dStream: An Online-Based Dynamic Operator-Level Query Mapping Scheme on Discrete CPU-GPU Architectures
In streaming systems with a discrete CPU-GPU architecture, leveraging the strengths of both processing units can significantly improve query performance. Existing studies assign entire queries to either the CPU or GPU to reduce data transfer overhead and boost throughput. However, this coarse-graine...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10776952/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841536207695118336 |
|---|---|
| author | Gyeonghwan Jung Yeonwoo Jeong Kyuli Park Dongjae Lee Hongsu Byun Suyeon Lee Sungyong Park |
| author_facet | Gyeonghwan Jung Yeonwoo Jeong Kyuli Park Dongjae Lee Hongsu Byun Suyeon Lee Sungyong Park |
| author_sort | Gyeonghwan Jung |
| collection | DOAJ |
| description | In streaming systems with a discrete CPU-GPU architecture, leveraging the strengths of both processing units can significantly improve query performance. Existing studies assign entire queries to either the CPU or GPU to reduce data transfer overhead and boost throughput. However, this coarse-grained approach can limit performance for two main reasons. Firstly, PCIe transfer overhead is minimal for small data sizes, and the device preference of each operator within a query may change with variations in data size. Secondly, it neglects performance fluctuations based on the placement location of consecutive operators within the device. To address these issues, we propose dSTREAM, a distributed stream processing system that dynamically maps queries at the operator level on discrete CPU-GPU architectures. dSTREAM adapts to runtime conditions by selecting the optimal device for each operator dynamically. Through dynamic operator-level query mapping without prior knowledge, dSTREAM consistently achieves high performance. Extensive evaluation has confirmed that dSTREAM enhances average throughput by up to 45% and reduces average latency by up to 42.5% across various types of stream SQL queries, regardless of traffic types. |
| format | Article |
| id | doaj-art-cb4cbe0129c64d1c9ab2237da2254178 |
| institution | Kabale University |
| issn | 2169-3536 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-cb4cbe0129c64d1c9ab2237da22541782025-01-15T00:02:55ZengIEEEIEEE Access2169-35362025-01-01138239825610.1109/ACCESS.2024.351088510776952dStream: An Online-Based Dynamic Operator-Level Query Mapping Scheme on Discrete CPU-GPU ArchitecturesGyeonghwan Jung0https://orcid.org/0009-0002-9393-3936Yeonwoo Jeong1https://orcid.org/0000-0003-4653-115XKyuli Park2https://orcid.org/0009-0001-2404-3430Dongjae Lee3https://orcid.org/0009-0005-8487-9888Hongsu Byun4Suyeon Lee5https://orcid.org/0000-0002-5526-6127Sungyong Park6https://orcid.org/0000-0002-0309-1820Department of Computer Science and Engineering, Sogang University, Seoul, South KoreaDepartment of Computer Science and Engineering, Sogang University, Seoul, South KoreaDepartment of Computer Science and Engineering, Sogang University, Seoul, South KoreaDepartment of Computer Science and Engineering, Sogang University, Seoul, South KoreaDepartment of Computer Science and Engineering, Sogang University, Seoul, South KoreaDepartment of Computer Science, Georgia Institute of Technology, Atlanta, GA, USADepartment of Computer Science and Engineering, Sogang University, Seoul, South KoreaIn streaming systems with a discrete CPU-GPU architecture, leveraging the strengths of both processing units can significantly improve query performance. Existing studies assign entire queries to either the CPU or GPU to reduce data transfer overhead and boost throughput. However, this coarse-grained approach can limit performance for two main reasons. Firstly, PCIe transfer overhead is minimal for small data sizes, and the device preference of each operator within a query may change with variations in data size. Secondly, it neglects performance fluctuations based on the placement location of consecutive operators within the device. To address these issues, we propose dSTREAM, a distributed stream processing system that dynamically maps queries at the operator level on discrete CPU-GPU architectures. dSTREAM adapts to runtime conditions by selecting the optimal device for each operator dynamically. Through dynamic operator-level query mapping without prior knowledge, dSTREAM consistently achieves high performance. Extensive evaluation has confirmed that dSTREAM enhances average throughput by up to 45% and reduces average latency by up to 42.5% across various types of stream SQL queries, regardless of traffic types.https://ieeexplore.ieee.org/document/10776952/Query planningdata stream processingheterogeneous architectures |
| spellingShingle | Gyeonghwan Jung Yeonwoo Jeong Kyuli Park Dongjae Lee Hongsu Byun Suyeon Lee Sungyong Park dStream: An Online-Based Dynamic Operator-Level Query Mapping Scheme on Discrete CPU-GPU Architectures IEEE Access Query planning data stream processing heterogeneous architectures |
| title | dStream: An Online-Based Dynamic Operator-Level Query Mapping Scheme on Discrete CPU-GPU Architectures |
| title_full | dStream: An Online-Based Dynamic Operator-Level Query Mapping Scheme on Discrete CPU-GPU Architectures |
| title_fullStr | dStream: An Online-Based Dynamic Operator-Level Query Mapping Scheme on Discrete CPU-GPU Architectures |
| title_full_unstemmed | dStream: An Online-Based Dynamic Operator-Level Query Mapping Scheme on Discrete CPU-GPU Architectures |
| title_short | dStream: An Online-Based Dynamic Operator-Level Query Mapping Scheme on Discrete CPU-GPU Architectures |
| title_sort | dstream an online based dynamic operator level query mapping scheme on discrete cpu gpu architectures |
| topic | Query planning data stream processing heterogeneous architectures |
| url | https://ieeexplore.ieee.org/document/10776952/ |
| work_keys_str_mv | AT gyeonghwanjung dstreamanonlinebaseddynamicoperatorlevelquerymappingschemeondiscretecpugpuarchitectures AT yeonwoojeong dstreamanonlinebaseddynamicoperatorlevelquerymappingschemeondiscretecpugpuarchitectures AT kyulipark dstreamanonlinebaseddynamicoperatorlevelquerymappingschemeondiscretecpugpuarchitectures AT dongjaelee dstreamanonlinebaseddynamicoperatorlevelquerymappingschemeondiscretecpugpuarchitectures AT hongsubyun dstreamanonlinebaseddynamicoperatorlevelquerymappingschemeondiscretecpugpuarchitectures AT suyeonlee dstreamanonlinebaseddynamicoperatorlevelquerymappingschemeondiscretecpugpuarchitectures AT sungyongpark dstreamanonlinebaseddynamicoperatorlevelquerymappingschemeondiscretecpugpuarchitectures |