) is becoming increasingly more resou...
The need to train DNN models on end-user devices (e.g., smartphones) is
...
The increasing demand for memory in hyperscale applications has led to m...
Model aggregation, the process that updates model parameters, is an impo...
Training deep neural networks (DNNs) is time-consuming. While most exist...
The Internet of Things (IoT) is on the verge of a major paradigm shift. ...
We present Memtrade, the first memory disaggregation system for public
c...
In this paper we propose Fed-ensemble: a simple approach that bringsmode...
We present FedScale, a diverse set of challenging and realistic benchmar...
Federated Learning (FL) is an emerging direction in distributed machine
...
Simultaneously supporting latency- and throughout-sensitive workloads in...
Memory disaggregation over RDMA can improve the performance of
memory-co...
Memory disaggregation has received attention in recent years as a promis...
The coflow scheduling problem has emerged as a popular abstraction in th...
Despite its increasing popularity, most of RDMA's benefits such as ultra...
Geo-distributed analytics (GDA) frameworks transfer large datasets over ...
GPU computing is becoming increasingly more popular with the proliferati...