Domain-Specialized Cache Management for Graph Analytics

Faldu, Priyank; Diamond, Jeff; Grot, Boris

doi:10.1109/hpca47549.2020.00028

Cited by 38 publications

(9 citation statements)

References 59 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It shows that TDGraph-H consistently outperforms the other schemes as the LLC size increases. Moreover, compared with the other LLC management methods, i.e., LRU [27], DRRIP [25], and P-OPT [9], TDGraph-H achieves better performance when GRASP [19] is employed. This is because GRASP can efficiently prevent the coalesced states of the frequently-accessed vertices from cache thrashing.…”

Section: Sensitivity Studiesmentioning

confidence: 99%

“…A number of hardware techniques have been developed to augment the many-core processor for efficient graph processing. GRASP [19] utilizes a domain-specialized cache policy to protect hot vertices against cache thrashing. P-OPT [9] is further designed to use the graph's transpose to ensure Belady's optimal cache replacement for graph processing.…”

Section: Hardware Support For Graph Processingmentioning

confidence: 99%

“…Although some hardware solutions [9,19,36,37,64,67,73] have been designed to augment the many-core processors for highperformance graph processing, they are designed to accelerate the processing of static graphs. Meanwhile, to support streaming graph processing, some ASIC-based accelerators, e.g., JetStream [44] and DREDGE [35], have been recently proposed.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

TDGraph

Zhao

Yang

Liao

et al. 2022

Proceedings of the 49th Annual International Symposium on Computer Architecture

View full text Add to dashboard Cite

Many solutions have been recently proposed to support the processing of streaming graphs. However, for the processing of each graph snapshot of a streaming graph, the new states of the vertices affected by the graph updates are propagated irregularly along the graph topology. Despite the years' research efforts, existing approaches still suffer from the serious problems of redundant computation overhead and irregular memory access, which severely underutilizes a many-core processor. To address these issues, this paper proposes a topology-driven programmable accelerator TDGraph, which is the first accelerator to augment the many-core processors to achieve high performance processing of streaming graphs. Specifically, we propose an efficient topology-driven incremental execution approach into the accelerator design for more regular state propagation and better data locality. TDGraph takes the vertices affected by graph updates as the roots to prefetch other vertices along the graph topology and synchronizes the incremental computations of them on the fly. In this way, most state propagations originated from multiple vertices affected by different graph updates can be conducted together along the graph topology, which help reduce the redundant computations and data access cost. Besides, through the efficient coalescing of the accesses to vertex states, TDGraph further improves the utilization of the cache and memory bandwidth. We have evaluated TDGraph on a simulated 64-core processor. The results show that, the state-of-the-art software system achieves the speedup of 7.1∼21.4 times after integrating with TDGraph, while incurring only 0.73% area cost. Compared with four cutting-edge accelerators, i.e., HATS, Minnow, PHI, and DepGraph, TDGraph gains the speedups of 4.6∼12.7, 3.2∼8.6, 3.8∼9.7, and 2.3∼6.1 times, respectively.

show abstract

Section: Sensitivity Studiesmentioning

confidence: 99%

Section: Hardware Support For Graph Processingmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

TDGraph

Zhao

Yang

Liao

et al. 2022

Proceedings of the 49th Annual International Symposium on Computer Architecture

View full text Add to dashboard Cite

show abstract

“…This optimization turns skew from problem to opportunity. This is not a surprise: researches have repeatedly observed that skew is a form of locality, and as such it can be leveraged to increase performance [15,18,43,45].…”

Section: Related Workmentioning

confidence: 99%

Odyssey

Gavrielatos

Katsarakis

Nagarajan

2021

Proceedings of the Sixteenth European Conference on Computer Systems

View full text Add to dashboard Cite

Get/Put Key-Value Stores (KVSes) rely on replication protocols to enforce consistency and guarantee availability. Today's modern hardware, with manycore servers and RDMAcapable networks, challenges the conventional wisdom on protocol design. In this paper, we investigate the impact of modern hardware on the performance of strongly-consistent replication protocols.First, we create an informal taxonomy of replication protocols, based on which we carefully select 10 protocols for analysis. Secondly, we present Odyssey, a framework tailored towards protocol implementation for multi-threaded, RDMA-enabled, in-memory, replicated KVSes. We implement all 10 protocols over Odyssey, and perform the first apples-to-apples comparison of replication protocols over modern hardware.Our comparison characterizes the protocol design space, revealing the performance capabilities of different classes of protocols on modern hardware. Among other things, our results demonstrate that some of the protocols that were efficient in yesterday's hardware are not so today because they cannot take advantage of the abundant parallelism and fast networking present in modern hardware. Conversely, some protocols that were inefficient in yesterday's hardware are very attractive today. We distill our findings in a concise set of general guidelines and recommendations for protocol selection and design in the era of modern hardware.

show abstract

“…Ozdal et al [51] designed dedicated caches for different types of graph data according to the access characteristics. In addition, recent studies aim at exposing flexibility [51][52][53] and releasing productivity [51,54,55] for ASIC-based graph accelerators.…”

Section: Introductionmentioning

confidence: 99%

Software Systems Implementation and Domain-Specific Architectures towards Graph Analytics

Jin

Zhao

et al. 2022

Intell Comput

View full text Add to dashboard Cite

Graph analytics, which mainly includes graph processing, graph mining, and graph learning, has become increasingly important in several domains, including social network analysis, bioinformatics, and machine learning. However, graph analytics applications suffer from poor locality, limited bandwidth, and low parallelism owing to the irregular sparse structure, explosive growth, and dependencies of graph data. To address those challenges, several programming models, execution modes, and messaging strategies are proposed to improve the utilization of traditional hardware and performance. In recent years, novel computing and memory devices have emerged, e.g., HMCs, HBM, and ReRAM, providing massive bandwidth and parallelism resources, making it possible to address bottlenecks in graph applications. To facilitate understanding of the graph analytics domain, our study summarizes and categorizes current software systems implementation and domain-specific architectures. Finally, we discuss the future challenges of graph analytics.

show abstract

Domain-Specialized Cache Management for Graph Analytics

Cited by 38 publications

References 59 publications

TDGraph

TDGraph

Odyssey

Software Systems Implementation and Domain-Specific Architectures towards Graph Analytics

Contact Info

Product

Resources

About