Parallelization of tree-to-TLV serialization

Makoto Nakayama, Kenichi Yamazaki, Satoshi Tanaka, Hironori Kasahara

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    1 Citation (Scopus)

    Abstract

    A serializer/deserializer (SerDe) is necessary to serialize a data object into a byte array and to deserialize in reverse direction. A SerDe that is used worldwide and runs quickly is the Protocol Buffer (ProtoBuf), which serializes a tree-structured data object into the Type-Length-Value (TLV) format. Acceleration of SerDe processing is beneficial because SerDes are used in various fields. This paper proposes a new method that accelerates the tree-to-TLV serialization through 2-way parallel processing called "parallelized serialization" and "parallelization with streaming". Experimental results show that parallelized serialization with 4 worker threads achieves a 1.97 fold shorter serialization time than when using a single worker thread, and the combination of 2-way parallel processing achieves a 2.11 fold shorter output time than that for ProtoBuf when 4 worker threads, File Output Stream and trees of 10,080 container nodes are used.

    Original languageEnglish
    Title of host publication2014 IEEE 33rd International Performance Computing and Communications Conference, IPCCC 2014
    PublisherInstitute of Electrical and Electronics Engineers Inc.
    Volume2014-January
    ISBN (Electronic)9781479975754
    DOIs
    Publication statusPublished - 2015 Jan 20
    Event33rd IEEE International Performance Computing and Communications Conference, IPCCC 2014 - Austin, United States
    Duration: 2014 Dec 52014 Dec 7

    Other

    Other33rd IEEE International Performance Computing and Communications Conference, IPCCC 2014
    CountryUnited States
    CityAustin
    Period14/12/514/12/7

    Fingerprint

    Processing
    Containers

    Keywords

    • Multi-Core
    • Parallel Processing
    • Serialization

    ASJC Scopus subject areas

    • Software
    • Computational Theory and Mathematics
    • Computer Networks and Communications

    Cite this

    Nakayama, M., Yamazaki, K., Tanaka, S., & Kasahara, H. (2015). Parallelization of tree-to-TLV serialization. In 2014 IEEE 33rd International Performance Computing and Communications Conference, IPCCC 2014 (Vol. 2014-January). [7017059] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/PCCC.2014.7017059

    Parallelization of tree-to-TLV serialization. / Nakayama, Makoto; Yamazaki, Kenichi; Tanaka, Satoshi; Kasahara, Hironori.

    2014 IEEE 33rd International Performance Computing and Communications Conference, IPCCC 2014. Vol. 2014-January Institute of Electrical and Electronics Engineers Inc., 2015. 7017059.

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Nakayama, M, Yamazaki, K, Tanaka, S & Kasahara, H 2015, Parallelization of tree-to-TLV serialization. in 2014 IEEE 33rd International Performance Computing and Communications Conference, IPCCC 2014. vol. 2014-January, 7017059, Institute of Electrical and Electronics Engineers Inc., 33rd IEEE International Performance Computing and Communications Conference, IPCCC 2014, Austin, United States, 14/12/5. https://doi.org/10.1109/PCCC.2014.7017059
    Nakayama M, Yamazaki K, Tanaka S, Kasahara H. Parallelization of tree-to-TLV serialization. In 2014 IEEE 33rd International Performance Computing and Communications Conference, IPCCC 2014. Vol. 2014-January. Institute of Electrical and Electronics Engineers Inc. 2015. 7017059 https://doi.org/10.1109/PCCC.2014.7017059
    Nakayama, Makoto ; Yamazaki, Kenichi ; Tanaka, Satoshi ; Kasahara, Hironori. / Parallelization of tree-to-TLV serialization. 2014 IEEE 33rd International Performance Computing and Communications Conference, IPCCC 2014. Vol. 2014-January Institute of Electrical and Electronics Engineers Inc., 2015.
    @inproceedings{0e7b41b3dbb74e43bf437f818c3e895e,
    title = "Parallelization of tree-to-TLV serialization",
    abstract = "A serializer/deserializer (SerDe) is necessary to serialize a data object into a byte array and to deserialize in reverse direction. A SerDe that is used worldwide and runs quickly is the Protocol Buffer (ProtoBuf), which serializes a tree-structured data object into the Type-Length-Value (TLV) format. Acceleration of SerDe processing is beneficial because SerDes are used in various fields. This paper proposes a new method that accelerates the tree-to-TLV serialization through 2-way parallel processing called {"}parallelized serialization{"} and {"}parallelization with streaming{"}. Experimental results show that parallelized serialization with 4 worker threads achieves a 1.97 fold shorter serialization time than when using a single worker thread, and the combination of 2-way parallel processing achieves a 2.11 fold shorter output time than that for ProtoBuf when 4 worker threads, File Output Stream and trees of 10,080 container nodes are used.",
    keywords = "Multi-Core, Parallel Processing, Serialization",
    author = "Makoto Nakayama and Kenichi Yamazaki and Satoshi Tanaka and Hironori Kasahara",
    year = "2015",
    month = "1",
    day = "20",
    doi = "10.1109/PCCC.2014.7017059",
    language = "English",
    volume = "2014-January",
    booktitle = "2014 IEEE 33rd International Performance Computing and Communications Conference, IPCCC 2014",
    publisher = "Institute of Electrical and Electronics Engineers Inc.",
    address = "United States",

    }

    TY - GEN

    T1 - Parallelization of tree-to-TLV serialization

    AU - Nakayama, Makoto

    AU - Yamazaki, Kenichi

    AU - Tanaka, Satoshi

    AU - Kasahara, Hironori

    PY - 2015/1/20

    Y1 - 2015/1/20

    N2 - A serializer/deserializer (SerDe) is necessary to serialize a data object into a byte array and to deserialize in reverse direction. A SerDe that is used worldwide and runs quickly is the Protocol Buffer (ProtoBuf), which serializes a tree-structured data object into the Type-Length-Value (TLV) format. Acceleration of SerDe processing is beneficial because SerDes are used in various fields. This paper proposes a new method that accelerates the tree-to-TLV serialization through 2-way parallel processing called "parallelized serialization" and "parallelization with streaming". Experimental results show that parallelized serialization with 4 worker threads achieves a 1.97 fold shorter serialization time than when using a single worker thread, and the combination of 2-way parallel processing achieves a 2.11 fold shorter output time than that for ProtoBuf when 4 worker threads, File Output Stream and trees of 10,080 container nodes are used.

    AB - A serializer/deserializer (SerDe) is necessary to serialize a data object into a byte array and to deserialize in reverse direction. A SerDe that is used worldwide and runs quickly is the Protocol Buffer (ProtoBuf), which serializes a tree-structured data object into the Type-Length-Value (TLV) format. Acceleration of SerDe processing is beneficial because SerDes are used in various fields. This paper proposes a new method that accelerates the tree-to-TLV serialization through 2-way parallel processing called "parallelized serialization" and "parallelization with streaming". Experimental results show that parallelized serialization with 4 worker threads achieves a 1.97 fold shorter serialization time than when using a single worker thread, and the combination of 2-way parallel processing achieves a 2.11 fold shorter output time than that for ProtoBuf when 4 worker threads, File Output Stream and trees of 10,080 container nodes are used.

    KW - Multi-Core

    KW - Parallel Processing

    KW - Serialization

    UR - http://www.scopus.com/inward/record.url?scp=84983121041&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84983121041&partnerID=8YFLogxK

    U2 - 10.1109/PCCC.2014.7017059

    DO - 10.1109/PCCC.2014.7017059

    M3 - Conference contribution

    AN - SCOPUS:84983121041

    VL - 2014-January

    BT - 2014 IEEE 33rd International Performance Computing and Communications Conference, IPCCC 2014

    PB - Institute of Electrical and Electronics Engineers Inc.

    ER -