Abelian: A Compiler for Graph Analytics on Distributed, Heterogeneous Platforms

Gill, Gurbinder; Dathathri, Roshan; Hoang, Loc; Lenharth, Andrew; Pingali, Keshav

doi:10.1007/978-3-319-96983-1_18

Cited by 16 publications

(6 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Galois programming model [39] is an efficient model for encoding graph analytics for shared memory systems. Besides Gluon, there are other prior works that propose various compilers for converting BSP style programs written in Galois model to Distributed systems [19] and on to GPUs [42]. Many shared memory frameworks [35,40,42,49,57,58] have also been proposed for graph analytics.…”

Section: Related Workmentioning

confidence: 99%

DisGCo

Rajendran

Nandivada

2020

ACM Trans. Archit. Code Optim.

View full text Add to dashboard Cite

Graph algorithms are widely used in various applications. Their programmability and performance have garnered a lot of interest among the researchers. Being able to run these graph analytics programs on distributed systems is an important requirement. Green-Marl is a popular Domain Specific Language (DSL) for coding graph algorithms and is known for its simplicity. However, the existing Green-Marl compiler for distributed systems (Green-Marl to Pregel) can only compile limited types of Green-Marl programs (in Pregel canonical form). This severely restricts the types of parallel Green-Marl programs that can be executed on distributed systems. We present DisGCo, the first compiler to translate any general Green-Marl program to equivalent MPI program that can run on distributed systems. Translating Green-Marl programs to MPI (SPMD/MPMD style of computation, distributed memory) presents many other exciting challenges, besides the issues related to differences in syntax, as Green-Marl gives the programmer a unified view of the whole memory and allows the parallel and serial code to be intermixed. We first present the set of challenges involved in translating Green-Marl programs to MPI and then present a systematic approach to do the translation. We also present a few optimization techniques to improve the performance of our generated programs. DisGCo is the first graph DSL compiler that can handle all syntactic capabilities of a practical graph DSL like Green-Marl and generate code that can run on distributed systems. Our preliminary evaluation of DisGCo shows that our generated programs are scalable. Further, compared to the state-of-the-art DH-Falcon compiler that translates a subset of Falcon programs to MPI, our generated codes exhibit a geomean speedup of 17.32×.

show abstract

Section: Related Workmentioning

confidence: 99%

DisGCo

Rajendran

Nandivada

2020

ACM Trans. Archit. Code Optim.

View full text Add to dashboard Cite

show abstract

“…GG supports ordered graph algorithms with a GPU-based two-bucket priority queue. Abelian [18] uses the Galois framework as an interface for shared-memory CPU, distributedmemory CPU, and GPU platforms. However, it lacks support for direction-optimization, various load balancing optimizations, and active vertexset creation optimizations, which are needed to achieve high performance.…”

Section: Related Workmentioning

confidence: 99%

Compilation Techniques for Graph Algorithms on GPUs

Brahmakshatriya,

Zhang,

Hong

et al. 2020

Preprint

View full text Add to dashboard Cite

The performance of graph programs depends highly on the algorithm, size, and structure of the input graphs, as well as the features of the underlying hardware. No single set of optimizations or one hardware platform works well across all settings. To achieve high performance, the programmer must carefully select which set optimizations as well as the hardware platform to use. However, currently when switching between CPUs and GPUs, programmer must re-implement the entire algorithm with different optimizations in order to achieve high performance.We propose the GG GPU compiler, a new graph processing framework that achieves high performance on both CPUs and GPUs using the same algorithm specification. GG significantly expands the optimization space of GPU graph processing frameworks with a novel GPU scheduling language and compiler that enables combining load balancing, edge traversal direction, active vertex set creation, active vertex set processing ordering, and kernel fusion optimizations. GG also introduces two performance optimizations, Edge-based Thread Warps CTAs load balancing (ETWC) and EdgeBlocking, to expand the optimization space for GPUs. ETWC improves load balance by dynamically partitioning the edges of each vertex into blocks that are assigned to threads, warps, and CTAs for execution. EdgeBlocking improves the locality of the program by reordering the edges and restricting random memory accesses to fit within L2 cache. We evaluate GG on 5 algorithms and 9 input graphs on both Pascal and Volta generation NVIDIA GPUs, and show that it achieves up to 5.11× speedup over state-of-the-art GPU graph processing frameworks, and is the fastest on 62 out of the 90 experiments.

show abstract

“…The compiler particularly addresses the challenges of optimizing throughput due to the special thread organization, the SIMD execution model, and the complex memory hierarchy of the GPU architecture. The Abelian [16] compiler takes a graph computation algorithm and generates code to execute on distributed systems with heterogeneous processors.…”

Section: Graph Computation Systemsmentioning

confidence: 99%

AutoMine

Mawhirter

2019

Proceedings of the 27th ACM Symposium on Operating Systems Principles

View full text Add to dashboard Cite

Graph mining algorithms that aim at identifying structural patterns of graphs are typically more complex than graph computation algorithms such as breadth first search. Researchers have implemented several systems with high-level and flexible interfaces customized for tackling graph mining problems. However, we find that for triangle counting, one of the simplest graph mining problems, such systems can be several times slower than a single-threaded implementation of a straightforward algorithm. In this paper, we reveal the root causes of the severe inefficiencies of state-of-the-art graph mining systems and the challenges to address the performance problems. We build AutoMine, a single-machine system to provide both highlevel interfaces and high performance for large-scale graph mining applications. The novelty of AutoMine comes from 1) a new representation of subgraph patterns and 2) compilation techniques that automatically generate efficient mining code with minimized memory consumption from a highlevel abstraction. We have extensively evaluated AutoMine against 3 graph mining systems on 8 real-world graphs of different scales. Our experimental results show that AutoMine often produces several orders of magnitude better performance and can process very large graphs existing systems cannot handle. CCS Concepts • Computing methodologies → Shared memory algorithms; • Software and its engineering → Compilers.

show abstract

Abelian: A Compiler for Graph Analytics on Distributed, Heterogeneous Platforms

Cited by 16 publications

References 27 publications

DisGCo

DisGCo

Compilation Techniques for Graph Algorithms on GPUs

AutoMine

Contact Info

Product

Resources

About