Smart environments (homes, factories, hospitals, buildings) contain an increasing number of IoT devices, making them complex to manage. Today, in smart homes when users or triggers initiate routines (i.e., a sequence of commands), concurrent routines and device failures can cause incongruent outcomes. We describe SafeHome, a system that provides notions of atomicity and serial equivalence for smart homes. Due to the human-facing nature of smart homes, SafeHome offers a spectrum of visibility models which trade off between responsiveness vs. isolation of the smart home. We implemented SafeHome and performed workload-driven experiments. We find that a weak visibility model, called eventual visibility, is almost as fast as today's status quo (up to 23% slower) and yet guarantees serially-equivalent end states.

Recently, a new class of "arbitrator-based" membership protocols have been proposed. These claim to provide time bounds on how long membership lists can stay inconsistent-this property is critical in many distributed applications which need to take timely recovery actions. In this paper, we: 1) present the first fully decentralized and stabilizing version of membership protocols in this class; 2) formally prove properties and claims about both our decentralized version and the original protocol; and 3) present experimental results from both a simulation and a real cluster implementation.

We describe Service Fabric (SF), Microsoft's distributed platform for building, running, and maintaining microservice applications in the cloud. SF has been running in production for 10+ years, powering many critical services at Microsoft. This paper outlines key design philosophies in SF. We then adopt a bottom-up approach to describe low-level components in its architecture, focusing on modular use and support for strong semantics like fault-tolerance and consistency within each component of SF. We discuss lessons learned, and present experimental results from production data.

Bandwidth is a shared and limited resource of the wireless LAN. The MAC protocols are mainly responsible for efficiently sharing it among all the stations. The IEEE 802.11 standard uses Distributed Coordination Function (DCF) as its default MAC protocol which employs Carrier Sense Multiple Access using Collision Avoidance (CSMA/CA) scheme with binary exponential backoff algorithm. Despite its ubiquitous acceptance, it has two major drawbacks: i) channel idle time and ii) collision overhead. Besides, this protocol uses explicit ack which comes with significant overheads, for example preamble, frame header etc. In this paper, we propose a scheme called O-ACK which reduces the overhead of explicit ack and backoff interval by leveraging piggybacking, packet overhearing and token based scheduling. Our NS2 based simulation results confirm that this protocol significantly outperforms the DCF protocol.

In wireless LANs, MAC protocols perform one of the most important tasks: efficiently sharing the limited bandwidth among the contending stations. In IEEE 802.11, Distributed Coordination Function (DCF) is used as a default MAC protocol. Despite its wide acceptance, it has two major downsides: i) Channel idle time and ii) Collision overheads. Besides, the use of the explicit ack frames incurs extra overhead due to the preamble, packet headers etc. This paper presents an optimized version of the Overheard-ACK (O-ACK) protocol (presented in [2]) which incorporates piggybacking, packet overhearing and token based scheduling to significantly reduce the overheads. The previous version of O-ACK creates chains of transmissions, where consecutive transmissions are separated by SIFS interval, which increases the throughput. But on the downside, as two adjacent transmissions are separated by only SIFS idle time, a newly arrived station cannot start transmission. Our current version of O-ACK ensures higher throughput and fairness in multiple AP scenario by continuously adapting the chain creation process based on the surrounding environment. Our NS-2 based simulation results confirm that this protocol significantly outperforms the DCF protocol.

Distributed Coordination Function (DCF) is defined in IEEE 802.11 standard, which is widely used in practice. Despite of its wide use, it has several limitations. Because of the idle and collision times, it suffers from poor channel utilization. Besides, the control packets, particularly, Acknowledgement (ACK), consume non-trivial amount of bandwidth. Though the number of control bits in an ACK frame is small, the added overheads like the preamble, packet header etc. make the situation worse. In this paper, we propose a scheme called Overheard ACK where the explicit ACK frame has been eliminated by using the leverage of packet overhearing. Also, by incorporating explicit and implicit token-passing, this protocol attempts to schedule transmissions without having to use random access, dramatically reducing the idle time and collision time. Simulation results using NS2 confirm that this protocol significantly outperforms the conventional 802.11 DCF.