Teseo and the analysis of structural dynamic graphs

Leo, Dean De; Boncz, Peter

doi:10.14778/3447689.3447708

Cited by 21 publications

(6 citation statements)

References 59 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…• How much memory does CSR++ consume with these workloads? • How does CSR++ perform in comparison to other read-friendly (i.e., CSR) and updatefriendly graph structures (i.e., ALs, LLAMA [3], STINGER [23] , GraphOne [22], and Teseo [24])?…”

Section: Discussionmentioning

confidence: 99%

“…Teseo [24] Transactional Fat Tree based on PMA Asynchronous rebalances delayed to 200 ms and 1 MB maximum leaf capacity.…”

Section: Discussionmentioning

confidence: 99%

“…We evaluate CSR++ on both read and update workloads, with various graphs and graph algorithms, and compare it against CSR, ALs, LLAMA [3], STINGER [23], Teseo [24], and GraphOne [22]. Our results indicate that CSR++ is much faster than the other updatefriendly data structures while being almost as fast as CSR with read-only workloads.…”

Section: Introductionmentioning

confidence: 98%

“…Another technique to support graph mutation is "in-place updates", which does not require rebuilding the entire vertex/edge structures; instead, an update is directly stored either by using extra space that already exists within the structure or by reallocating only part of the edge/vertex tables. Existing graph structures with in-place mutations include STINGER [23] and Teseo [24]. STINGER relies on ALs, which store edges in linked lists of blocks, while Teseo extends CSR by using Packed Memory Arrays (PMAs) [25] to store the edge table.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

A Scalable Data Structure for Efficient Graph Analytics and In-Place Mutations

Firmli,

Chiadmi

2023

Data

View full text Add to dashboard Cite

The graph model enables a broad range of analyses; thus, graph processing (GP) is an invaluable tool in data analytics. At the heart of every GP system lies a concurrent graph data structure that stores the graph. Such a data structure needs to be highly efficient for both graph algorithms and queries. Due to the continuous evolution, the sparsity, and the scale-free nature of real-world graphs, GP systems face the challenge of providing an appropriate graph data structure that enables both fast analytical workloads and fast, low-memory graph mutations. Existing graph structures offer a hard tradeoff among read-only performance, update friendliness, and memory consumption upon updates. In this paper, we introduce CSR++, a new graph data structure that removes these tradeoffs and enables both fast read-only analytics, and quick and memory-friendly mutations. CSR++ combines ideas from CSR, the fastest read-only data structure, and adjacency lists (ALs) to achieve the best of both worlds. We compare CSR++ to CSR, ALs from the Boost Graph Library (BGL), and the following state-of-the-art update-friendly graph structures: LLAMA, STINGER, GraphOne, and Teseo. In our evaluation, which is based on popular GP algorithms executed over real-world graphs, we show that CSR++ remains close to CSR in read-only concurrent performance (within 10% on average) while significantly outperforming CSR (by an order of magnitude) and LLAMA (by almost 2×) with frequent updates. We also show that both CSR++’s update throughput and analytics performance exceed those of several state-of-the-art graph structures while maintaining low memory consumption when the workload includes updates.

show abstract

Section: Discussionmentioning

confidence: 99%

“…Teseo [24] Transactional Fat Tree based on PMA Asynchronous rebalances delayed to 200 ms and 1 MB maximum leaf capacity.…”

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 98%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Scalable Data Structure for Efficient Graph Analytics and In-Place Mutations

Firmli,

Chiadmi

2023

Data

View full text Add to dashboard Cite

show abstract

“…Manually implemented dynamic graphs are typically inefficient and inflexible. Although there are some dynamic graph storages [10,22,77], they focus on graph analysis workloads and do not natively support storing node vectors. Dynamic GNNs require storage structures supporting node embedding storage and multi-version embedding storage for backpropagation.…”

Section: Neutronstream 51 Dynamic Graph Storagementioning

confidence: 99%

NeutronStream: A Dynamic GNN Training Framework with Sliding Window for Graph Streams

Chen,

Gao,

Zhang

et al. 2023

Proc. VLDB Endow.

View full text Add to dashboard Cite

Existing Graph Neural Network (GNN) training frameworks have been designed to help developers easily create performant GNN implementations. However, most existing GNN frameworks assume that the input graphs are static, but ignore that most real-world graphs are constantly evolving. Though many dynamic GNN models have emerged to learn from evolving graphs, the training process of these dynamic GNNs is dramatically different from traditional GNNs in that it captures both the spatial and temporal dependencies of graph updates. This poses new challenges for designing dynamic GNN training frameworks. First, the traditional batched training method fails to capture real-time structural evolution information. Second, the time-dependent nature makes parallel training hard to design. Third, it lacks system supports for users to efficiently implement dynamic GNNs. In this paper, we present NeutronStream, a framework for training dynamic GNN models. NeutronStream abstracts the input dynamic graph into a chronologically updated stream of events and processes the stream with an optimized sliding window to incrementally capture the spatial-temporal dependencies of events. Furthermore, NeutronStream provides a parallel execution engine to tackle the sequential event processing challenge to achieve high performance. NeutronStream also integrates a built-in graph storage structure that supports dynamic updates and provides a set of easy-to-use APIs that allow users to express their dynamic GNNs. Our experimental results demonstrate that, compared to state-of-the-art dynamic GNN implementations, NeutronStream achieves speedups ranging from 1.48X to 5.87X and an average accuracy improvement of 3.97%.

show abstract

Adaptive update handling for graph HTAP

Jibril

Baumstark

Sattler

2023

Distrib Parallel Databases

View full text Add to dashboard Cite

Hybrid transactional/analytical processing (HTAP) workloads on graph data can significantly benefit from GPU accelerators. However, to exploit the full potential of GPU processing, dedicated graph representations are necessary, which mostly make in-place updates difficult. In this paper, we discuss an adaptive update handling approach in a graph database system for HTAP workloads. We discuss and evaluate strategies for propagating transactional updates from an update-friendly table storage to a GPU-optimized sparse matrix format for analytics.

show abstract

Teseo and the analysis of structural dynamic graphs

Cited by 21 publications

References 59 publications

A Scalable Data Structure for Efficient Graph Analytics and In-Place Mutations

A Scalable Data Structure for Efficient Graph Analytics and In-Place Mutations

NeutronStream: A Dynamic GNN Training Framework with Sliding Window for Graph Streams

Adaptive update handling for graph HTAP

Contact Info

Product

Resources

About