Network bandwidth continues to scale rapidly, yet Internet data transmission performance remains constrained by the legacy 1500 B MTU. This small MTU translates high bandwidth into high packet rates that strain CPU processing at middleboxes and end hosts. While increasing the MTU could substantially improve performance, coordinating upgrades across arbitrary Internet paths is impractical.

Workload traces are essential to understand complex computer systems' behavior and manage processing and memory resources.Since real-world traces are hard to obtain, synthetic trace generation is a promising alternative.This paper proposes a first-of-a-kind approach that relies on training a large language model (LLM) to generate synthetic workload traces, specifically microservice call graphs.To capture complex and arbitrary hierarchical structures and implicit constraints in such traces, we propose to train LLMs to generate recursively, making call graph generation a sequence of more manageable steps.To further enforce learning constraints on the traces and generate uncommon situations, we apply additional instruction tuning steps to align our model with the desired trace features.With this method, we train TraceLLM, an LLM for microservice trace generation, and demonstrate that it produces diverse, realistic traces under varied conditions, outperforming existing approaches in both accuracy and validity.The synthetically generated traces can effectively replace real data to optimize important microservice management tasks.Additionally, TraceLLM adapts to downstream trace-related tasks, such as predicting key trace features and infilling missing data.

Policy design for various systems controllers has conventionally been a manual process, with domain experts carefully tailoring heuristics for the specific instance in which the policy will be deployed. In this paper, we re-imagine policy design via a novel automated search technique fueled by recent advances in generative models, specifically Large Language Model (LLM)-driven code generation. We outline the design and implementation of PolicySmith, a framework that applies LLMs to synthesize instance-optimal heuristics. We apply PolicySmith to two long-standing systems policies - web caching and congestion control, highlighting the opportunities unraveled by this LLM-driven heuristic search. For caching, PolicySmith discovers heuristics that outperform established baselines on standard open-source traces. For congestion control, we show that PolicySmith can generate safe policies that integrate directly into the Linux kernel.

This paper proposes a systematic approach to incrementally enabling large MTUs in the Internet. We demonstrate that increasing the MTU size significantly enhances the performance of both middleboxes and end hosts. To bridge MTU mismatches at network borders, we introduce PacketExpress gateway (PXGW), an MTU-translating gateway that dynamically adjusts packet sizes for cross-traffic. PXGW merges and splits TCP payloads on the fly and tunnels UDP packets, ensuring seamless adaptation. Also, we propose F-PMTUD, a new path MTU discovery algorithm that determines the path MTU within a single round-trip without relying on ICMP. Our preliminary evaluation shows that the PXGW prototype achieves 1.45 Tbps of packet forwarding throughput using only 8 CPU cores. After dynamic conversion, 94% of transmitted TCP packets are 9000 B jumbo frames, indicating that most flows were effectively converted into large segments, thereby demonstrating the system's efficiency and scalability. We also find that large-MTU packets, made available via PXGW, enhance end-host performance by up to 2.5X.

The growing use of service robots in dynamic environments requires flexible management of on-board compute resources to optimize the performance of diverse tasks such as navigation, localization, and perception. Current robot deployments often rely on static OS configurations and system over-provisioning. However, they are suboptimal because they ignore variations in resource usage, leading to system-wide issues like robot instability or inefficient resource utilization. This paper presents ConfigBot, a novel system designed to adaptively reconfigure robot applications to meet a predefined performance specification by leveraging runtime profiling and automated configuration tuning. Through experiments on multiple real robots, each running a different stack with diverse performance requirements, which could be context-dependent, we illustrate ConfigBot's efficacy in maintaining system stability and optimizing resource allocation. Our findings highlight the promise of automatic system configuration tuning for robot deployments, including adaptation to dynamic changes. Code available at: https://github.com/ldos-project/configbot

While mobile live video analytics apps require end-to-end latency guarantee for responsiveness and immersiveness, achieving consistent low latency is challenging due to complex fluctuations of wireless channel and scene complexity; for example, latency SLO satisfaction rate drops to as low as 26% in commercial 5G MEC platforms. Prior works mostly focus on either app-only (bitrate, DNN adaptation, or GPU allocation) or RAN-only (radio resource allocation) scheduling, with mutual ignorance of the other side resulting in mismatched scheduling decisions and frequent SLO violations. Coordinating the two schedulers is also challenging, as they are run separately by network and cloud operators with disjoint control. We present ARMA, an end-to-end live video analytics system with app-RAN mutual-awareness for high end-to-end latency SLO satisfaction in MEC. We design a mutually-aware decoupled scheduling mechanism on top of RAN Intelligent Controller (RIC) in Open-RAN architecture that fosters cooperative interaction between the two operators' schedulers while preserving operational proprietaries. We prototype an Open RAN-enabled 5G MEC testbed and evaluate ARMA, showing that ARMA achieves 97% SLO satisfaction rate.

Video Corpus Moment Retrieval aims to select a temporal video moment pertinent to a given language query from a large video corpus. Existing systems are prone to rely on a retrieval bias as a shortcut, which hinders the systems from accurately learning vision-language association. The retrieval bias is spurious correlations between query and scene. For a given query, systems tend to retrieve incorrectly correlated scenes due to biased annotations that have predominant binding in a dataset. To this end, we present a Counterfactual Two-stage Debiasing Learning (CTDL), which incorporates a counterfactual bias network that intentionally learns the retrieval bias by providing a shortcut to learn the spurious correlation between keyword and scene, and performs two-stage debiasing learning that mitigates the bias via contrasting factual retrievals with counterfactually biased retrievals. Extensive experiments show the effectiveness of CTDL paradigm.

SmartNICs have become an indispensable communication fabric and computing substrate in today’s data centers and enterprise clusters, providing in-network computing capabilities for traversed packets and benefiting a range of applications across the system stack. Building an efficient SmartNIC-assisted solution is generally non-trivial and tedious as it requires programmers to understand the SmartNIC architecture, refactor application logic to match the device’s capabilities and limitations, and correlate an application execution with traffic characteristics. A high-level SmartNIC performance model can decouple the underlying SmartNIC hardware device from its offloaded software implementations and execution contexts, thereby drastically simplifying and facilitating the development process. However, prior architectural models can hardly be applied due to their limited capabilities in dissecting the SmartNIC-offloaded program’s complexity, capturing the nondeterministic overlapping between computation and I/O, and perceiving diverse traffic profiles.

Open RAN is an emerging paradigm in mobile networks where the Radio Access Network (RAN) functions are disaggregated and virtualized on commodity servers. Despite the importance of Open RAN research, existing platforms often lack the fidelity and stability required to address a wide range of research problems. In response to this limitation, we have developed an enterprise-scale Open RAN testbed aimed at conducting state-of-the-art research in key areas that have received limited attention due to the lack of suitable platforms. In this poster, we provide an overview of the testbed we have created and examples of the research it has enabled, with the hope of catalyzing future open RAN research and innovation.

Virtualized radio access networks (vRANs), which allow running RAN processing on commodity servers instead of proprietary hardware, are gaining adoption in cellular networks. Two properties of the vRAN's "Distributed Unit (DU)" that implements the lower RAN layers---its real-time deadlines and its black-box nature---make it challenging to provide resilience features such as upgrades and failover without long service disruptions. These properties preclude the use of existing resilience techniques like virtual machine migration or state replication that are used for typical workloads. This paper presents Atlas, the first system that provides resilience for the DU. The central insight in Atlas is to repurpose existing cellular mechanisms for wireless resilience, namely handovers and cell reselection, to provide software resilience for the DU. For planned resilience events like upgrades, we design a novel technique that simultaneously serves cells from both the old and new DUs via the same radio, and uses handovers between these cells to migrate user devices. For unplanned failures, we identify deficiencies in existing RAN protocols that disrupt cell reselection after DU failure, and show how we can eliminate these disruptions using a middlebox between the DU and higher layers. Our evaluation with a state-of-the-art 5G vRAN testbed shows that Atlas achieves minimal disruption to cellular connectivity during resilience events, while incurring low overhead.

In cellular networks, there is a growing adoption of virtualized radio access networks (vRANs), where operators are replacing the traditional specialized hardware for RAN processing with software running on commodity servers. Today's vRAN deployments lack resilience, since there is no support for vRAN failover or upgrades without long service interruptions. Enabling these features for vRANs is challenging because of their strict real-time latency requirements and black-box nature. Slingshot is a new system that transparently provides resilience for the vRAN's most performance-critical layer: the physical layer (PHY). We design new techniques for realtime workload migration with fast RAN protocol middle-boxes, and realtime RAN failure detection. A key insight in our design is to view the transient disruptions from resilience events to RAN computation state and I/O similarly to regular wireless signal impairments, and leverage the inherent resilience of cellular networks to these events. Experiments with a state-of-the-art 5G vRAN testbed show that Slingshot handles PHY failover with no disruption to video conferencing, and under 110 ms disruption to a TCP connection, and it also enables zero-downtime upgrades.

Many recent efforts have demonstrated the performance benefits of running datacenter functions (\emph{e.g.,} NATs, load balancers, monitoring) on programmable switches. However, a key missing piece remains: fault tolerance. This is especially critical as the network is no longer stateless and pure endpoint recovery does not suffice. In this paper, we design and implement RedPlane, a fault-tolerant state store for stateful in-switch applications. This provides in-switch applications consistent access to their state, even if the switch they run on fails or traffic is rerouted to an alternative switch. We address key challenges in devising a practical, provably correct replication protocol and implementing it in the switch data plane. Our evaluations show that RedPlane incurs negligible overhead and enables end-to-end applications to rapidly recover from switch failures.

Programmable switches have been touted as an attractive alternative for deploying network functions (NFs) such as network address translators (NATs), load balancers, and firewalls. However, their limited memory capacity has been a major stumbling block that has stymied their adoption for supporting state-intensive NFs such as cloud-scale NATs and load balancers that maintain millions of flow-table entries. In this paper, we explore a new approach that leverages DRAM on servers available in typical NFV clusters. Our new system architecture, called TEA (Table Extension Architecture), provides a virtual table abstraction that allows NFs on programmable switches to look up large virtual tables built on external DRAM. Our approach enables switch ASICs to access external DRAM purely in the data plane without involving CPUs on servers. We address key design and implementation challenges in realizing this idea. We demonstrate its feasibility and practicality with our implementation on a Tofino-based programmable switch. Our evaluation shows that NFs built with TEA can look up table entries on external DRAM with low and predictable latency (1.8-2.2 μs) and the lookup throughput can be linearly scaled with additional servers (138 million lookups per seconds with 8 servers).

Network switches are an attractive vantage point to serve various network applications and functions such as load balancing and virtual switching because of their in-network location and high packet processing rate. Recent advances in programmable switch ASICs open more opportunities for offloading various functionality to switches. However, the limited memory capacity on switches has been a major challenge that such applications struggle to deal with. In this paper, we envision that by enabling network switches to access remote memory purely from data planes, the performance of a wide range of applications can be improved. We design three remote memory primitives, leveraging RDMA operations, and show the feasibility of accessing remote memory from switches using our prototype implementation.

Storage systems in data centers are an important component of large-scale online services. They typically perform replicated transactional operations for high data availability and integrity. Today, however, such operations suffer from high tail latency even with recent kernel bypass and storage optimizations, and thus affect the predictability of end-to-end performance of these services. We observe that the root cause of the problem is the involvement of the CPU, a precious commodity in multi-tenant settings, in the critical path of replicated transactions. In this paper, we present HyperLoop, a new framework that removes CPU from the critical path of replicated transactions in storage systems by offloading them to commodity RDMA NICs, with non-volatile memory as the storage medium. To achieve this, we develop new and general NIC offloading primitives that can perform memory operations on all nodes in a replication group while guaranteeing ACID properties without CPU involvement. We demonstrate that popular storage applications can be easily optimized using our primitives. Our evaluation results with microbenchmarks and application benchmarks show that HyperLoop can reduce 99th percentile latency ≈ 800X with close to 0% CPU consumption on replicas.

Wireless video traffic has grown at an unprecedented rate and put significant burden on wireless networks. Multicast can significantly reduce traffic by sending a single video to multiple receivers simultaneously. On the other hand, wireless receivers are heterogeneous due to both channel and antenna heterogeneity, the latter of which is rapidly increasing with the emergence of 802.11n and 802.11ac. In this paper, we develop optimized layered integrated video encoding (LIVE) to guarantee reasonable performance to weaker receivers (with worse channel and/or fewer antennas) and allow stronger receivers to enjoy better quality. Our approach has three distinct features: (i) It uses a novel layered coding to naturally accommodate the heterogeneity of different video receivers; (ii) It uses an optimization framework to optimize the amount of time used for transmission and the amount of information to transmit at each layer under the current channel condition; and (iii) It uses an integrated modulation, where most video data are transmitted using soft modulation to enjoy efficiency and resilience while the most important video data are transmitted using a combination of soft modulation and conventional hard modulation to further enhance their reliability. To our knowledge, this is the first approach that handles MIMO antenna heterogeneity in wireless video multicast. We demonstrate its effectiveness through extensive Matlab simulation and USRP testbed experiments.

A variety of advantages from sounds such as measurement and accessibility introduces a new opportunity for mobile applications to offer broad types of interesting, valuable functionalities, supporting a richer user experience. However, in spite of the growing interests on mobile sound applications, few or no works have been done in focusing on managing an audio device effectively. More specifically, their low level of real-time capability for audio resources makes it challenging to satisfy tight timing requirements of mobile sound applications, e.g., a high sensing rate of acoustic sensing applications. To address this problem, this work presents the SounDroid framework, an audio device management framework for real-time audio requests from mobile sound applications. The design of SounDroid is based on the requirement analysis of audio requests as well as an understanding of the audio playback procedure including the audio request scheduling and dispatching on Android. It then incorporates both real-time audio request scheduling algorithms, called EDF-V and AFDS, and dispatching optimization techniques into mobile platforms, and thus improves the quality-of-service of mobile sound applications. Our experimental results with the prototype implementation of SounDroid demonstrate that it is able to enhance scheduling performance for audio requests, compared to traditional mechanisms (by up to 40% of improvement), while allowing deterministic dispatching latency.

Hardware-based external monitors have been proposed as a trustworthy method for protecting the kernel integrity. We introduce the design and implementation of Address Translation Redirection Attack (ATRA) that enables complete evasion of the hardware-based external monitor that anchors its trust on a separate processor. ATRA circumvents the external monitor by redirecting the memory access to critical kernel objects into a non-monitored region. Despite the seriousness of the ATRA issue, the address translation integrity has been assumed in many hardware-based external monitors and the possibility of its exploitation has been suggested yet many considered hypothetical. We explore the intricate details of ATRA, explain major challenges in realizing ATRA in practice, and address them with two types of ATRA called Memory-bound ATRA and Register-bound ATRA. Our evaluations with benchmarks show that ATRA does not introduce a noticeable performance degradation to the host system, proving practical applicability of the attack to alert the researchers to seriously address ATRA in designing future external monitors.

Explosive growth of WiFi traffic calls for new technologies to dramatically improve spectrum efficiency. In this paper, we propose an approach to adapt the spectrum on a per-frame basis. It consists of three major components: (i) a fine-grained spectrum access design that allows a sender and receiver to change their transmission and reception spectrum on demand, (ii) fast and accurate spectrum detection that allows a receiver to determine which spectrum is used by its sender on a per-frame basis by exploiting the IEEE 802.11 preamble structure, and (iii) an efficient spectrum allocation algorithm that determines which spectrum to use for each transmission by taking into account frequency diversity and interference. It can further be adapted to perform a joint assignment of spectrum, schedule, and access point (AP) for each frame. Using a SORA implementation and trace-driven simulation, we demonstrate the feasibility of per-frame spectrum adaptation and its significant benefit over existing channel assignment approaches. To our knowledge, this is the first per-frame spectrum adaptation prototype for WiFi networks.

Recently, wireless communication methods that exploit the broadcast nature of the wireless medium have been attracting growing attention. Among these methods, opportunistic routing and network coding are regarded as the most promising techniques. While there have been some attempts to combine opportunistic routing with network coding to capture the advantages of both techniques, none of these attempts has considered bit-rate selection for data transmission in multi-rate wireless networks. In this paper, we study the potential benefits of the combination of opportunistic routing and network coding with the bit-rate selection mechanism from an optimization perspective. We develop a theoretical model and algorithm for finding the optimal forwarding scheme for a multi-rate combination of opportunistic routing and network coding in a given network. MIT Roofnet trace-based simulations show that considering bit-rate selection in combination with opportunistic routing and network coding has substantial benefits in terms of the expected transmission time compared to multi-rate opportunistic routing, multi-rate network coding, and a fixed-rate combination approach.