arXiv subjects

Peter Robinson

Publications and source records attributed to Peter Robinson.

At least 19 recordsLinked to original sources

arXiv2026

Polynomial Lower Bounds for Distributed Graph Sketching with Tiny Error: Connectivity and Spanning Tree Construction

We present the first polynomial lower bounds for several fundamental problems in the distributed graph sketching model in the tiny-error regime, which includes deterministic algorithms as a special case. In the graph sketching model, every node sends a single message to the referee who does not have any prior knowledge of the graph and must output the answer. While the work of Nelson and Yu (SODA 2019) and Yu (SODA 2021) showed that $\Theta( \log^3n )$ is optimal for constructing a spanning forest or deciding whether the graph is connected with error at most $\frac{1}{\text{poly}(n)}$ , their approach does not yield any stronger bounds for significantly smaller error probabilities. Our main result is to show that solving either connectivity or spanning tree construction with error at most $\delta$ requires messages of length $\Omega( \min\{n, \log_2 \frac{1}{\delta}\}^{1/3} )$, which implies that algorithms with exponentially small error must send messages of $\Omega( n^{1/3} )$ bits in the worst case. Our results significantly narrows the current gap between the Jelani-Yu threshold of $\Theta( \log^3n )$ and the trivial upper bound of sending $O(n)$ bits per node for deterministic graph sketching. We also extend our results to $k$-edge connectivity. For any $k=O(n^{1/7})$, we recover the same bound of $\Omega( k )$ on the message length for algorithms with exponentially small error that was shown by Robinson and Tan (PODS 2026) only for deterministic algorithms. Finally, for $k=n^{o(1)}$, our result implies a stronger lower bound of $\Omega_\epsilon( n^{\epsilon} )$ bits, for any constant $\epsilon<\tfrac{1}{3}$.

cs.DS↗

arXiv2026

Tight Energy Lower Bounds for Distributed Graph Algorithms

There has been a significant recent interest in designing distributed algorithms in the SLEEPING model that minimize the {energy (a.k.a awake) complexity, which measures the number of rounds a node is awake during the algorithm. A node spends non-trivial resources (messages, energy, etc.) only when it is awake and not while sleeping. Energy complexity has been studied for various fundamental problems with respect to minimizing the maximum (worst-case) or the average number of rounds a node is awake. It has been shown that the energy complexities of several fundamental problems such as leader election (LE), broadcast, Minimum Spanning Tree (MST), Maximal Independent Set (MIS) is exponentially smaller compared to their respective best-possible round complexities in the standard CONGEST model (where nodes can only send messages of small size). This raises a fundamental question of whether such significant energy gains are possible for many other fundamental problems. Our main contribution is a general and powerful technique for showing energy lower bounds using information theory. It gives almost a "plug-in" way to show energy lower bounds for various problems in the standard CONGEST model. Our information-theoretic technique allows us to leverage known lower bounds on communication complexity to obtain new, almost optimal (up to logarithmic factors) polynomial (in $n$) lower bounds on energy complexity --- for both worst-case and average-case --- for fundamental graph problems such as triangle enumeration, All-Pairs Shortest Paths (APSP), diameter computation, minimum weight cycle, Maximum Independent Set (MaxIS), Minimum Dominating Set (MinDS), Minimum Vertex Cover (MinVC). The energy lower bounds of these problems match their respective round lower bounds, implying that one cannot obtain any significant gains in energy complexity.

cs.DC↗

arXiv2026

The Quantum Message Complexity of Distributed Wake-Up with Advice

We consider the distributed wake-up problem with advice, where nodes are equipped with initial knowledge about the network at large. After the adversary awakens a subset of nodes, an oracle computes a bit string (``the advice'') for each node, and the goal is to wake up all sleeping nodes efficiently. We present the first upper and lower bounds on the message complexity for wake-up in the quantum routing model, introduced by Dufoulon, Magniez, and Pandurangan (PODC 2025). In more detail, we give a distributed advising scheme that, given $\alpha$ bits of advice per node, wakes up all nodes with a message complexity of $O( \sqrt{\frac{n^3}{2^{\max\{\lfloor (\alpha-1)/2 \rfloor},0\}}}\cdot\log n )$ with high probability. Our result breaks the $\Omega( \frac{n^2}{2^\alpha} )$ barrier known for the classical port numbering model in sufficiently dense graphs. To complement our algorithm, we give a lower bound on the message complexity for distributed quantum algorithms: By leveraging a lower bound result for the single-bit descriptor problem in the query complexity model, we show that wake-up has a quantum message complexity of $\Omega( n^{3/2} )$ without advice, which holds independently of how much time we allow. In the setting where an adversary decides which nodes start the algorithm, most graph problems of interest implicitly require solving wake-up, and thus the same lower bound also holds for other fundamental problems such as single-source broadcast and spanning tree construction.

quant-ph↗

arXiv2025

Deterministic Lower Bounds for $k$-Edge Connectivity in the Distributed Sketching Model

We study the $k$-edge connectivity problem on undirected graphs in the distributed sketching model, where we have $n$ nodes and a referee. Each node sends a single message to the referee based on its 1-hop neighborhood in the graph, and the referee must decide whether the graph is $k$-edge connected by taking into account the received messages. We present the first lower bound for deciding a graph connectivity problem in this model with a deterministic algorithm. Concretely, we show that the worst case message length is $\Omega( k )$ bits for $k$-edge connectivity, for any super-constant $k = O(\sqrt{n})$. Previously, only a lower bound of $\Omega( \log^3 n )$ bits was known for ($1$-edge) connectivity, due to Yu (SODA 2021). In fact, our result is the first super-polylogarithmic lower bound for a connectivity decision problem in the distributed graph sketching model. To obtain our result, we introduce a new lower bound graph construction, as well as a new 3-party communication complexity problem that we call UniqueOverlap. As this problem does not appear to be amenable to reductions to existing hard problems such as set disjointness or indexing due to correlations between the inputs of the three players, we leverage results from cross-intersecting set families to prove the hardness of UniqueOverlap for deterministic algorithms. Finally, we obtain the sought lower bound for deciding $k$-edge connectivity via a novel simulation argument that, in contrast to previous works, does not introduce any probability of error and thus works for deterministic algorithms.

cs.DS↗

arXiv2025

Time-Optimal and Energy-Efficient Deterministic Consensus

We study fault-tolerant consensus in a variant of the synchronous message passing model, where, in each round, every node can choose to be awake or asleep. This is known as the sleeping model (Chatterjee, Gmyr, Pandurangan PODC 2020) and defines the awake complexity (also called \emph{energy complexity}), which measures the maximum number of rounds that any node is awake throughout the execution. Only awake nodes can send and receive messages in a given round and all messages sent to sleeping nodes are lost. We present new deterministic consensus algorithms that tolerate up to $f<n$ crash failures, where $n$ is the number of nodes. Our algorithms match the optimal time complexity lower bound of $f+1$ rounds. For multi-value consensus, where the input values are chosen from some possibly large set, we achieve an energy complexity of ${O}(\lceil f^2 / n \rceil)$ rounds, whereas for binary consensus, we show that ${O}(\lceil f / \sqrt{n} \rceil)$ rounds are possible.

cs.DC↗

arXiv2025

Perfect Matching with Few Link Activations

We consider the problem of computing a perfect matching problem in a synchronous distributed network, where the network topology corresponds to a complete bipartite graph. The communication between nodes is restricted to activating communication links, which means that instead of sending messages containing a number of bits, each node can only send a pulse over some of its incident links in each round. In the port numbering model, where nodes are unaware of their neighbor's IDs, we give a randomized algorithm that terminates in $O( \log n )$ rounds and has a pulse complexity of $O( n\log n )$, which corresponds to the number of pulses sent over all links. We also show that randomness is crucial in the port numbering model, as any deterministic algorithm must send at least $\Omega( n^2 )$ messages in the standard LOCAL model, where the messages can be of unbounded size. Then, we turn our attention to the KT_1 assumption, where each node starts out knowing its neighbors' IDs. We show that this additional knowledge enables significantly improved bounds even for deterministic algorithms. First, we give an $O( \log n )$ time deterministic algorithm that sends only $O( n )$ pulses. Finally, we apply this algorithm recursively to obtain an exponential reduction in the time complexity to $O( \log^*n\log\log n )$, while slightly increasing the pulse complexity to $O( n\log^*n )$. All our bounds also hold in the standard CONGEST model with single-bit messages.

cs.DC↗

arXiv2025

Message Optimality and Message-Time Trade-offs for APSP and Beyond

Round complexity is an extensively studied metric of distributed algorithms. In contrast, our knowledge of the \emph{message complexity} of distributed computing problems and its relationship (if any) with round complexity is still quite limited. To illustrate, for many fundamental distributed graph optimization problems such as (exact) diameter computation, All-Pairs Shortest Paths (APSP), Maximum Matching etc., while (near) round-optimal algorithms are known, message-optimal algorithms are hitherto unknown. More importantly, the existing round-optimal algorithms are not message-optimal. This raises two important questions: (1) Can we design message-optimal algorithms for these problems? (2) Can we give message-time tradeoffs for these problems in case the message-optimal algorithms are not round-optimal? In this work, we focus on a fundamental graph optimization problem, \emph{All Pairs Shortest Path (APSP)}, whose message complexity is still unresolved. We present two main results in the CONGEST model: (1) We give a message-optimal (up to logarithmic factors) algorithm that solves weighted APSP, using $\tilde{O}(n^2)$ messages. This algorithm takes $\tilde{O}(n^2)$ rounds. (2) For any $0 \leq \varepsilon \le 1$, we show how to solve unweighted APSP in $\tilde{O}(n^{2-\varepsilon })$ rounds and $\tilde{O}(n^{2+\varepsilon })$ messages. At one end of this smooth trade-off, we obtain a (nearly) message-optimal algorithm using $\tilde{O}(n^2)$ messages (for $\varepsilon = 0$), whereas at the other end we get a (nearly) round-optimal algorithm using $\tilde{O}(n)$ rounds (for $\varepsilon = 1$). This is the first such message-time trade-off result known.

cs.DC↗

arXiv2025

Dynamic Approximate Maximum Matching in the Distributed Vertex Partition Model

We initiate the study of approximate maximum matching in the vertex partition model, for graphs subject to dynamic changes. We assume that the $n$ vertices of the graph are partitioned among $k$ players, who execute a distributed algorithm and communicate via message passing. An adaptive adversary may perform dynamic updates to the graph topology by inserting or removing edges between the nodes, and the algorithm needs to respond to these changes by adapting the output of the players, with the goal of maintaining an approximate maximum matching. The main performance metric in this setting is the algorithm's update time, which corresponds to the number of rounds required for updating the solution upon an adversarial change. For the standard setting of single-edge insertions and deletions, we give a randomized Las Vegas algorithm with an expected update time of $O( \lceil \frac{\sqrt{m}}{\beta k} \rceil )$ rounds that maintains a $\frac{2}{3}$-approximate maximum matching that is also maximal, where $m$ is the number of edges in the graph and $\beta$ is the available link bandwidth. For batch-dynamic updates, where the adversary may insert up to $\ell\ge 1$ edges at once, we prove the following. There is a randomized algorithm that succeeds with high probability in maintaining a $\frac{2}{3}$-approximate maximum matching and has a worst case update time of $O(\lceil\frac{\ell\log n}{\sqrt{\beta k}}\rceil )$ rounds. Any algorithm for maintaining a maximal matching without 3-augmenting paths under batches of $\ell$-edge insertions has an update time of $\Omega( \frac{\ell}{\beta k \log n} )$ rounds in the worst case.

cs.DC↗

arXiv2024

The Singular Optimality of Distributed Computation in LOCAL

It has been shown that one can design distributed algorithms that are (nearly) singularly optimal, meaning they simultaneously achieve optimal time and message complexity (within polylogarithmic factors), for several fundamental global problems such as broadcast, leader election, and spanning tree construction, under the $\text{KT}_0$ assumption. With this assumption, nodes have initial knowledge only of themselves, not their neighbors. In this case the time and message lower bounds are $\Omega(D)$ and $\Omega(m)$, respectively, where $D$ is the diameter of the network and $m$ is the number of edges, and there exist (even) deterministic algorithms that simultaneously match these bounds. On the other hand, under the $\text{KT}_1$ assumption, whereby each node has initial knowledge of itself and the identifiers of its neighbors, the situation is not clear. For the $\text{KT}_1$ CONGEST model (where messages are of small size), King, Kutten, and Thorup (KKT) showed that one can solve several fundamental global problems (with the notable exception of BFS tree construction) such as broadcast, leader election, and spanning tree construction with $\tilde{O}(n)$ message complexity ($n$ is the network size), which can be significantly smaller than $m$. Randomization is crucial in obtaining this result. While the message complexity of the KKT result is near-optimal, its time complexity is $\tilde{O}(n)$ rounds, which is far from the standard lower bound of $\Omega(D)$. In this paper, we show that in the $\text{KT}_1$ LOCAL model (where message sizes are not restricted), singular optimality is achievable. Our main result is that all global problems, including BFS tree construction, can be solved in $\tilde{O}(D)$ rounds and $\tilde{O}(n)$ messages, where both bounds are optimal up to polylogarithmic factors. Moreover, we show that this can be achieved deterministically.

cs.DC↗

arXiv2024

Rise and Shine Efficiently! Tight Bounds for Adversarial Wake-up

We study the wake-up problem in distributed networks, where an adversary awakens a subset of nodes at arbitrary times, and the goal is to wake up all other nodes as quickly as possible by sending only few messages. We prove the following lower bounds: * We first consider the setting where each node receives advice from an oracle who can observe the entire network, but does not know which nodes are awake initially. More specifically, we consider the $KT_0$ $LOCAL$ model with advice. We prove that any randomized algorithm must send $\Omega( \frac{n^{2}}{2^{\beta}\log n} )$ messages if nodes receive only $O(\beta)$ bits of advice on average. * For the $KT_1$ assumption, we show that any $(k+1)$-time algorithm requires $\Omega( n^{1+1/k} )$ messages. Our result is the first super-linear (in $n$) lower bound, for a problem that does not require individual nodes to learn a large amount of information about the network topology. To complement our lower bound results, we present several new algorithms: * We give an asynchronous $KT_1$ $LOCAL$ algorithm that solves the wake-up problem with a time and message complexity of $O( n\log n )$ with high probability. * We introduce the notion of \emph{awake distance} $\rho_{\text{awk}}$, which is upper-bounded by the network diameter, and present a synchronous $KT_1$ $LOCAL$ algorithm that takes $O( \rho_{\text{awk}} )$ rounds and sends $O( n^{3/2}\sqrt{\log n} )$ messages with high probability. We also extend these ideas to obtain a near-optimal time- and message complexity of $O\( \rho_{awk} \log^3n )$ rounds $O( n \log^3n )$ messages. * We give deterministic advising schemes in the asynchronous $KT_0$ $CONGEST$ model (with advice). In particular, we obtain an $O( \rho_{\text{awk}}\log^2n )$-time advising scheme that sends $O( n\log^2n )$ messages, while requiring $O( \log^2n )$ bits of advice per node.

cs.DC↗

arXiv2024

Dynamic Maximal Matching in Clique Networks

We consider the problem of computing a maximal matching with a distributed algorithm in the presence of batch-dynamic changes to the graph topology. We assume that a graph of $n$ nodes is vertex-partitioned among $k$ players that communicate via message passing. Our goal is to provide an efficient algorithm that quickly updates the matching even if an adversary determines batches of $\ell$ edge insertions or deletions. Assuming a link bandwidth of $O(\beta\log n)$ bits per round, for a parameter $\beta \ge 1$, we first show a lower bound of $\Omega( \frac{\ell\,\log k}{\beta\,k^2\log n})$ rounds for recomputing a matching assuming an oblivious adversary who is unaware of the initial (random) vertex partition as well as the current state of the players, and a stronger lower bound of $\Omega(\frac{\ell}{\beta\,k\log n})$ rounds against an adaptive adversary, who may choose any balanced (but not necessarily random) vertex partition initially and who knows the current state of the players. We also present a randomized algorithm that has an initialization time of $O( \lceil\frac{n}{\beta\,k}\rceil\log n )$ rounds, while achieving an update time that that is independent of $n$: In more detail, the update time is $O( \lceil \frac{\ell}{\beta\,k} \rceil \log(\beta\,k))$ against an oblivious adversary, who must fix all updates in advance. If we consider the stronger adaptive adversary, the update time becomes $O( \lceil \frac{\ell}{\sqrt{\beta\,k}}\rceil \log(\beta\,k))$ rounds.

cs.DC↗

arXiv2024

Tight Bounds on the Message Complexity of Distributed Tree Verification

We consider the message complexity of verifying whether a given subgraph of the communication network forms a tree with specific properties both in the KT-$\rho$ (nodes know their $\rho$-hop neighborhood, including node IDs) and the KT-$0$ (nodes do not have this knowledge) models. We develop a rather general framework that helps in establishing tight lower bounds for various tree verification problems. We also consider two different verification requirements: namely that every node detects in the case the input is incorrect, as well as the requirement that at least one node detects. The results are stronger than previous ones in the sense that we assume that each node knows the number $n$ of nodes in the graph (in some cases) or an $\alpha$ approximation of $n$ (in other cases). For spanning tree verification, we show that the message complexity inherently depends on the quality of the given approximation of $n$: We show a tight lower bound of $\Omega(n^2)$ for the case $\alpha \ge \sqrt{2}$ and a much better upper bound (i.e., $O(n \log n)$) when nodes are given a tighter approximation. On the other hand, our framework also yields an $\Omega(n^2)$ lower bound on the message complexity of verifying a minimum spanning tree (MST), which reveals a polynomial separation between ST verification and MST verification. This result holds for randomized algorithms with perfect knowledge of the network size, and even when just one node detects illegal inputs, thus improving over the work of Kor, Korman, and Peleg (2013). For verifying a $d$-approximate BFS tree, we show that the same lower bound holds even if nodes know $n$ exactly, however, the lower bound is sensitive to $d$, which is the stretch parameter.

cs.DC↗

arXiv2023

The Message Complexity of Distributed Graph Optimization

The message complexity of a distributed algorithm is the total number of messages sent by all nodes over the course of the algorithm. This paper studies the message complexity of distributed algorithms for fundamental graph optimization problems. We focus on four classical graph optimization problems: Maximum Matching (MaxM), Minimum Vertex Cover (MVC), Minimum Dominating Set (MDS), and Maximum Independent Set (MaxIS). In the sequential setting, these problems are representative of a wide spectrum of hardness of approximation. While there has been some progress in understanding the round complexity of distributed algorithms (for both exact and approximate versions) for these problems, much less is known about their message complexity and its relation with the quality of approximation. We almost fully quantify the message complexity of distributed graph optimization by showing the following results...[see paper for full abstract]

cs.DC↗

arXiv2023

Improved Tradeoffs for Leader Election

We consider leader election in clique networks, where $n$ nodes are connected by point-to-point communication links. For the synchronous clique under simultaneous wake-up, i.e., where all nodes start executing the algorithm in round $1$, we show a tradeoff between the number of messages and the amount of time. More specifically, we show that any deterministic algorithm with a message complexity of $n f(n)$ requires $\Omega\left(\frac{\log n}{\log f(n)+1}\right)$ rounds, for $f(n) = \Omega(\log n)$. Our result holds even if the node IDs are chosen from a relatively small set of size $\Theta(n\log n)$, as we are able to avoid using Ramsey's theorem. We also give an upper bound that improves over the previously-best tradeoff. Our second contribution for the synchronous clique under simultaneous wake-up is to show that $\Omega(n\log n)$ is in fact a lower bound on the message complexity that holds for any deterministic algorithm with a termination time $T(n)$. We complement this result by giving a simple deterministic algorithm that achieves leader election in sublinear time while sending only $o(n\log n)$ messages, if the ID space is of at most linear size. We also show that Las Vegas algorithms (that never fail) require $\Theta(n)$ messages. For the synchronous clique under adversarial wake-up, we show that $\Omega(n^{3/2})$ is a tight lower bound for randomized $2$-round algorithms. Finally, we turn our attention to the asynchronous clique: Assuming adversarial wake-up, we give a randomized algorithm that achieves a message complexity of $O(n^{1 + 1/k})$ and an asynchronous time complexity of $k+8$. For simultaneous wake-up, we translate the deterministic tradeoff algorithm of Afek and Gafni to the asynchronous model, thus partially answering an open problem they pose.

cs.DC↗

arXiv2022

What Can We Compute in a Single Round of the Congested Clique?

We show that any one-round algorithm that computes a minimum spanning tree (MST) in the unicast congested clique must use a link bandwidth of $\Omega(\log^3 n)$ bits in the worst case. Consequently, computing an MST under the standard assumption of $O(\log n)$-size messages requires at least $2$ rounds. This is the first round complexity lower bound in the unicast congested clique for a problem where the output size is small, i.e., $O(n\log n)$ bits. Our lower bound holds as long as every edge of the MST is output by an incident node. To the best of our knowledge, all prior lower bounds for the unicast congested clique either considered problems with large output sizes (e.g., triangle enumeration) or required every node to learn the entire output.

cs.DC↗

arXiv2022

Byzantine-Resilient Counting in Networks

We present two distributed algorithms for the {\em Byzantine counting problem}, which is concerned with estimating the size of a network in the presence of a large number of Byzantine nodes. In an $n$-node network ($n$ is unknown), our first algorithm, which is {\em deterministic}, finishes in $O(\log{n})$ rounds and is time-optimal. This algorithm can tolerate up to $O(n^{1 - \gamma})$ arbitrarily (adversarially) placed Byzantine nodes for any arbitrarily small (but fixed) positive constant $\gamma$. It outputs a (fixed) constant factor estimate of $\log{n}$ that would be known to all but $o(1)$ fraction of the good nodes. This algorithm works for \emph{any} bounded degree expander network. However, this algorithms assumes that good nodes can send arbitrarily large-sized messages in a round. Our second algorithm is {\em randomized} and most good nodes send only small-sized messages (Throughout this paper, a small-sized message is defined to be one that contains $O(\log{n})$ bits in addition to at most a constant number of node IDs.). This algorithm works in \emph{almost all} $d$-regular graphs. It tolerates up to $B(n) = n^{\frac{1}{2} - \xi}$ (note that $n$ and $B(n)$ are unknown to the algorithm) arbitrarily (adversarially) placed Byzantine nodes, where $\xi$ is any arbitrarily small (but fixed) positive constant. This algorithm takes $O(B(n)\log^2{n})$ rounds and outputs a (fixed) constant factor estimate of $\log{n}$ with probability at least $1 - o(1)$. The said estimate is known to most nodes, i.e., $\geq (1 - \beta)n$ nodes for any arbitrarily small (but fixed) positive constant $\beta$. To complement our algorithms, we also present an impossibility result that shows that it is impossible to estimate the network size with any reasonable approximation with any non-trivial probability of success if the network does not have sufficient vertex expansion.

cs.DC↗

arXiv2021

Can We Break Symmetry with o(m) Communication?

We study the communication cost (or message complexity) of fundamental distributed symmetry breaking problems, namely, coloring and MIS. While significant progress has been made in understanding and improving the running time of such problems, much less is known about the message complexity of these problems. In fact, all known algorithms need at least $\Omega(m)$ communication for these problems, where $m$ is the number of edges in the graph. We address the following question in this paper: can we solve problems such as coloring and MIS using sublinear, i.e., $o(m)$ communication, and if so under what conditions? [See full abstract in pdf]

cs.DC↗

arXiv2021

The Complexity of Symmetry Breaking in Massive Graphs

The goal of this paper is to understand the complexity of symmetry breaking problems, specifically maximal independent set (MIS) and the closely related $\beta$-ruling set problem, in two computational models suited for large-scale graph processing, namely the $k$-machine model and the graph streaming model. We present a number of results. For MIS in the $k$-machine model, we improve the $\tilde{O}(m/k^2 + \Delta/k)$-round upper bound of Klauck et al. (SODA 2015) by presenting an $\tilde{O}(m/k^2)$-round algorithm. We also present an $\tilde{\Omega}(n/k^2)$ round lower bound for MIS, the first lower bound for a symmetry breaking problem in the $k$-machine model. For $\beta$-ruling sets, we use hierarchical sampling to obtain more efficient algorithms in the $k$-machine model and also in the graph streaming model. More specifically, we obtain a $k$-machine algorithm that runs in $\tilde{O}(\beta n\Delta^{1/\beta}/k^2)$ rounds and, by using a similar hierarchical sampling technique, we obtain one-pass algorithms for both insertion-only and insertion-deletion streams that use $O(\beta \cdot n^{1+1/2^{\beta-1}})$ space. The latter result establishes a clear separation between MIS, which is known to require $\Omega(n^2)$ space (Cormode et al., ICALP 2019), and $\beta$-ruling sets, even for $\beta = 2$. Finally, we present an even faster 2-ruling set algorithm in the $k$-machine model, one that runs in $\tilde{O}(n/k^{2-\epsilon} + k^{1-\epsilon})$ rounds for any $\epsilon$, $0 \le \epsilon \le 1$.

cs.DC↗