Given the wide adoption of disaggregated storage in public clouds, networking is the key to
enabling high performance and high reliability in a cloud storage service. In Azure, we …

RDMA is expected to be highly scalable: to perform well in large-scale data center networks
where packet losses are inevitable (ie, high network scalability), and to support a large …

Data serialization is critical for many datacenter applications, but the memory copies
required to move application data into packets are costly. Recent zero-copy APIs expose …

Modern network hardware is able to meet the stringent bandwidth demands of applications
like GPU-accelerated AI. However, existing host network stacks offer a hard tradeoff …

Recent years have witnessed the wide adoption of RDMA in the cloud to accelerate first-
party workloads and achieve cost savings by freeing up CPU cycles. Now cloud providers …

The switch buffers in datacenters today are shared by traffic classes with different loss
tolerance and reaction to congestion signals. In particular, while legacy applications use …

Hardware offloaded network stacks are widely adopted in modern datacenters to meet the
demand for high throughput, ultra-low latency and low CPU overhead. To fully leverage their …

Remote Procedure Call (RPC) is a widely used abstraction for cloud computing. The
programmer specifies type information for each remote procedure, and a compiler generates …

SmartNICs have become an indispensable communication fabric and computing substrate
in today's data centers and enterprise clusters, providing in-network computing capabilities …

Intra-host networking was considered robust in the RDMA (Remote Direct Memory Access)
network and received little attention. However, as the RNIC (RDMA NIC) line rate increases …