Data communication is a critical element of applications inside the data center and cloud server infrastructure. Muti-tiered architectures require communication among edge servers, application servers, network appliances, and database servers. The overhead of traditional networking protocols limits the performance of communication-centric, distributed applications, but these protocols provide well-known standard interfaces for application programmers. Cross-layer optimizations are needed to increase throughput and reduce latency and power, while still providing programmer-friendly abstractions.

This workshop will bring together researchers from all layers of the network stack: physical interfaces, data transport and transformation, middleware services, and application-layer APIs and languages. Papers that discuss original research are preferred, but excellent survey/tutorial papers will also be considered. Topics of interest include, but are not limited to:

- High-performance system-area networks (SANs)
- Low-overhead network protocols
- APIs and language constructs for high-performance communication
- Performance evaluation of cloud/data center communication infrastructure
- Workload characterization
- RDMA hardware, protocol stacks, and applications
- Network appliances and hardware accelerators
- Data transformation techniques
- Communication services (e.g., caching, migration) that exploit high-performance SANs
- Interaction of network hardware with processors, memory
- Techniques for full-system design and cross-layer optimization
- Virtualization of communication resources
- Power and energy tradeoffs in communication architectures
- Integration of intra-server and inter-server communication
- Storage area networks