Logo des Repositoriums

Auflistung nach Schlagwort "accelerator"

1 - 2 von 2
  • Konferenzbeitrag
    An FPGA Avro Parser Generator for Accelerated Data Stream Processing
    (BTW 2023, 2023) Hahn, Tobias; Schüll, Daniel; Wildermann, Stefan; Teich, Jürgen
    Big Data applications frequently involve processing data streams encoded in semi-structured data formats such as JSON, Protobuf, or Avro.A major challenge in accelerating data stream processing on FPGAs is that the parsing of such data formats is usually highly complex.This is especially true for JSON parsing on FPGAs, which lies in the focus of related work.The parsing of the binary Avro format, on the other hand, is perfectly suited for being processed on FPGAs and can thus serve as an enabler for data stream processing on FPGAs.In this realm, we present a methodology for parsing, projection, and selection of Avro objects, which enforces an output format suitable for further processing on the FPGA.Moreover, we provide a generator to automatically create accelerators based on this methodology.The obtained accelerators can achieve significant speedups compared to CPU-based parsers, and at the same time require only very few FPGA resources.
  • Textdokument
    Partial Reload of Incrementally Updated Tables in Analytic Database Accelerators
    (BTW 2019, 2019) Stolze, Knut; Beier, Felix; Müller, Jens
    The IBM Db2 Analytics Accelerator (IDAA) is a state-of-the art hybrid database system that seamlessly extends the strong transactional capabilities of Db2 for z/OS (Db2z) with very fast column-store processing in Db2 Database for Linux, Unix, and Windows. IDAA maintains a copy of the data from Db2z in its backend database. The data can be synchronized in batch with a granularity of table partitions, or incrementally using replication technology for individual rows. In this paper we present the enablement of combining the batch loading of a true subset of a table’s partitions for replicated tables. The primary goal for such an integration is to ensure data consistency. A specific challenge is that no duplicated rows stemming from the two data transfer paths come into existence. We present a robust and yet simple approach that is based on IDAA’s implementation of multi-version concurrency control.