The paper introduces Spandana, an architecture that decouples SLO enforcement from cost optimization in cloud-based online services, achieving high utilization, strict SLO adherence, and cost savings.

SPARK introduces a predictive, traffic-aware autoscaling toolchain for Kubernetes that uses eBPF to enhance security and significantly reduce timeout errors during sudden traffic spikes.

This paper proposes a dependency-aware autoscaling framework for serverless computing, integrating graph-based bottleneck identification, short-term workload forecasting, multi-model consensus, and co…

This paper proposes a scheme to coordinate 5G and TSN schedulers for supporting deterministic communications with bounded latencies in industrial applications.

This paper proposes a distributed architecture for automating and controlling multi-vendor, multi-layer IP over DWDM networks using SDN, enabling end-to-end service lifecycle automation and closed-loo…

This paper proposes a new approach for designing Radio Access Network (RAN) slices in 5G and beyond networks using descriptors that consider both transmission rate and latency requirements to support…

This paper introduces Selective Field Transmission (SFT), a mechanism that dynamically adapts transmitted message components to each receiver's needs in publish-subscribe systems, achieving significan…

SOCI (Seekable OCI) is a lazy-loading architecture that reduces container image pulling time in Kubernetes environments by building an index over standard OCI images and serving file accesses via HTTP…

A cloud-scale gateway system for MCP services is presented, which breaks the direct-connect model and offloads legacy service integration, consolidates incompatible MCP variants, and reduces tool sele…

This paper proposes CAPS, a scheduling layer for data centers that separates rate computation and packet scheduling, reducing queue occupancy by up to 10x without throughput loss.

OpenURMA provides the first open, clean-room implementation of Huawei's Unified Bus (UB) protocol, demonstrating a significant reduction in latency and increase in throughput for remote memory access…

This paper proposes a proactive fair scheduling framework to prevent fabric oversubscription and eliminate incast in Mixture of Experts (MoE) architectures, demonstrating consistent link utilization a…

This paper proposes a carbon-aware routing policy for geo-distributed cloud deployments, achieving up to 46.8% carbon reduction while maintaining zero SLA violations.

The ZCube topology, which eliminates path multiplicity and reduces switching hardware, delivers better performance for large model training and inference than traditional multipath datacenter networks…