Enabling Robust and Efficient Distributed Computation in Dynamic Peer-to-Peer Networks

Augustine, John; Pandurangan, Gopal; Robinson, Peter; Roche, Scott; Upfal, Eli

doi:10.1109/focs.2015.29

Cited by 42 publications

(69 citation statements)

References 54 publications

(77 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Research has shown that expander topologies may provide the right framework for solving several different problems in P2P networks such as consensus [8], leader election [3], and even storage and search [4] for a single data item. The good news is that we now have ways to build and maintain overlays with good expansion properties despite heavy churn [7]. However, such expander topologies tend to be unstructured and prove to be inefficient for storing, searching, and retrieving data items, which is a key requirement for DHT applications.…”

Section: Introductionmentioning

confidence: 99%

Spartan: A Framework For Sparse Robust Addressable Networks

Augustine

Sivasubramaniam

2018

2018 IEEE International Parallel and Distributed Processing Symposium (IPDPS)

Self Cite

View full text Add to dashboard Cite

A Peer-to-Peer (P2P) network is a dynamic collection of nodes that connect with each other via virtual overlay links built upon an underlying network (usually, the Internet). Typical P2P networks, however, are highly dynamic and can experience very heavy churn, i.e., a large number of nodes join/leave the network every time step. Building and maintaining a stable overlay network despite such heavy churn is therefore an important problem that has been studied extensively for nearly two decades.We present an overlay design called Sparse Robust Addressable Network (Spartan) that can tolerate heavy adversarial churn. We show that Spartan can be built efficiently in a fully distributed manner within O(log n) rounds. Furthermore, the Spartan overlay structure can be maintained, again, in a fully distributed manner despite adversarially controlled churn (i.e., nodes joining and leaving) and significant variation in the number of nodes. When the number of nodes in the network lies in [n, f n] for any fixed f ≥ 1 the adversary can remove up to n nodes and add up to n nodes (for some small but fixed > 0) within any period of P rounds for some P ∈ O(log log n). Moreover, the adversary can add or remove nodes from the network at will and without any forewarning.Despite such uncertainty in the network, Spartan maintains Θ(n/ log n) committees that are stable and addressable collections of Θ(log n) nodes each. Any node that enters the network will be able to gain membership in one of these committees within O(1) rounds. The committees are also capable of performing sustained computation and passing messages between each other. Thus, any protocol designed for static networks can be simulated on Spartan with minimal overhead. This makes Spartan an ideal platform for developing applications. All our results hold with high probability. *

show abstract

Section: Introductionmentioning

confidence: 99%

Spartan: A Framework For Sparse Robust Addressable Networks

Augustine

Sivasubramaniam

2018

2018 IEEE International Parallel and Distributed Processing Symposium (IPDPS)

Self Cite

View full text Add to dashboard Cite

show abstract

“…(Our model is described in detail in Section 1.1, it is similar to the model considered in prior work, e.g., [6,4,5].) We consider a dynamic network as a sparse bounded degree expander graph whose topology -both nodes and edges -can change arbitrarily from round to round and is controlled by an adversary.…”

Section: Introductionmentioning

confidence: 99%

Fast Byzantine Leader Election in Dynamic Networks

Augustine

Pandurangan

Robinson

2015

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

Abstract. We study the fundamental Byzantine leader election problem in dynamic networks where the topology can change from round to round and nodes can also experience heavy churn (i.e., nodes can join and leave the network continuously over time). We assume the full information model where the Byzantine nodes have complete knowledge about the entire state of the network at every round (including random choices made by all the nodes), have unbounded computational power and can deviate arbitrarily from the protocol. The churn is controlled by an adversary that has complete knowledge and control over which nodes join and leave and at what times and also may rewire the topology in every round and has unlimited computational power, but is oblivious to the random choices made by the algorithm. Our main contribution is an O(log 3 n) round algorithm that achieves Byzantine leader election under the presence of up to O(n 1/2−ε ) Byzantine nodes (for a small constant ε > 0) and a churn of up to O( √ n/ polylog(n)) nodes per round (where n is the stable network size). The algorithm elects a leader with probability at least 1 − n −Ω(1) and guarantees that it is an honest node with probability at least 1 − n −Ω(1) ; assuming the algorithm succeeds, the leader's identity will be known to a 1 − o(1) fraction of the honest nodes. Our algorithm is fully-distributed, lightweight, and is simple to implement. It is also scalable, as it runs in polylogarithmic (in n) time and requires nodes to send and receive messages of only polylogarithmic size per round. To the best of our knowledge, our algorithm is the first scalable solution for Byzantine leader election in a dynamic network with a high rate of churn; our protocol can also be used to solve Byzantine agreement in a straightforward way. We also show how to implement an (almost-everywhere) public coin with constant bias in a dynamic network with Byzantine nodes and provide a mechanism for enabling honest nodes to store information reliably in the network, which might be of independent interest.

show abstract

“…In P2P systems, peer dynamic refers to the ability of nodes to join, leave, or even fail without any delay or restriction imposed [16]. Thus, it is important for the DHT to have a simplified joining procedure in order to maintain its resilience and scalability when multiple nodes attempt to join.…”

Section: Node Joinmentioning

confidence: 99%

“…The latency between nodes within the same class is drawn uniformly at random from [ 5,15] intervals. Similarly, the interclass latency is selected from the space [ 16,45]. The scenario here is to send queries from source nodes to their most distant nodes in the overlay and measure the latency these queries accumulate in a heterogeneous and homogeneous addressing schemes.…”

Section: Homogeneous Addressingmentioning

confidence: 99%

“…To examine this venue, we run the simulation on fully occupied 2 16 -node 3-d SHAM and Chord overlays and measure the number of failed queries in each system. The scenario of this second experiment is to fail nodes at different capacities in both systems by increasing the departure rate, then monitor the percentage of queries that encounter failed nodes through their paths to destination; see Table 4.…”

Section: Failed Lookupsmentioning

confidence: 99%

See 1 more Smart Citation

SHAM: Scalable Homogeneous Addressing Mechanism for structured P2P networks

Zghaibeh

Hassan

2017

J Wireless Com Network

View full text Add to dashboard Cite

In designing structured P2P networks, scalability, resilience, and load balancing are features that are needed to be handled meticulously. The P2P overlay has to handle large scale of nodes while maintaining minimized path lengths in performing lookups. It has also to be resilient to nodes' failure and be able to distribute the load uniformly over its participant. In this paper, we introduce SHAM: a Scalable, Homogenous, Addressing Mechanism for structured P2P networks. SHAM is a multi-dimensional overlay that places nodes in the network based on geometric addressing and maps keys onto values using consistent hashing. Our simulation results show that SHAM locates keys in the network efficiently, is highly resilient to major nodes' failure, and has an effective load balancing property. Furthermore, unlike other DHTs and due to its distinguished naming scheme, SHAM deploys homogenous addressing which drastically reduces latency in the underlying network.

show abstract

Enabling Robust and Efficient Distributed Computation in Dynamic Peer-to-Peer Networks

Cited by 42 publications

References 54 publications

Spartan: A Framework For Sparse Robust Addressable Networks

Spartan: A Framework For Sparse Robust Addressable Networks

Fast Byzantine Leader Election in Dynamic Networks

SHAM: Scalable Homogeneous Addressing Mechanism for structured P2P networks

Contact Info

Product

Resources

About