publications

2026

1. SIGCOMM 26

   Credit-Guided Congestion Control on Wafer-Scale On-Chip Networks for Molecular Dynamics (To Appear)

   Tao Jiang, Shixiong Qi*, Zhan Wang, Ning Kang, Fan Yang, Yuanzhe Wang, Zejun Li, Guanglei Chen, Jing Xu, Wenzhe Li, Guangming Tan, and Guojun Yuan
2. EuroSys 26

   Not A DPU in Name Only! Unleashing RDMA-capable DPUs in Multi-Tenant Serverless Clouds with NADINO

   Shixiong Qi, Songyu Zhang, K. K. Ramakrishnan, Diman Zad Tootaghaj, Hardik Soni, and Puneet Sharma

   Abs DOI Code

   Serverless computing promises enhanced resource efficiency and lower user costs, yet is burdened by a heavyweight, CPU-bound data plane. Prior efforts exploiting shared memory reduce overhead locally but fall short when scaling across nodes. Furthermore, serverless environments can have unpredictable and large-scale multi-tenancy, leading to contention for shared network resources.We present NADINO, a DPU-centric serverless data plane that reduces the CPU burden and enables efficient, zero-copy communication in multi-tenant serverless clouds. Despite the limited general-purpose processing capability of the DPU cores, NADINO strategically exploits the DPU’s potential by (1) offloading data transmission to high-performance NIC cores via RDMA, combined with intra-node shared memory to eliminate data copies across nodes, and (2) enabling cross-processor (CPU-DPU) shared memory to eliminate redundant data movement, which overwhelms wimpy DPU cores. At the core of NADINO is the DPU-enabled network engine (DNE) - a lightweight reverse proxy that isolates RDMA resources from tenant functions, orchestrates inter-node RDMA flows, and enforces fairness under contention.To further reduce CPU involvement, NADINO performs early HTTP/TCP-to-RDMA transport conversion at the cloud ingress, bridging the protocol mismatch before client traffic enters the RDMA fabric, thus avoiding costly protocol translation along the critical path. We show that careful selection of RDMA primitives (i.e., two-sided instead of one-sided) significantly affects the zero-copy data plane.Our preliminary experimental results show that enabling DPU offloading in NADINO improves RPS by 20.9\texttimes. The latency is reduced by a factor of 21\texttimes in the best case, all the while saving up to 7 CPU cores, and only consuming two wimpy DPU cores.

3. IEEE TCC

   SURE: A High-performance, Efficient, and Secure Serverless Framework based on Unikernels

   Federico Parola, Shixiong Qi*, Anvaya B. Narappa, K. K. Ramakrishnan, and Fulvio Risso

   DOI Code

2025

1. SIGCOMM 25

   Palladium: A DPU-enabled Multi-Tenant Serverless Cloud over Zero-copy Multi-node RDMA Fabrics

   Shixiong Qi, Songyu Zhang, K. K. Ramakrishnan, Diman Zad Tootaghaj, Hardik Soni, and Puneet Sharma

   Abs DOI

   Serverless computing offers resource efficiency but suffers from a heavyweight data plane. We present Palladium, a DPU-offloaded serverless data plane enabling distributed zero-copy communication. Palladium uses two-sided RDMA and cross-processor shared memory to mitigate limitations of wimpy DPU cores. Its DPU-enabled network engine (DNE) isolates RDMA resources and manages flows across tenants. By converting HTTP/TCP to RDMA at ingress, Palladium reduces protocol overhead on the critical path.

2. free5GC 25

   Rethinking 5G Core for Performance, Fairness, and Flexibility

   Yang-Zhe Lin, Chun-Ting Lin, Po-Yi Lin, Shixiong Qi*, Jyh-Cheng Chen, and K. K. Ramakrishnan

   Best Paper Award Abs DOI Code

   This paper received the Best Paper Award at the 1st free5GC World Forum 2025

   The 5G Core (5GC) adopts a service-based architecture that enables flexible deployment of network functions (NFs) across the control and user planes. Despite its advantages, building a 3GPP-compliant, high-performance 5GC remains a significant challenge. In this paper, we present L25GC+, a 3GPP-compliant 5GC platform designed for low-latency 5GC operations. L25GC+ advances our previous prototype, L25GC, by replacing the kernel-based Service-Based Interface (SBI) with X-IO, a shared-memory-based SBI that maintains HTTP-style semantics while significantly reducing inter-NF communication latency. The platform integrates Golang-based NFs from free5GC, supports seamless upgrades to 3GPP Release 17, and introduces several key features: (1) enhanced QoS enforcement in the UPF with fairness guarantees, (2) rate limiting for hotspot control plane NFs (e.g., AMF) to prevent thrashing, (3) network slicing to enable multi-tenant isolation and amortize polling overhead.We evaluate L25GC+ on the NSF FABRIC testbed in a geographically distributed deployment, integrating the RAN, core, and data network across multiple sites. L25GC+ is being made available as an open Core-as-a-Service to support the broader 5G research community.

2024

1. ACM SoCC 24

   SURE: Secure Unikernels Make Serverless Computing Rapid and Efficient

   Federico Parola, Shixiong Qi, Anvaya B. Narappa, K. K. Ramakrishnan, and Fulvio Risso

   Abs DOI Code

   Current serverless platforms introduce non-trivial overheads when chaining and orchestrating loosely-coupled microservices. Containerized function runtimes are also constrained by insufficient isolation and excessive startup time. This motivates our exploration of a more efficient, secure, and rapid serverless design. We describe SURE, a unikernel-based serverless framework for fast function startup, equipped with a high-performance and secure data plane. SURE’s data plane supports distributed zero-copy communication via the seamless interaction between zero-copy protocol stack (Z-stack) and local shared memory processing. To establish a lightweight service mesh, SURE uses library-based sidecars instead of individual userspace sidecars. We leverage Intel’s Memory Protection Keys (MPK) as a lightweight capability to ensure safe access to the shared memory data plane. It also isolates the Trusted Computing Base (TCB) components in SURE’s function runtime (e.g., library-based sidecar, scheduler, etc) from untrusted user code, while preserving the efficient single-address-space nature of unikernels. In particular, SURE prevents unintended privilege escalation involving MPK with an enhanced TCB. These combined efforts create a more secure and robust data plane while improving throughput up to 79X over Knative, a representative open-source serverless platform.

2. LANMAN 24

   Z-Stack: A High-Performance DPDK-Based Zero-Copy TCP/IP Protocol Stack

   Anvaya B. Narappa, Federico Parola, Shixiong Qi, and K. K. Ramakrishnan

   DOI Code
3. MLSys 24

   LIFL: A Lightweight, Event-driven Serverless Platform for Federated Learning

   Shixiong Qi, K. K. Ramakrishnan, and Myungjin Lee

   PDF Code
4. Internet Computing

   L26GC: Evolving the Low-Latency Core for Future Cellular Networks

   Shixiong Qi, K. K. Ramakrishnan, and Jyh-Cheng Chen

   DOI
5. IEEE/ACM ToN

   SPRIGHT: High-Performance eBPF-Based Event-Driven, Shared-Memory Processing for Serverless Computing

   Shixiong Qi, Leslie Monis, Ziteng Zeng, Ian-Chin Wang, and K. K. Ramakrishnan

   DOI Code

2023

1. NetSoft 23

   X-IO: A High-performance Unified I/O Interface using Lock-free Shared Memory Processing

   Shixiong Qi, Han-Sing Tsai, Yu-Sheng Liu, K. K. Ramakrishnan, and Jyh-Cheng Chen

   DOI Code
2. CloudNet 23

   L25GC+: An Improved, 3GPP-compliant 5G Core for Low-latency Control Plane Operations

   Yu-Sheng Liu, Shixiong Qi, Po-Yi Lin, Han-Sing Tsai, K. K. Ramakrishnan, and Jyh-Cheng Chen

   DOI Code
3. IEEE TNSM

   MiddleNet: A Unified, High-Performance NFV and Middlebox Framework With eBPF and DPDK

   Shixiong Qi, Ziteng Zeng, Leslie Monis, and K. K. Ramakrishnan

   DOI

2022

1. SIGCOMM 22

   L25GC: a low latency 5G core network based on high-performance NFV platforms

   Vivek Jain, Hao-Tse Chu, Shixiong Qi, Chia-An Lee, Hung-Cheng Chang, Cheng-Ying Hsieh, K. K. Ramakrishnan, and Jyh-Cheng Chen

   Abs DOI Code

   Cellular network control procedures (e.g., mobility, idle-active transition to conserve energy) directly influence data plane behavior, impacting user-experienced delay. Recognizing this control-data plane interdependence, L25GC re-architects the 5G Core (5GC) network, and its processing, to reduce latency of control plane operations and their impact on the data plane. Exploiting shared memory, L25GC eliminates message serialization and HTTP processing overheads, while being 3GPP-standards compliant. We improve data plane processing by factoring the functions to avoid control-data plane interference, and using scalable, flow-level packet classifiers for forwarding-rule lookups. Utilizing buffers at the 5GC, L25GC implements paging, and an intelligent handover scheme avoiding 3GPP’s hairpin routing, and data loss caused by limited buffering at 5G base stations, reduces delay and unnecessary message processing. L25GC’s integrated failure resiliency transparently recovers from failures of 5GC software network functions and hardware much faster than 3GPP’s reattach recovery procedure. L25GC is built based on free5GC, an open-source kernel-based 5GC implementation. L25GC reduces event completion time by  50% for several control plane events and improves data packet latency (due to improved control plane communication) by  2\texttimes, during paging and handover events, compared to free5GC. L25GC’s design is general, although current implementation supports a limited number of user sessions.

2. SIGCOMM 22

   SPRIGHT: extracting the server from serverless computing! high-performance eBPF-based event-driven, shared-memory processing

   Shixiong Qi, Leslie Monis, Ziteng Zeng, Ian-chin Wang, and K. K. Ramakrishnan

   Abs DOI Code

   Serverless computing promises an efficient, low-cost compute capability in cloud environments. However, existing solutions, epitomized by open-source platforms such as Knative, include heavyweight components that undermine this goal of serverless computing. Additionally, such serverless platforms lack dataplane optimizations to achieve efficient, high-performance function chains that facilitate the popular microservices development paradigm. Their use of unnecessarily complex and duplicate capabilities for building function chains severely degrades performance. ’Cold-start’ latency is another deterrent.We describe SPRIGHT, a lightweight, high-performance, responsive serverless framework. SPRIGHT exploits shared memory processing and dramatically improves the scalability of the dataplane by avoiding unnecessary protocol processing and serialization-deserialization overheads. SPRIGHT extensively leverages event-driven processing with the extended Berkeley Packet Filter (eBPF). We creatively use eBPF’s socket message mechanism to support shared memory processing, with overheads being strictly load-proportional. Compared to constantly-running, polling-based DPDK, SPRIGHT achieves the same dataplane performance with 10\texttimes less CPU usage under realistic workloads. Additionally, eBPF benefits SPRIGHT, by replacing heavyweight serverless components, allowing us to keep functions ’warm’ with negligible penalty.Our preliminary experimental results show that SPRIGHT achieves an order of magnitude improvement in throughput and latency compared to Knative, while substantially reducing CPU usage, and obviates the need for ’cold-start’.

3. 6GNet 22

   Evolving to 6G: Improving the Cellular Core to lower control and data plane latency

   Vivek Jain, Sourav Panda, Shixiong Qi, and K. K. Ramakrishnan

   Best Paper Award DOI

   This paper received the Best Paper Award at the 2022 6GNet Conference

4. NetSoft 22

   MiddleNet: A High-Performance, Lightweight, Unified NFV and Middlebox Framework

   Ziteng Zeng, Leslie Monis, Shixiong Qi, and K. K. Ramakrishnan

   DOI

2021

1. ACM SoCC 21

   Mu: An Efficient, Fair and Responsive Serverless Framework for Resource-Constrained Edge Clouds

   Viyom Mittal, Shixiong Qi, Ratnadeep Bhattacharya, Xiaosu Lyu, Junfeng Li, Sameer G. Kulkarni, Dan Li, Jinho Hwang, K. K. Ramakrishnan, and Timothy Wood

   Abs DOI

   Serverless computing platforms simplify development, deployment, and automated management of modular software functions. However, existing serverless platforms typically assume an over-provisioned cloud, making them a poor fit for Edge Computing environments where resources are scarce. In this paper we propose a redesigned serverless platform that comprehensively tackles the key challenges for serverless functions in a resource constrained Edge Cloud.Our Mu platform cleanly integrates the core resource management components of a serverless platform: autoscaling, load balancing, and placement. Each worker node in Mu transparently propagates metrics such as service rate and queue length in response headers, feeding this information to the load balancing system so that it can better route requests, and to our autoscaler to anticipate workload fluctuations and proactively meet SLOs. Data from the Autoscaler is then used by the placement engine to account for heterogeneity and fairness across competing functions, ensuring overall resource efficiency, and minimizing resource fragmentation. We implement our design as a set of extensions to the Knative serverless platform and demonstrate its improvements in terms of resource efficiency, fairness, and response time.Evaluating Mu, shows that it improves fairness by more than 2x over the default Kubernetes placement engine, improves 99th percentile response times by 62% through better load balancing, reduces SLO violations and resource consumption by pro-active and precise autoscaling. Mu reduces the average number of pods required by more than  15% for a set of real Azure workloads.

2. NAS 21

   Towards a Proactive Lightweight Serverless Edge Cloud for Internet-of-Things Applications

   Ian-Chin Wang, Shixiong Qi, Elizabeth Liri, and K. K. Ramakrishnan

   DOI
3. LANMAN 21

   Fast Function Instantiation with Alternate Virtualization Approaches

   Vivek Jain, Shixiong Qi, and K. K. Ramakrishnan

   DOI
4. IEEE TNSM

   Assessing Container Network Interface Plugins: Functionality, Performance, and Scalability

   Shixiong Qi, Sameer G. Kulkarni, and K. K. Ramakrishnan

   DOI

2020

1. LANMAN 20

   Understanding Container Network Interface Plugins: Design Considerations and Performance

   Shixiong Qi, Sameer G Kulkarni, and K. K. Ramakrishnan

   DOI

2019

1. OSN

   SMONoC: Optical network-on-chip using a statistical multiplexing strategy

   Kun Wang, Shixiong Qi, Zheng Chen, Yintang Yang, and Huaxi Gu

   Abs DOI

   Optical interconnect can provide high bandwidth and low latency, which is considered as a good candidate for future multicore systems. To maintain the scalability of optical networks and simultaneously support high performance communication, we propose a Statistical Multiplexing Optical Network-on-Chip (SMONoC), a new architecture targeting dynamic channel allocation for high channel utilization, a modest amount of photonic devices for low integration cost, and statistical multiplexing for a simple arbitration scheme. The SMONoC consists of two layers, one optical layer for data transmission and one electrical layer for controlling and scheduling. With the introduction of the statistical multiplexing strategy, the SMONoC avoids the confliction between different cores with low complexity. The simulation results show that the SMONoC is a low-latency and high-throughput architecture with high channel utilization under both synthetic traffic patterns and real traffic patterns.

2. IEICE ELEX

   ALPHA: A hybrid topology for memory-centric network

   Jiaxiang Li, Huaxi Gu, Shixiong Qi, Haoran Wang, and Kang Wang

   DOI

2017

1. GLOBECOM 17

   Testudo: A Low Latency and High-Efficient Memory-Centric Network Using Optical Interconnect

   Shixiong Qi, Huaxi Gu, Haibo Zhang, and Yawen Chen

   DOI
2. ICOCN 17

   LACE: A non-blocking on-chip optical router by utilizing the wavelength routing technology

   Lei Huang, Shixiong Qi, Kun Wang, and Huaxi Gu

   DOI
3. ICOCN 17

   A low-crosstalk optical router using multi-layer coupled MR for ONoC

   Xinglong Diao, Lei Huang, Wei Tan, Shixiong Qi, and Huaxi Gu

   DOI

2016

1. IEICE ELEX

   Panzer: A 6 × 6 photonic router for optical network on chip

   Lei Huang, Kun Wang, Shixiong Qi, Huaxi Gu, and Yintang Yang

   DOI

2015

1. ICCP 15

   Crosstalk analysis for closed ring-based optical network-on-chip

   Shixiong Qi, Kun Wang, Huaxi Gu, Kang Wang, and Xiaolu Wang

   DOI