In modern blockchains, efficient, fair, and fault-tolerant information dissemination is critical for performance and security. Several stages of the transaction lifecycle are affected, from the creation and dissemination of transactions to the dissemination of blocks in the consensus layer. Mempool protocols, such as LØ, already address some of modern blockchains’ security threats. However, others remain unless fairness is embedded most fundamentally in the dissemination layer used by these protocols to share transactions and blocks and to reconcile mempools. This paper introduces HERMES, a novel dissemination protocol for mempools that leverages robust minimal structures to optimize data propagation, balance load, and ensure fairness despite faults and Byzantine actors. Specifically, we address front-running attacks by randomizing the choice of an overlay structure while forcing nodes to prove adherence to the mempools’ dissemination policies and the random choice made. Experimental results show significant performance improvements compared to traditional broadcast protocols for both permissioned and permissionless blockchains.

Geo-replication provides disaster recovery after catastrophic accidental failures or attacks, such as fires, blackouts or denial-of-service attacks to a data center or region. Naturally distributed data structures, such as Blockchains, when well designed, are immune against such disruptions, but they also benefit from leveraging locality. In this work, we consolidate the performance of geo-replicated consensus by leveraging novel insights about hierarchical consensus and a construction methodology that allows creating novel protocols from existing building blocks. In particular we show that cluster confirmation, paired with subgroup rotation, allows protocols to safely operate through situations where all members of the global consensus group are Byzantine. We demonstrate our compositional construction by combining the recent HotStuff and Damysus protocols into a hierarchical geo-replicated blockchain with global durability guarantees. We present a compositionality proof and demonstrate the correctness of our protocol, including its ability to tolerate cluster crashes. Our protocol — Orion1— achieves a 20% higher throughput than GeoBFT, the latest hierarchical Byzantine Fault-Tolerant (BFT) protocol.

Peer sampling is a crucial primitive in distributed systems, used to manage overlays and disseminate information in large-scale scenarios such as permissionless blockchain systems. Its purpose is to maintain and regularly update a local and partial snapshot, or view, of the complete system’s membership. These protocols are often targeted by malicious actors who aim to disrupt higher-level protocols. Typically, an adversary who controls a set of Byzantine nodes attempts to manipulate how legitimate nodes perceive the presence of Byzantine ones by increasing their representation in the view of honest nodes. While state-of-the-art Byzantine-tolerant peer sampling protocols mitigate this bias, their effectiveness decreases significantly as the number of malicious nodes increases. This paper introduces AUPE, the first collaborative Byzantine-tolerant peer sampling protocol that leverages the presence of trusted nodes, such as Intel’s SGX capable devices, to collaboratively track the spread of identifiers in the system and locally debias the representation of Byzantine nodes. Simulations with $\mathbf{1 0, 0 0 0}$ nodes demonstrate that AUPE outperforms state-of-the-art solutions, achieving nearperfect resilience even when faced with an adversary controlling $\mathbf{2 6 \%}$ of the nodes. Overall, by including as few as $\mathbf{1 0 \%}$ of trusted nodes, AUPE increases the tolerance of Brahms by up to $\mathbf{6 0 \%}$ while limiting the impact of the adversary’s attack, even when possessing up to 40% of the nodes.

Peer sampling is a first-class abstraction used in distributed systems for overlay management and information dissemination. The goal of peer sampling is to continuously build and refresh a partial and local view of the full membership of a dynamic, large-scale distributed system. Malicious nodes under the control of an adversary may aim at being over-represented in the views of correct nodes, increasing their impact on the proper operation of protocols built over peer sampling. State-of-the-art Byzantine resilient peer sampling protocols reduce this bias as long as Byzantines are not overly present. This paper studies the benefits brought to the resilience of peer sampling services when considering that a small portion of trusted nodes can run code whose authenticity and integrity can be assessed within a trusted execution environment, and specifically Intel’s software guard extensions technology (SGX). We present RAPTEE, a protocol that builds and leverages trusted gossip-based communications to hamper an adversary’s ability to increase its system-wide representation in the views of all nodes. We apply RAPTEE to BRAHMS, the most resilient peer sampling protocol to date. Experiments with 10,000 nodes show that with only 1% of SGX-capable devices, RAPTEE can reduce the proportion of identifiers of Byzantine nodes in the view of honest ones by up to 17%, when the system contains 10% of Byzantine nodes. In addition, the security guarantees of RAPTEE hold even in the presence of a powerful attacker attempting to identify trusted nodes and injecting view-poisoned trusted nodes.

Gossip-based dissemination protocols are important building blocks of large-scale distributed systems as they may impact both the systems’ efficiency and fault tolerance. There exist many flavors of gossip dissemination protocols. IDA-Gossip is one of the gossip dissemination protocols proposed in the context of blockchains to efficiently disseminate large messages. It relies on multi-chunk gossip dissemination, erasure coding, and Merkle hash trees. However, despite its claimed efficiency, there is no in-depth analysis of this protocol to understand its behavior under different conditions (e.g., with injected faults). In this work, we evaluate the behavior of IDA-Gossip by relying on extensive experiments and simulations. Specifically, we evaluate IDA-Gossip both in terms of performance and resilience to faults by varying its configuration parameters and the number of faulty nodes, respectively. This study results in several takeaways. First, IDA-Gossip provides excellent dissemination latency compared to classic gossip. Second, it provides excellent coverage even with 40 percent of faulty nodes in the system. Finally, the use of erasure coding provides an important advantage to IDA-Gossip compared to classic multi-chunk gossip dissemination protocols.

Most of today’s online services (e.g., social networks, search engines, marketplace places) are centralized, which most users recognize as unsatisfactory for various reasons (e.g., centralized governance, censorship, loss of control over personal data). Blockchain technologies promise a new Web revolution (Web 3.0) through the decentralization of online services. However, one of the fundamental limitations for this revolution to happen at a planetary scale is the poor performance of today’s permissionless blockchains. In this paper, we propose ALDER, a generic construction that multiplexes off-the-shelf permissionless blockchain protocols to address the performance bottleneck due to store-validate-forward block dissemination techniques in blockchain protocols. We apply ALDER to two representative blockchains, namely Algorand (Proof-of-Stake) and Bitcoin (Proof-of-Work), to illustrate the benefits it brings to blockchain performance. Our evaluations show that ALDER can drastically improve the throughput of blockchains when bottlenecks exist.

Delivering video content with a high and fairly shared quality of experience is a challenging task in view of the drastic video traffic increase forecasts. Currently, content delivery networks provide numerous servers hosting replicas of the video content, and consuming clients are re-directed to the closest server. Then, the video content is streamed using adaptive streaming solutions. However, some servers become overloaded, and clients may experience a poor or unfairly distributed quality of experience.

Single-source HTTP adaptive streaming solutions (HAS) have become the de facto solutions to deliver live video over the Internet. By avoiding video stalling events that are mainly caused by the lack of throughput at client or at server side, HAS solutions increase the end users’ quality of experience (QoE). We propose to pragmatically extend HAS with our MS-Stream solution that simultaneously utilizes several servers. MS-Stream aims at offering high QoE for live content delivery by exploiting expanded bandwidth and link diversity in distributed heterogeneous infrastructures. By leveraging end users’ connectivity capacities, we further extend the QoE and scalability capabilities of our proposal by exposing a hybrid P2P/multisource live-streaming solution (P2P/MS-Stream (PMS)), achieving trade-offs between the system’s scale and the end users’ QoE. We propose a distributed quality adaptation algorithm run by every peer, along with a local optimization method of the usage of the server infrastructure made available. Large-scale evaluations conducted with 300 peers located in France permits validating our approach and algorithms over flash crowd events and allow us to conclude that PMS can reach the optimal trade-offs between QoE and system scale.

Adaptive bitrate streaming protocols, such as DASH, have seen extensive interests for their adaptation capabilities to increase consumers' Quality of Experience (QoE) over the Internet, and have become de-facto standards in web video delivery. Compared with traditional single-server approaches, multiple-server streaming offers the opportunity to exploit expanded bandwidth, link diversity, and reliability. In this paper, we expose our solution for multiple-server support to dynamic adaptive streaming applications: Multiple-Source Streaming (MS-Stream). Thanks to its codec agnosticism and DASH-compliance our contribution is a pragmatic and evolving solution for QoE enhancement that can be applied to many streaming architectures (CDNs, Clouds) and is particularly suited for distributed environments such as P2P or Set-Top-Box overlays. In addition, splitting content into multiple independent sub-streams provides the opportunity to achieve easy-to-design bitrate adaptation and server-switching mechanisms. We empirically validate our approach using an extensive collection of network profiles provided by the DASH Industry Forum. Our solution is compared with the full potential of DASH with several servers over several QoE criteria. Results show the QoE gain of using MS-Stream against DASH; an online demonstration is made available.

HTTP Adaptive Streaming (HAS) protocols have become the de-facto solutions to deliver video over the Internet, mainly due to their ability to limit video freezing and thus enhance consumers' Quality of Experience (QoE). Nevertheless, they do not have the possibility to improve the actual delivered video quality, limited by the available throughput between the delivering server and the chent. Compared with this single-server approach, multiple-server streaming offers the opportunity to obtain enhanced QoE by benefiting from expanded bandwidth, link diversity and rehability in distributed streaming infrastructures. We present a prototype for a pragmatic evolving HAS-compliant streaming solution, simultaneously using several servers.

Single-source HTTP Adaptive Streaming protocols (HAS), such as MPEG-DASH, have become the de-facto solutions to deliver video over the Internet. By avoiding buffer stalling events that are mainly caused by the lack of throughput at client or at server side, HAS protocols increase end-user's Quality of Experience (QoE). We propose to extend HAS capabilities to a pragmatic DASH-compliant Multiple-Source Streaming solution (MS-Stream) that simultaneously utilizes several servers. MS-Stream offers the opportunity to obtain higher QoE by exploiting expanded bandwidth and link diversity in heterogeneous distributed streaming infrastructures, such as distributed home-gateways or geographically distributed set-top-boxes belonging to Over-The-Top video service providers. This paper exposes a cost-effective two-phase adaptation process with dual (i.e., bitrate and number of sources) adaptation decisioning prior segment request and in-segment download adaptation. Our approach was empirically evaluated for on-demand video streaming over the Internet. An online demonstration is also made available [1].

HTTP Adaptive Streaming have become the de-facto solutions to deliver video over the Internet due to their ability to enhance consumers' Quality of Experience (QoE). Nevertheless, they do not have the possibility to improve the actual delivered video quality, limited by the available client-server throughput. In comparison, multiple-server and P2P streaming offer the opportunity to obtain enhanced QoE by benefiting from expanded bandwidth, link diversity and reliability in distributed streaming infrastructures. We present a prototype for a hybrid P2P/multi-server quality-adaptive streaming solution, simultaneously using several servers and peers, and trading off the server infrastructure capacities and QoE gains.

Over the past few years, adaptive bitrate streaming protocols, such as DASH, have risen to enhance the End-Users' experience towards video consumption over the Internet. Current methods perform a smoother playback (i.e., fewer re-buffering states) trading off with quality fluctuations due to the client-server link state. The proposed work goes a step further and introduces an innovative lightweight streaming solution by taking advantage of bandwidth aggregation over multiple paths using multiple content sources at the same time. This pragmatic evolving approach outperforms the QoE delivered by current DASH-based or P2P-based solutions and leverages the scope of action made available to the media delivery chain actors. The proposed solution has been implemented and is compared to a DASH-based approach used in most existing VoD use-cases through a first set of experiments. Results demonstrate the strong advantages in terms of quality delivered at the End-User's side and buffer occupancy. Discussion on the necessary trade-offs is also tackled.