15552

Automatic Data Layout Generation and Kernel Mapping for CPU+GPU Architectures

Deepak Majeti, Kuldeep S. Meel, Rajkishore Barik, Vivek Sarkar
Rice University, USA
International Conference on Compiler Construction (CC), 2016

@proceedings{MMBV16,

   author={Majeti, Deepak and Meel, Kuldeep S. and Barik, Raj and Sarkar, Vivek},

   title={Automatic Data Layout Generation and Kernel Mapping for CPU+GPU Architectures},

   year={2016},

   month={3}

}

Download Download (PDF)   View View   Source Source   

508

views

The ubiquity of hybrid CPU+GPU architectures has led to renewed interest in automatic data layout generation owing to the fact that data layouts have a large impact on performance, and that different data layouts yield the best performance on CPUs vs. GPUs. Unfortunately, current programming models still fail to provide an effective solution to the problem of automatic data layout generation for CPU+GPU processors. Specifically, the interaction among whole-program data layout optimizations, data movement optimizations, and mapping of kernels across heterogeneous cores poses a major challenge to current programming systems. In this paper, we introduce a novel two-level hierarchical formulation of the data layout and kernel mapping problem for modern heterogeneous architectures. The top level formulation targets data layouts and kernel mapping for the entire program for which we provide a polynomial- time solution using a graph-based shortest path algorithm that uses the data layouts for the code regions (sections) for a given processor computed in the bottom level formulation. The bottom level formulation deals with the data layout problem for a parallel code region on a given processor, which is NP-Hard, and we provide a greedy algorithm that uses an affinity graph to obtain approximate solutions. We have implemented this data layout transformation in the new Heterogeneous Habanero-C (H2C) parallel programming framework and propose performance models to characterize the data layout impact on both the CPU and GPU. Our data layout framework shows significant performance improvements of up to 2.9 (geometric mean 1.5) on a multicore CPU+GPU compared to the manually specified layouts for a set of parallel programs running on a heterogeneous platform consisting of an Intel Xeon CPU and a NVIDIA GPU. Further, our framework also shows performance improvements of up to 2.7 (geometric mean 1.6) on just the multicore CPU, demonstrating the applicability of our approach to both heterogeneous and homogeneous hardware platforms.
VN:F [1.9.22_1171]
Rating: 0.0/5 (0 votes cast)

* * *

* * *

TwitterAPIExchange Object
(
    [oauth_access_token:TwitterAPIExchange:private] => 301967669-yDz6MrfyJFFsH1DVvrw5Xb9phx2d0DSOFuLehBGh
    [oauth_access_token_secret:TwitterAPIExchange:private] => o29ji3VLVmB6jASMqY8G7QZDCrdFmoTvCDNNUlb7s
    [consumer_key:TwitterAPIExchange:private] => TdQb63pho0ak9VevwMWpEgXAE
    [consumer_secret:TwitterAPIExchange:private] => Uq4rWz7nUnH1y6ab6uQ9xMk0KLcDrmckneEMdlq6G5E0jlQCFx
    [postfields:TwitterAPIExchange:private] => 
    [getfield:TwitterAPIExchange:private] => ?cursor=-1&screen_name=hgpu&skip_status=true&include_user_entities=false
    [oauth:protected] => Array
        (
            [oauth_consumer_key] => TdQb63pho0ak9VevwMWpEgXAE
            [oauth_nonce] => 1481242382
            [oauth_signature_method] => HMAC-SHA1
            [oauth_token] => 301967669-yDz6MrfyJFFsH1DVvrw5Xb9phx2d0DSOFuLehBGh
            [oauth_timestamp] => 1481242382
            [oauth_version] => 1.0
            [cursor] => -1
            [screen_name] => hgpu
            [skip_status] => true
            [include_user_entities] => false
            [oauth_signature] => r4DDIR553//dKCR19Z8VudWeczo=
        )

    [url] => https://api.twitter.com/1.1/users/show.json
)
Follow us on Facebook
Follow us on Twitter

HGPU group

2081 peoples are following HGPU @twitter

HGPU group © 2010-2016 hgpu.org

All rights belong to the respective authors

Contact us: