Optimizing burst-heavy datacenter workloads necessitates finegrained network control and visibility. We introduce CSIG, a protocol that delivers precise, multi-bit bottleneck congestion signals via a fixed-length Ethernet header. The architecture captures μsgranularity switch metrics, such as available bandwidth, and signals them to end-hosts using in-band, line-rate operations. We propose Fast Ramp-Up, a congestion control primitive that leverages these bottleneck signals to reduce median RPC latency by 20% and unclaimed bandwidth by 60% in production. Beyond transport-level performance, CSIG enables flow-aware observability by embedding μs-scale metrics into every packet, allowing individual application transfers to pinpoint their bottleneck location, such as the topology tier limiting their performance. CSIG thus transforms network telemetry from post-hoc correlation into a real time, context-aware capability. We demonstrate CSIG's broad deployability by validating it across five generations of commodity switch hardware (up to 102.4 Tbps), four NIC generations, and five transport stacks. Our design proves that a streamlined Layer 2 approach, focusing exclusively on the principal path bottleneck, provides transport-agnostic gains without requiring forklift hardware upgrades.

We present the design of Espresso, Google's SDN-based Internet peering edge routing infrastructure. This architecture grew out of a need to exponentially scale the Internet edge cost-effectively and to enable application-aware routing at Internet-peering scale. Espresso utilizes commodity switches and host-based routing/packet processing to implement a novel fine-grained traffic engineering capability. Overall, Espresso provides Google a scalable peering edge that is programmable, reliable, and integrated with global traffic systems. Espresso also greatly accelerated deployment of new networking features at our peering edge. Espresso has been in production for two years and serves over 22% of Google's total traffic to the Internet.

Now that our smartphones have multiple interfaces (WiFi, 3G, 4G, etc.), we have preferences for which interfaces an application may use. We may prefer to stream video over WiFi because it is fast, but VoIP over 3G because it gives continued connectivity. We also have relative preferences, such as giving Netflix twice as much capacity as Dropbox. This means our mobile devices need to schedule packets in keeping with our preferences while making use of all the capacity available. This is the natural domain of fair queuing, and this paper is about the design of a packet scheduler to meet these requirements. We show that traditional fair queueing schedulers cannot take into account a user's preferences for some interfaces over others. We present a novel packet scheduler called miDRR that meets our needs by generalizing DRR for multiple interfaces. We demonstrate a prototype running in Linux and show that it works correctly and can easily run at the speeds we need.

Policy-makers, ISPs and content providers are locked in a debate about who can control the Internet traffic that flows into our homes. In this paper we argue that the user, not the ISP or the content provider, should decide how traffic is prioritized to and from the home. Home users know most about their preferences, and if they can express them well to the ISP, then both the ISP and user are better off. To test the idea we built a prototype that lets users express highlevel preferences that are translated to low-level semantics and used to control the network.

Body Area Networks (BANs) can perform the task of continuous, remote monitoring of a patient's physiological signals in diverse environments. Apart from providing healthcare professionals with extensive logs of a patient's physiological history, BANs can be used to identify and react to emergency situations. We identify three important factors that afflict wireless communication in BANs: impermeability of the human body to radio waves at frequencies commonly used in BANs, efficient operation in mobile and time-varying environments, and mission-critical requirements for quick response to emergencies. An understanding of the link layer behavior of wireless sensor nodes placed on the body is crucial to address these and other challenges such as reducing energy consumption and increasing network lifetime.

As body sensor networks (BSNs) advance to fulfill the promise of continuous, non-intrusive, remote monitoring of patients, it is important that we achieve efficient communication between energy constrained on-body sensors. An important design choice which has significant impact on achieving this goal is the network architecture. Star architecture has been the natural choice for BSNs due to the short distances between the nodes. In this paper, we revisit this choice by quantitatively studying architecture choices using data from experiments conducted by deploying nodes operating at the 2.4 GHz band on actual human volunteers. We compare the star and multihop architectures to highlight their respective performance characteristics. In particular, we use our data to construct multihop networks with routes that maximize end-to-end packet delivery ratio (PDR) and routes that minimize the average number of retransmissions. Since BSNs span an entire spectrum of applications, each with its unique constraints and requirements, there is no solution that is optimal for all applications. Instead, we show the performance across a variety of metrics and the trade-offs that are achievable. We see that a multihop network minimizing retransmissions has several advantages including having better network lifetime as well as the lowest delay and energy consumption.

We propose MAX, a system that facilitates human-centric search of the physical world. Instead of organizing objects a priori, it allows humans to search for and locate them as needed. Designed for the following objectives: (i) human-centric operation, (ii) privacy, and (iii) efficient searching of any tagged object, MAX provides location information in a form natural to humans, that is, with reference to identifiable landmarks (such as, “on the dining table”) rather than precise coordinates. In the system, all physical objects—from documents to clothing—can be tagged, users then locate objects using an intuitive search interface. To make searching efficient, MAX adopts a hierarchical architecture consisting of tags (bound to objects), substations (bound to landmarks), and base-stations (bound to localities). Tags can be marked as either public or private, with private tags searchable only by the owner. MAX also provides for privacy of physical spaces. It requires minimal initial configuration, and is robust to reconfiguration of the physical space. We also present a methodology to design energy-optimal and delay-optimal query protocols for a variety of device choices, this optimizes system performance, and affords insight into the appropriate actions for various scenarios. We have implemented a simple prototype of MAX, demonstrating the feasibility of the system for human-centric search over several locations across a wide area. We contend that a MAX-like search system will enable sharing (e.g., books on a college campus) and trading (e.g., buying and selling used books) of physical resources, and will be the engine for a host of new applications.

10.1109/INFOCOM.2006.118

MAX is a system that facilitates human-centric search of the physical world. It allows humans to search for and locate objects as and when they need it instead of organizing them a priori. It provides location information in a form natural to humans, i.e., with reference to identifiable landmarks (e.g., on the dining table) rather than precise coordinates. MAX was designed with the following objectives: (i) human-centric operation, (ii) privacy, and (iii) efficient search of any tagged object. In the system, all physical objects, from documents to clothing, can be tagged and people locate objects using an intuitive search interface. To make search efficient, MAX adopts a hierarchical architecture consisting of tags (bound to objects), sub-stations (bound to landmarks) and base-stations (bound to localities). Tags can be marked as either public or private, with private tags searchable only by the owner. MAX also provides for privacy of physical spaces.MAX requires minimal initial configuration, and is robust to reconfiguration of the physical space. To optimize system performance, we present a methodology to design energy and delay optimal query protocols for a variety of device choices. We have implemented MAX using Crossbow motes and conducted user trials in a 5m by 5m cluttered office. The user feedback was positive, demonstrating the feasibility of MAX for human-centric search. We contend that a MAX-like search system will enable sharing (e.g., books on a college campus) and trading (e.g., buying and selling used books) of physical resources, and will be the engine for a host of new applications.