Search

This paper proposes efficient solutions for $k$-core decomposition with high parallelism. The problem of $k$-core decomposition is fundamental in graph analysis and has applications across various domains. However, existing algorithms face significant challenges in achieving work-efficiency in theory and/or high parallelism in practice, and suffer from various performance bottlenecks. We present a simple, work-efficient parallel framework for $k$-core decomposition that is easy to implement and adaptable to various strategies for improving work-efficiency. We introduce two techniques to enhance parallelism: a sampling scheme to reduce contention on high-degree vertices, and vertical granularity control (VGC) to mitigate scheduling overhead for low-degree vertices. Furthermore, we design a hierarchical bucket structure to optimize performance for graphs with high coreness values. We evaluate our algorithm on a diverse set of real-world and synthetic graphs. Compared to state-of-the-art parallel algorithms, including ParK, PKC, and Julienne, our approach demonstrates superior performance on 23 out of 25 graphs when tested on a 96-core machine. Our algorithm shows speedups of up to 315$\times$ over ParK, 33.4$\times$ over PKC, and 52.5$\times$ over Julienne., Comment: 18 pages, 15 figures

Contraction Hierarchies (CH) (Geisberger et al., 2008) is one of the most widely used algorithms for shortest-path queries on road networks. Compared to Dijkstra's algorithm, CH enables orders of magnitude faster query performance through a preprocessing phase, which iteratively categorizes vertices into hierarchies and adds shortcuts. However, constructing a CH is an expensive task. Existing solutions, including parallel ones, may suffer from long construction time. Especially in our experiments, we observe that existing parallel solutions demonstrate unsatisfactory scalability and have close performance to sequential algorithms. In this paper, we present \textsf{SPoCH} (\textbf{S}calable \textbf{P}arallelization \textbf{o}f \textbf{C}ontraction \textbf{H}ierarchies), an efficient and scalable CH construction algorithm in parallel. To address the challenges in previous work, our improvements focus on both redesigning the algorithm and leveraging parallel data structures. %to maintain the original and shortcut edges dynamically. We implement our algorithm and compare it with the state-of-the-art sequential and parallel implementations on 13 graphs, including road networks, synthetic graphs, and $k$-NN graphs. Our experiments show that \textsf{SPoCH} achieves $17$--$131\times$ speedups in CH construction over the best baseline, while maintaining competitive query performance and CH graph size.

The $k$d-tree is one of the most widely used data structures to manage multi-dimensional data. Due to the ever-growing data volume, it is imperative to consider parallelism in $k$d-trees. However, we observed challenges in existing parallel kd-tree implementations, for both constructions and updates. The goal of this paper is to develop efficient in-memory $k$d-trees by supporting high parallelism and cache-efficiency. We propose the Pkd-tree (Parallel $k$d-tree), a parallel $k$d-tree that is efficient both in theory and in practice. The Pkd-tree supports parallel tree construction, batch update (insertion and deletion), and various queries including k-nearest neighbor search, range query, and range count. We proved that our algorithms have strong theoretical bounds in work (sequential time complexity), span (parallelism), and cache complexity. Our key techniques include 1) an efficient construction algorithm that optimizes work, span, and cache complexity simultaneously, and 2) reconstruction-based update algorithms that guarantee the tree to be weight-balanced. With the new algorithmic insights and careful engineering effort, we achieved a highly optimized implementation of the Pkd-tree. We tested Pkd-tree with various synthetic and real-world datasets, including both uniform and highly skewed data. We compare the Pkd-tree with state-of-the-art parallel $k$d-tree implementations. In all tests, with better or competitive query performance, Pkd-tree is much faster in construction and updates consistently than all baselines. We released our code.

Breadth-first Search (BFS) is one of the most important graph processing subroutines, especially for computing the unweighted distance. Many applications may require running BFS from multiple sources. Sequentially, when running BFS on a cluster of nearby vertices, a known optimization is using bit-parallelism. Given a subset of vertices with size $k$ and the distance between any pair of them is no more than $d$, BFS can be applied to all of them in total work $O(dm(k/w+1))$, where $w$ is the length of a word in bits and $m$ is the number of edges. We will refer to this approach as cluster-BFS (C-BFS). Such an approach has been studied and shown effective both in theory and in practice in the sequential setting. However, it remains unknown how this can be combined with thread-level parallelism. In this paper, we focus on designing efficient parallel C-BFS based on BFS to answer unweighted distance queries. Our solution combines the strengths of bit-level parallelism and thread-level parallelism, and achieves significant speedup over the plain sequential solution. We also apply our algorithm to real-world applications. In particular, we identified another application (landmark-labeling for the approximate distance oracle) that can take advantage of parallel C-BFS. Under the same memory budget, our new solution improves accuracy and/or time on all the 18 tested graphs.

In this paper, we introduce PASGAL (Parallel And Scalable Graph Algorithm Library), a parallel graph library that scales to a variety of graph types, many processors, and large graph sizes. One special focus of PASGAL is the efficiency on \textit{large-diameter graphs}, which is a common challenge for many existing parallel graph processing systems: many existing graph processing systems can be even slower than the standard sequential algorithm on large-diameter graphs due to the lack of parallelism. Such performance degeneration is caused by the high overhead in scheduling and synchronizing threads when traversing the graph in the breadth-first order. The core technique in PASGAL to achieve high parallelism is a technique called \textit{vertical granularity control (VGC)} to hide synchronization overhead, as well as careful redesign of parallel graph algorithms and data structures. In our experiments, we compare PASGAL with state-of-the-art parallel implementations on BFS, SCC, BCC, and SSSP. PASGAL achieves competitive performance on small-diameter graphs compared to the parallel baselines, and is significantly faster on large-diameter graphs.

The idea of dynamic programming (DP), proposed by Bellman in the 1950s, is one of the most important algorithmic techniques. However, in parallel, many fundamental and sequentially simple problems become more challenging, and open to a (nearly) work-efficient solution (i.e., the work is off by at most a polylogarithmic factor over the best sequential solution). In fact, sequential DP algorithms employ many advanced optimizations such as decision monotonicity or special data structures, and achieve better work than straightforward solutions. Many such optimizations are inherently sequential, which creates extra challenges for a parallel algorithm to achieve the same work bound. The goal of this paper is to achieve (nearly) work-efficient parallel DP algorithms by parallelizing classic, highly-optimized and practical sequential algorithms. We show a general framework called the Cordon Algorithm for parallel DP algorithms, and use it to solve several classic problems. Our selection of problems includes Longest Increasing Subsequence (LIS), sparse Longest Common Subsequence (LCS), convex/concave generalized Least Weight Subsequence (LWS), Optimal Alphabetic Tree (OAT), and more. We show how the Cordon Algorithm can be used to achieve the same level of optimization as the sequential algorithms, and achieve good parallelism. Many of our algorithms are conceptually simple, and we show some experimental results as proofs-of-concept.

Integer sorting is a fundamental problem in computer science. This paper studies parallel integer sort both in theory and in practice. In theory, we show tighter bounds for a class of existing practical integer sort algorithms, which provides a solid theoretical foundation for their widespread usage in practice and strong performance. In practice, we design a new integer sorting algorithm, \textsf{DovetailSort}, that is theoretically-efficient and has good practical performance. In particular, \textsf{DovetailSort} overcomes a common challenge in existing parallel integer sorting algorithms, which is the difficulty of detecting and taking advantage of duplicate keys. The key insight in \textsf{DovetailSort} is to combine algorithmic ideas from both integer- and comparison-sorting algorithms. In our experiments, \textsf{DovetailSort} achieves competitive or better performance than existing state-of-the-art parallel integer and comparison sorting algorithms on various synthetic and real-world datasets.

Influence Maximization (IM) is a crucial problem in data science. The goal is to find a fixed-size set of highly-influential seed vertices on a network to maximize the influence spread along the edges. While IM is NP-hard on commonly-used diffusion models, a greedy algorithm can achieve $(1-1/e)$-approximation, repeatedly selecting the vertex with the highest marginal gain in influence as the seed. Due to theoretical guarantees, rich literature focuses on improving the performance of the greedy algorithm. To estimate the marginal gain, existing work either runs Monte Carlo (MC) simulations of influence spread or pre-stores hundreds of sketches (usually per-vertex information). However, these approaches can be inefficient in time (MC simulation) or space (storing sketches), preventing the ideas from scaling to today's large-scale graphs. This paper significantly improves the scalability of IM using two key techniques. The first is a sketch-compression technique for the independent cascading model on undirected graphs. It allows combining the simulation and sketching approaches to achieve a time-space tradeoff. The second technique includes new data structures for parallel seed selection. Using our new approaches, we implemented PaC-IM: Parallel and Compressed IM. We compare PaC-IM with state-of-the-art parallel IM systems on a 96-core machine with 1.5TB memory. PaC-IM can process large-scale graphs with up to 900M vertices and 74B edges in about 2 hours. On average across all tested graphs, our uncompressed version is 5--18$\times$ faster and about 1.4$\times$ more space-efficient than existing parallel IM systems. Using compression further saves 3.8$\times$ space with only 70% overhead in time on average.

Given two strings $A[1..n]$ and $B[1..m]$, and a set of operations allowed to edit the strings, the edit distance between $A$ and $B$ is the minimum number of operations required to transform $A$ into $B$. Sequentially, a standard Dynamic Programming (DP) algorithm solves edit distance with $\Theta(nm)$ cost. In many real-world applications, the strings to be compared are similar and have small edit distances. To achieve highly practical implementations, we focus on output-sensitive parallel edit-distance algorithms, i.e., to achieve asymptotically better cost bounds than the standard $\Theta(nm)$ algorithm when the edit distance is small. We study four algorithms in the paper, including three algorithms based on Breadth-First Search (BFS) and one algorithm based on Divide-and-Conquer (DaC). Our BFS-based solution is based on the Landau-Vishkin algorithm. We implement three different data structures for the longest common prefix (LCP) queries needed in the algorithm: the classic solution using parallel suffix array, and two hash-based solutions proposed in this paper. Our DaC-based solution is inspired by the output-insensitive solution proposed by Apostolico et al., and we propose a non-trivial adaption to make it output-sensitive. All our algorithms have good theoretical guarantees, and they achieve different tradeoffs between work (total number of operations), span (longest dependence chain in the computation), and space. We test and compare our algorithms on both synthetic data and real-world data. Our BFS-based algorithms outperform the existing parallel edit-distance implementation in ParlayLib in all test cases. By comparing our algorithms, we also provide a better understanding of the choice of algorithms for different input patterns. We believe that our paper is the first systematic study in the theory and practice of parallel edit distance.

Approximate nearest-neighbor search (ANNS) algorithms are a key part of the modern deep learning stack due to enabling efficient similarity search over high-dimensional vector space representations (i.e., embeddings) of data. Among various ANNS algorithms, graph-based algorithms are known to achieve the best throughput-recall tradeoffs. Despite the large scale of modern ANNS datasets, existing parallel graph based implementations suffer from significant challenges to scale to large datasets due to heavy use of locks and other sequential bottlenecks, which 1) prevents them from efficiently scaling to a large number of processors, and 2) results in nondeterminism that is undesirable in certain applications. In this paper, we introduce ParlayANN, a library of deterministic and parallel graph-based approximate nearest neighbor search algorithms, along with a set of useful tools for developing such algorithms. In this library, we develop novel parallel implementations for four state-of-the-art graph-based ANNS algorithms that scale to billion-scale datasets. Our algorithms are deterministic and achieve high scalability across a diverse set of challenging datasets. In addition to the new algorithmic ideas, we also conduct a detailed experimental study of our new algorithms as well as two existing non-graph approaches. Our experimental results both validate the effectiveness of our new techniques, and lead to a comprehensive comparison among ANNS algorithms on large scale datasets with a list of interesting findings.

Semisort is a fundamental algorithmic primitive widely used in the design and analysis of efficient parallel algorithms. It takes input as an array of records and a function extracting a \emph{key} per record, and reorders them so that records with equal keys are contiguous. Since many applications only require collecting equal values, but not fully sorting the input, semisort is broadly applicable, e.g., in string algorithms, graph analytics, and geometry processing, among many other domains. However, despite dozens of recent papers that use semisort in their theoretical analysis and the existence of an asymptotically optimal parallel semisort algorithm, most implementations of these parallel algorithms choose to implement semisort by using comparison or integer sorting in practice, due to potential performance issues in existing semisort implementations. In this paper, we revisit the semisort problem, with the goal of achieving a high-performance parallel semisort implementation with a flexible interface. Our approach can easily extend to two related problems, \emph{histogram} and \emph{collect-reduce}. Our algorithms achieve strong speedups in practice, and importantly, outperform state-of-the-art parallel sorting and semisorting methods for almost all settings we tested, with varying input sizes, distribution, and key types. We also test two important applications with real-world data, and show that our algorithms improve the performance over existing approaches. We believe that many other parallel algorithm implementations can be accelerated using our results.

Computing strongly connected components (SCC) is a fundamental problems in graph processing. As today's real-world graphs are getting larger and larger, parallel SCC is increasingly important. SCC is challenging in the parallel setting and is particularly hard on large-diameter graphs. Many existing parallel SCC implementations can be even slower than Tarjan's sequential algorithm on large-diameter graphs. To tackle this challenge, we propose an efficient parallel SCC implementation using a new parallel reachability algorithm. Our solution is based on a novel idea referred to as vertical granularity control (VGC). It breaks the synchronization barriers to increase parallelism and hide scheduling overhead. To use VGC in our SCC algorithm, we also design an efficient data structure called the \emph{parallel hash bag}. It uses parallel dynamic resizing to avoid redundant work in maintaining frontiers (vertices processed in a round). We implement the parallel SCC algorithm by Blelloch et al.\ (J.\ ACM, 2020) using our new parallel reachability algorithm. We compare our implementation to the state-of-the-art systems, including GBBS, iSpan, Multi-step, and our highly optimized Tarjan's (sequential) algorithm, on 18 graphs, including social, web, $k$-NN, and lattice graphs. On a machine with 96 cores, our implementation is the fastest on 16 out of 18 graphs. On average (geometric means) over all graphs, our SCC is 6.0$\times$ faster than the best previous parallel code (GBBS), 12.8$\times$ faster than Tarjan's sequential algorithms, and 2.7$\times$ faster than the \emph{best existing implementation on each graph}. We believe that our techniques are of independent interest. We also apply our parallel hash bag and VGC scheme to other graph problems, including connectivity and least-element lists (LE-lists).

Biconnectivity is one of the most fundamental graph problems. The canonical parallel biconnectivity algorithm is the Tarjan-Vishkin algorithm, which has $O(n+m)$ optimal work (number of operations) and polylogarithmic span (longest dependent operations) on a graph with $n$ vertices and $m$ edges. However, Tarjan-Vishkin is not widely used in practice. We believe the reason is the space-inefficiency (it generates an auxiliary graph with $O(m)$ edges). In practice, existing parallel implementations are based on breath-first search (BFS). Since BFS has span proportional to the diameter of the graph, existing parallel BCC implementations suffer from poor performance on large-diameter graphs and can be even slower than the sequential algorithm on many real-world graphs. We propose the first parallel biconnectivity algorithm (FAST-BCC) that has optimal work, polylogarithmic span, and is space-efficient. Our algorithm first generates a skeleton graph based on any spanning tree of the input graph. Then we use the connectivity information of the skeleton to compute the biconnectivity of the original input. All the steps in our algorithm are highly-parallel. We carefully analyze the correctness of our algorithm, which is highly non-trivial. We implemented FAST-BCC and compared it with existing implementations, including GBBS, Slota and Madduri's algorithm, and the sequential Hopcroft-Tarjan algorithm. We ran them on a 96-core machine on 27 graphs, including social, web, road, $k$-NN, and synthetic graphs, with significantly varying sizes and edge distributions. FAST-BCC is the fastest on all 27 graphs. On average (geometric means), FAST-BCC is 5.1$\times$ faster than GBBS, and 3.1$\times$ faster than the best existing baseline on each graph.

This paper studies parallel algorithms for the longest increasing subsequence (LIS) problem. Let $n$ be the input size and $k$ be the LIS length of the input. Sequentially, LIS is a simple problem that can be solved using dynamic programming (DP) in $O(n\log n)$ work. However, parallelizing LIS is a long-standing challenge. We are unaware of any parallel LIS algorithm that has optimal $O(n\log n)$ work and non-trivial parallelism (i.e., $\tilde{O}(k)$ or $o(n)$ span). This paper proposes a parallel LIS algorithm that costs $O(n\log k)$ work, $\tilde{O}(k)$ span, and $O(n)$ space, and is much simpler than the previous parallel LIS algorithms. We also generalize the algorithm to a weighted version of LIS, which maximizes the weighted sum for all objects in an increasing subsequence. To achieve a better work bound for the weighted LIS algorithm, we designed parallel algorithms for the van Emde Boas (vEB) tree, which has the same structure as the sequential vEB tree, and supports work-efficient parallel batch insertion, deletion, and range queries. We also implemented our parallel LIS algorithms. Our implementation is light-weighted, efficient, and scalable. On input size $10^9$, our LIS algorithm outperforms a highly-optimized sequential algorithm (with $O(n\log k)$ cost) on inputs with $k\le 3\times 10^5$. Our algorithm is also much faster than the best existing parallel implementation by Shen et al. (2022) on all input instances., Comment: to be published in Proceedings of the 35th ACM Symposium on Parallelism in Algorithms and Architectures (SPAA '23)

To design efficient parallel algorithms, some recent papers showed that many sequential iterative algorithms can be directly parallelized but there are still challenges in achieving work-efficiency and high-parallelism. Work-efficiency can be hard for certain problems where the number of dependences is asymptotically more than optimal sequential work bound. To achieve high-parallelism, we want to process as many objects as possible in parallel. The goal is to achieve $\tilde{O}(D)$ span for a problem with the deepest dependence length $D$. We refer to this property as round-efficiency. In this paper, we show work-efficient and round-efficient algorithms for a variety of classic problems and propose general approaches to do so. To efficiently parallelize many sequential iterative algorithms, we propose the phase-parallel framework. The framework assigns a rank to each object and processes them accordingly. All objects with the same rank can be processed in parallel. To enable work-efficiency and high parallelism, we use two types of general techniques. Type 1 algorithms aim to use range queries to extract all objects with the same rank, such that we avoid evaluating all the dependences. We discuss activity selection, unlimited knapsack, and more using Type 1 framework. Type 2 algorithms aim to wake up an object when the last object it depends on is finished. We discuss activity selection, longest increasing subsequence (LIS), and many other algorithms using Type 2 framework. All of our algorithms are (nearly) work-efficient and round-efficient. Many of them improve previous best bounds, and some of them are the first to achieve work-efficiency with round-efficiency. We also implement many of them. On inputs with reasonable dependence depth, our algorithms are highly parallelized and significantly outperform their sequential counterparts.

Many modern programming languages are shifting toward a functional style for collection interfaces such as sets, maps, and sequences. Functional interfaces offer many advantages, including being safe for parallelism and providing simple and lightweight snapshots. However, existing high-performance functional interfaces such as PAM, which are based on balanced purely-functional trees, incur large space overheads for large-scale data analysis due to storing every element in a separate node in a tree. This paper presents PaC-trees, a purely-functional data structure supporting functional interfaces for sets, maps, and sequences that provides a significant reduction in space over existing approaches. A PaC-tree is a balanced binary search tree which blocks the leaves and compresses the blocks using arrays. We provide novel techniques for compressing and uncompressing the blocks which yield practical parallel functional algorithms for a broad set of operations on PaC-trees such as union, intersection, filter, reduction, and range queries which are both theoretically and practically efficient. Using PaC-trees we designed CPAM, a C++ library that implements the full functionality of PAM, while offering significant extra functionality for compression. CPAM consistently matches or outperforms PAM on a set of microbenchmarks on sets, maps, and sequences while using about a quarter of the space. On applications including inverted indices, 2D range queries, and 1D interval queries, CPAM is competitive with or faster than PAM, while using 2.1--7.8x less space. For static and streaming graph processing, CPAM offers 1.6x faster batch updates while using 1.3--2.6x less space than the state-of-the-art graph processing system Aspen., Comment: This is a preliminary version of a paper that will appear at the ACM SIGPLAN Conference on Programming Language Design and Implementation (PLDI 2022)

Parallelism has become extremely popular over the past decade, and there have been a lot of new parallel algorithms and software. The randomized work-stealing (RWS) scheduler plays a crucial role in this ecosystem. In this paper, we study two important topics related to the randomized work-stealing scheduler. Our first contribution is a simplified, classroom-ready version of analysis for the RWS scheduler. The theoretical efficiency of the RWS scheduler has been analyzed for a variety of settings, but most of them are quite complicated. In this paper, we show a new analysis, which we believe is easy to understand, and can be especially useful in education. We avoid using the potential function in the analysis, and we assume a highly asynchronous setting, which is more realistic for today's parallel machines. Our second and main contribution is some new parallel cache complexity for algorithms using the RWS scheduler. Although the sequential I/O model has been well-studied over the past decades, so far very few results have extended it to the parallel setting. The parallel cache bounds of many existing algorithms are affected by a polynomial of the span, which causes a significant overhead for high-span algorithms. Our new analysis decouples the span from the analysis of the parallel cache complexity. This allows us to show new parallel cache bounds for a list of classic algorithms. Our results are only a polylogarithmic factor off the lower bounds, and significantly improve previous results.

We present a general technique for garbage collecting old versions for multiversion concurrency control that simultaneously achieves good time and space complexity. Our technique takes only $O(1)$ time on average to reclaim each version and maintains only a constant factor more versions than needed (plus an additive term). It is designed for multiversion schemes using version lists, which are the most common. Our approach uses two components that are of independent interest. First, we define a novel range-tracking data structure which stores a set of old versions and efficiently finds those that are no longer needed. We provide a wait-free implementation in which all operations take amortized constant time. Second, we represent version lists using a new lock-free doubly-linked list algorithm that supports efficient (amortized constant time) removals given a pointer to any node in the list. These two components naturally fit together to solve the multiversion garbage collection problem--the range-tracker identifies which versions to remove and our list algorithm can then be used to remove them from their version lists. We apply our garbage collection technique to generate end-to-end time and space bounds for the multiversioning system of Wei et al. (PPoPP 2021)., Comment: This is the full version of the paper appearing in the International Symposium on Distributed Computing (DISC 2021)

In this paper, we study the single-source shortest-path (SSSP) problem with positive edge weights, which is a notoriously hard problem in the parallel context. In practice, the $\Delta$-stepping algorithm proposed by Meyer and Sanders has been widely adopted. However, $\Delta$-stepping has no known worst-case bounds for general graphs. The performance of $\Delta$-stepping also highly relies on the parameter $\Delta$. There have also been lots of algorithms with theoretical bounds, such as Radius-stepping, but they either have no implementations available or are much slower than $\Delta$-stepping in practice. We propose a stepping algorithm framework that generalizes existing algorithms such as $\Delta$-stepping and Radius-stepping. The framework allows for similar analysis and implementations of all stepping algorithms. We also propose a new ADT, lazy-batched priority queue (LaB-PQ), that abstracts the semantics of the priority queue needed by the stepping algorithms. We provide two data structures for LaB-PQ, focusing on theoretical and practical efficiency, respectively. Based on the new framework and LaB-PQ, we show two new stepping algorithms, $\rho$-stepping and $\Delta^*$-stepping, that are simple, with non-trivial worst-case bounds, and fast in practice. The stepping algorithm framework also provides almost identical implementations for three algorithms: Bellman-Ford, $\Delta^*$-stepping, and $\rho$-stepping. We compare our code with four state-of-the-art implementations. On five social and web graphs, $\rho$-stepping is 1.3--2.5x faster than all the existing implementations. On two road graphs, our $\Delta^*$-stepping is at least 14\% faster than existing implementations, while $\rho$-stepping is also competitive. The almost identical implementations for stepping algorithms also allow for in-depth analyses and comparisons among the stepping algorithms in practice.

We present an approach for efficiently taking snapshots of the state of a collection of CAS objects. Taking a snapshot allows later operations to read the value that each CAS object had at the time the snapshot was taken. Taking a snapshot requires a constant number of steps and returns a handle to the snapshot. Reading a snapshotted value of an individual CAS object using this handle is wait-free, taking time proportional to the number of successful CASes on the object since the snapshot was taken. Our fast, flexible snapshots yield simple, efficient implementations of atomic multi-point queries on concurrent data structures built from CAS objects. For example, in a search tree where child pointers are updated using CAS, once a snapshot is taken, one can atomically search for ranges of keys, find the first key that matches some criteria, or check if a collection of keys are all present, simply by running a standard sequential algorithm on a snapshot of the tree. To evaluate the performance of our approach, we apply it to two search trees, one balanced and one not. Experiments show that the overhead of supporting snapshots is low across a variety of workloads. Moreover, in almost all cases, range queries on the trees built from our snapshots perform as well as or better than state-of-the-art concurrent data structures that support atomic range queries., Comment: To appear in PPoPP'21