arXiv subjects

Ümit V. Çatalyürek

Publications and source records attributed to Ümit V. Çatalyürek.

At least 19 recordsLinked to original sources

arXiv2025

Poseidon: A OneGraph Engine

We present the Poseidon engine behind the Neptune Analytics graph database service. Customers interact with Poseidon using the declarative openCypher query language, which enables requests that seamlessly combine traditional querying paradigms (such as graph pattern matching, variable length paths, aggregation) with algorithm invocations and has been syntactically extended to facilitate OneGraph interoperability, such as the disambiguation between globally unique IRIs (as exposed via RDF) vs. local identifiers (as encountered in LPG data). Poseidon supports a broad range of graph workloads, from simple transactions, to top-k beam search algorithms on dynamic graphs, to whole graph analytics requiring multiple full passes over the data. For example, real-time fraud detection, like many other use cases, needs to reflect current committed state of the dynamic graph. If a users cell phone is compromised, then all newer actions by that user become immediately suspect. To address such dynamic graph use cases, Poseidon combines state-of-the-art transaction processing with novel graph data indexing, including lock-free maintenance of adjacency lists, secondary succinct indices, partitioned heaps for data tuple storage with uniform placement, and innovative statistics for cost-based query optimization. The Poseidon engine uses a logical log for durability, enabling rapid evolution of in-memory data structures. Bulk data loads achieve more than 10 million property values per second on many data sets while simple transactions can execute in under 20ms against the storage engine.

cs.DB↗

arXiv2024

Cooperative Minibatching in Graph Neural Networks

Training large scale Graph Neural Networks (GNNs) requires significant computational resources, and the process is highly data-intensive. One of the most effective ways to reduce resource requirements is minibatch training coupled with graph sampling. GNNs have the unique property that items in a minibatch have overlapping data. However, the commonly implemented Independent Minibatching approach assigns each Processing Element (PE, i.e., cores and/or GPUs) its own minibatch to process, leading to duplicated computations and input data access across PEs. This amplifies the Neighborhood Explosion Phenomenon (NEP), which is the main bottleneck limiting scaling. To reduce the effects of NEP in the multi-PE setting, we propose a new approach called Cooperative Minibatching. Our approach capitalizes on the fact that the size of the sampled subgraph is a concave function of the batch size, leading to significant reductions in the amount of work as batch sizes increase. Hence, it is favorable for processors equipped with a fast interconnect to work on a large minibatch together as a single larger processor, instead of working on separate smaller minibatches, even though global batch size is identical. We also show how to take advantage of the same phenomenon in serial execution by generating dependent consecutive minibatches. Our experimental evaluations show up to 4x bandwidth savings for fetching vertex embeddings, by simply increasing this dependency without harming model convergence. Combining our proposed approaches, we achieve up to 64% speedup over Independent Minibatching on single-node multi-GPU systems, using same resources.

cs.LG↗

arXiv2024

A Scalable and Effective Alternative to Graph Transformers

Graph Neural Networks (GNNs) have shown impressive performance in graph representation learning, but they face challenges in capturing long-range dependencies due to their limited expressive power. To address this, Graph Transformers (GTs) were introduced, utilizing self-attention mechanism to effectively model pairwise node relationships. Despite their advantages, GTs suffer from quadratic complexity w.r.t. the number of nodes in the graph, hindering their applicability to large graphs. In this work, we present Graph-Enhanced Contextual Operator (GECO), a scalable and effective alternative to GTs that leverages neighborhood propagation and global convolutions to effectively capture local and global dependencies in quasilinear time. Our study on synthetic datasets reveals that GECO reaches 169x speedup on a graph with 2M nodes w.r.t. optimized attention. Further evaluations on diverse range of benchmarks showcase that GECO scales to large graphs where traditional GTs often face memory and time limitations. Notably, GECO consistently achieves comparable or superior quality compared to baselines, improving the SOTA up to 4.5%, and offering a scalable and effective solution for large-scale graph learning.

cs.LG↗

arXiv2023

Open Problems in (Hyper)Graph Decomposition

Large networks are useful in a wide range of applications. Sometimes problem instances are composed of billions of entities. Decomposing and analyzing these structures helps us gain new insights about our surroundings. Even if the final application concerns a different problem (such as traversal, finding paths, trees, and flows), decomposing large graphs is often an important subproblem for complexity reduction or parallelization. This report is a summary of discussions that happened at Dagstuhl seminar 23331 on "Recent Trends in Graph Decomposition" and presents currently open problems and future directions in the area of (hyper)graph decomposition.

cs.DS↗

arXiv2023

SGORP: A Subgradient-based Method for d-Dimensional Rectilinear Partitioning

Partitioning for load balancing is a crucial first step to parallelize any type of computation. In this work, we propose SGORP, a new spatial partitioning method based on Subgradient Optimization, to solve the $d$-dimensional Rectilinear Partitioning Problem (RPP). Our proposed method allows the use of customizable objective functions as well as some user-specific constraints, such as symmetric partitioning on selected dimensions. Extensive experimental evaluation using over 600 test matrices shows that our algorithm achieves favorable performance against the state-of-the-art RPP and Symmetric RPP algorithms. Additionally, we show the effectiveness of our algorithm to do application-specific load balancing using two applications as motivation: Triangle Counting and Sparse Matrix Multiplication (SpGEMM), where we model their load-balancing problems as $3$-dimensional RPPs.

cs.DS↗

arXiv2023

Layer-Neighbor Sampling -- Defusing Neighborhood Explosion in GNNs

Graph Neural Networks (GNNs) have received significant attention recently, but training them at a large scale remains a challenge. Mini-batch training coupled with sampling is used to alleviate this challenge. However, existing approaches either suffer from the neighborhood explosion phenomenon or have poor performance. To address these issues, we propose a new sampling algorithm called LAyer-neighBOR sampling (LABOR). It is designed to be a direct replacement for Neighbor Sampling (NS) with the same fanout hyperparameter while sampling up to 7 times fewer vertices, without sacrificing quality. By design, the variance of the estimator of each vertex matches NS from the point of view of a single vertex. Moreover, under the same vertex sampling budget constraints, LABOR converges faster than existing layer sampling approaches and can use up to 112 times larger batch sizes compared to NS.

cs.LG↗

arXiv2022

A Simple and Elegant Mathematical Formulation for the Acyclic DAG Partitioning Problem

This work addresses the NP-Hard problem of acyclic directed acyclic graph (DAG) partitioning problem. The acyclic partitioning problem is defined as partitioning the vertex set of a given directed acyclic graph into disjoint and collectively exhaustive subsets (parts). Parts are to be assigned such that the total sum of the vertex weights within each part satisfies a common upper bound and the total sum of the edge costs that connect nodes across different parts is minimized. Additionally, the quotient graph, i.e., the induced graph where all nodes that are assigned to the same part are contracted to a single node and edges of those are replaced with cumulative edges towards other nodes, is also a directed acyclic graph. That is, the quotient graph itself is also a graph that contains no cycles. Many computational and real-life applications such as in computational task scheduling, RTL simulations, scheduling of rail-rail transshipment tasks and Very Large Scale Integration (VLSI) design make use of acyclic DAG partitioning. We address the need for a simple and elegant mathematical formulation for the acyclic DAG partitioning problem that enables easier understanding, communication, implementation, and experimentation on the problem.

cs.DS↗

arXiv2022

More Recent Advances in (Hyper)Graph Partitioning

In recent years, significant advances have been made in the design and evaluation of balanced (hyper)graph partitioning algorithms. We survey trends of the last decade in practical algorithms for balanced (hyper)graph partitioning together with future research directions. Our work serves as an update to a previous survey on the topic. In particular, the survey extends the previous survey by also covering hypergraph partitioning and streaming algorithms, and has an additional focus on parallel algorithms.

cs.DS↗

arXiv2022

Efficient Hierarchical State Vector Simulation of Quantum Circuits via Acyclic Graph Partitioning

Early but promising results in quantum computing have been enabled by the concurrent development of quantum algorithms, devices, and materials. Classical simulation of quantum programs has enabled the design and analysis of algorithms and implementation strategies targeting current and anticipated quantum device architectures. In this paper, we present a graph-based approach to achieve efficient quantum circuit simulation. Our approach involves partitioning the graph representation of a given quantum circuit into acyclic sub-graphs/circuits that exhibit better data locality. Simulation of each sub-circuit is organized hierarchically, with the iterative construction and simulation of smaller state vectors, improving overall performance. Also, this partitioning reduces the number of passes through data, improving the total computation time. We present three partitioning strategies and observe that acyclic graph partitioning typically results in the best time-to-solution. In contrast, other strategies reduce the partitioning time at the expense of potentially increased simulation times. Experimental evaluation demonstrates the effectiveness of our approach.

quant-ph↗

arXiv2022

Batch Dynamic Algorithm to Find k-Cores and Hierarchies

Finding $k$-cores in graphs is a valuable and effective strategy for extracting dense regions of otherwise sparse graphs. We focus on the important problem of maintaining cores on rapidly changing dynamic graphs, where batches of edge changes need to be processed quickly. Prior batch core algorithms have only addressed half the problem of maintaining cores, the problem of maintaining a core decomposition. This finds vertices that are dense, but not regions; it misses connectivity. To address this, we bring an efficient index from community search into the core domain, the Shell Tree Index. We develop a novel dynamic batch algorithm to maintain it that improves efficiency over processing edge-by-edge. We implement our algorithm and experimentally show that with it core queries can be returned on rapidly changing graphs quickly enough for interactive applications. For 1 million edge batches, on many graphs we run over $100\times$ faster than processing edge-by-edge while remaining under re-computing from scratch.

cs.DS↗

arXiv2021

MG-GCN: Scalable Multi-GPU GCN Training Framework

Full batch training of Graph Convolutional Network (GCN) models is not feasible on a single GPU for large graphs containing tens of millions of vertices or more. Recent work has shown that, for the graphs used in the machine learning community, communication becomes a bottleneck and scaling is blocked outside of the single machine regime. Thus, we propose MG-GCN, a multi-GPU GCN training framework taking advantage of the high-speed communication links between the GPUs present in multi-GPU systems. MG-GCN employs multiple High-Performance Computing optimizations, including efficient re-use of memory buffers to reduce the memory footprint of training GNN models, as well as communication and computation overlap. These optimizations enable execution on larger datasets, that generally do not fit into memory of a single GPU in state-of-the-art implementations. Furthermore, they contribute to achieve superior speedup compared to the state-of-the-art. For example, MG-GCN achieves super-linear speedup with respect to DGL, on the Reddit graph on both DGX-1 (V100) and DGX-A100.

cs.LG↗

arXiv2020

A Block-Based Triangle Counting Algorithm on Heterogeneous Environments

Triangle counting is a fundamental building block in graph algorithms. In this paper, we propose a block-based triangle counting algorithm to reduce data movement during both sequential and parallel execution. Our block-based formulation makes the algorithm naturally suitable for heterogeneous architectures. The problem of partitioning the adjacency matrix of a graph is well-studied. Our task decomposition goes one step further: it partitions the set of triangles in the graph. By streaming these small tasks to compute resources, we can solve problems that do not fit on a device. We demonstrate the effectiveness of our approach by providing an implementation on a compute node with multiple sockets, cores and GPUs. The current state-of-the-art in triangle enumeration processes the Friendster graph in 2.1 seconds, not including data copy time between CPU and GPU. Using that metric, our approach is 20 percent faster. When copy times are included, our algorithm takes 3.2 seconds. This is 5.6 times faster than the fastest published CPU-only time.

cs.DS↗

arXiv2020

On Symmetric Rectilinear Matrix Partitioning

Even distribution of irregular workload to processing units is crucial for efficient parallelization in many applications. In this work, we are concerned with a spatial partitioning called rectilinear partitioning (also known as generalized block distribution) of sparse matrices. More specifically, in this work, we address the problem of symmetric rectilinear partitioning of a square matrix. By symmetric, we mean the rows and columns of the matrix are identically partitioned yielding a tiling where the diagonal tiles (blocks) will be squares. We first show that the optimal solution to this problem is NP-hard, and we propose four heuristics to solve two different variants of this problem. We present a thorough analysis of the computational complexities of those proposed heuristics. To make the proposed techniques more applicable in real life application scenarios, we further reduce their computational complexities by utilizing effective sparsification strategies together with an efficient sparse prefix-sum data structure. We experimentally show the proposed algorithms are efficient and effective on more than six hundred test matrices. With sparsification, our methods take less than 3 seconds in the Twitter graph on a modern 24 core system and output a solution whose load imbalance is no worse than 1%.

cs.DS↗

arXiv2019

Heuristics for Symmetric Rectilinear Matrix Partitioning

Partitioning sparse matrices and graphs is a common and important problem in many scientific and graph analytics applications. In this work, we are concerned with a spatial partitioning called rectilinear partitioning (also known as generalized block distribution) of sparse matrices, which is needed for tiled (or {\em blocked}) execution of sparse matrix and graph analytics kernels. More specifically, in this work, we address the problem of symmetric rectilinear partitioning of square matrices. By symmetric, we mean having the same partition on rows and columns of the matrix, yielding a special tiling where the diagonal tiles (blocks) will be squares. We propose five heuristics to solve two different variants of this problem, and present a thorough experimental evaluation showing the effectiveness of the proposed algorithms.

cs.DS↗

arXiv2018

Programming Strategies for Irregular Algorithms on the Emu Chick

The Emu Chick prototype implements migratory memory-side processing in a novel hardware system. Rather than transferring large amounts of data across the system interconnect, the Emu Chick moves lightweight thread contexts to near-memory cores before the beginning of each remote memory read. Previous work has characterized the performance of the Chick prototype in terms of memory bandwidth and programming differences from more typical, non-migratory platforms, but there has not yet been an analysis of algorithms on this system. This work evaluates irregular algorithms that could benefit from the lightweight, memory-side processing of the Chick and demonstrates techniques and optimization strategies for achieving performance in sparse matrix-vector multiply operation (SpMV), breadth-first search (BFS), and graph alignment across up to eight distributed nodes encompassing 64 nodelets in the Chick system. We also define and justify relative metrics to compare prototype FPGA-based hardware with established ASIC architectures. The Chick currently supports up to 68x scaling for graph alignment, 80 MTEPS for BFS on balanced graphs, and 50\% of measured STREAM bandwidth for SpMV.

cs.DC↗

arXiv2018

An Iterative Global Structure-Assisted Labeled Network Aligner

Integrating data from heterogeneous sources is often modeled as merging graphs. Given two or more 'compatible', but not-isomorphic graphs, the first step is to identify a graph alignment, where a potentially partial mapping of vertices between two graphs is computed. A significant portion of the literature on this problem only takes the global structure of the input graphs into account. Only more recent ones additionally use vertex and edge attributes to achieve a more accurate alignment. However, these methods are not designed to scale to map large graphs arising in many modern applications. We propose a new iterative graph aligner, gsaNA, that uses the global structure of the graphs to significantly reduce the problem size and align large graphs with a minimal loss of information. Concretely, we show that our proposed technique is highly flexible, can be used to achieve higher recall, and it is orders of magnitudes faster than the current state of the art techniques.

cs.SI↗

arXiv2014

On Distributed Graph Coloring with Iterative Recoloring

Identifying the sets of operations that can be executed simultaneously is an important problem appearing in many parallel applications. By modeling the operations and their interactions as a graph, one can identify the independent operations by solving a graph coloring problem. Many efficient sequential algorithms are known for this NP-Complete problem, but they are typically unsuitable when the operations and their interactions are distributed in the memory of large parallel computers. On top of an existing distributed-memory graph coloring algorithm, we investigate two compatible techniques in this paper for fast and scalable distributed-memory graph coloring. First, we introduce an improvement for the distributed post-processing operation, called recoloring, which drastically improves the number of colors. We propose a novel and efficient communication scheme for recoloring which enables it to scale gracefully. Recoloring must be seeded with an existing coloring of the graph. Our second contribution is to introduce a randomized color selection strategy for initial coloring which quickly produces solutions of modest quality. We extensively evaluate the impact of our new techniques on existing distributed algorithms and show the time-quality tradeoffs. We show that combining an initial randomized coloring with multiple recoloring iterations yields better quality solutions with the smaller runtime at large scale.

cs.DC↗

arXiv2012

Diversifying Citation Recommendations

Literature search is arguably one of the most important phases of the academic and non-academic research. The increase in the number of published papers each year makes manual search inefficient and furthermore insufficient. Hence, automatized methods such as search engines have been of interest in the last thirty years. Unfortunately, these traditional engines use keyword-based approaches to solve the search problem, but these approaches are prone to ambiguity and synonymy. On the other hand, bibliographic search techniques based only on the citation information are not prone to these problems since they do not consider textual similarity. For many particular research areas and topics, the amount of knowledge to humankind is immense, and obtaining the desired information is as hard as looking for a needle in a haystack. Furthermore, sometimes, what we are looking for is a set of documents where each one is different than the others, but at the same time, as a whole we want them to cover all the important parts of the literature relevant to our search. This paper targets the problem of result diversification in citation-based bibliographic search. It surveys a set of techniques which aim to find a set of papers with satisfactory quality and diversity. We enhance these algorithms with a direction-awareness functionality to allow the users to reach either old, well-cited, well-known research papers or recent, less-known ones. We also propose a set of novel techniques for a better diversification of the results. All the techniques considered are compared by performing a rigorous experimentation. The results show that some of the proposed techniques are very successful in practice while performing a search in a bibliographic database.

cs.IR↗