Development of a Streaming Application Benchmark Set

نویسندگان

  • Georgia Giannopoulou
  • Wolfgang Haid
  • Kai Huang
  • Lothar Thiele
چکیده

Future requirements of embedded systems can only be fulfilled using multiprocessor systems. One challenge of these systems is to efficiently program applications. The DOL framework addresses this problem by allowing the specification of applications as Kahn process networks and by mapping these applications to multiprocessor architectures. This semester thesis proposes a set of benchmark applications which can be used to evaluate the peak computational performance and peak communication bandwidth of applications executing on several heterogeneous multiprocessor systems in which inter-processor communication is achieved via message passing. These applications have been designed as synchronous data flows and have been developed within the DOL framework. In a second step, several alternative configurations of them have been used to evaluate the performance of the Sony/Toshiba/IBM Cell Broadband Engine. To this end, the existing DOL code generation back-end for the CBE has been revisited and improved so that the corresponding run-time environment can be used as a reliable platform for experiments.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

SRBench: A Streaming RDF/SPARQL Benchmark

We introduce SRBench, a general-purpose benchmark primarily designed for streaming RDF/SPARQL engines, completely based on real-world data sets from the Linked Open Data cloud. With the increasing problem of too much streaming data but not enough tools to gain knowledge from them, researchers have set out for solutions in which Semantic Web technologies are adapted and extended for publishing, ...

متن کامل

Understanding and Optimizing Performance on Two “Stream” Architectures: VIRAM and Imagine

Methods for understanding and optimizing the performance of streaming applications are examined in the context of two prototype processors, VIRAM and Imagine. We compare their performance on a set of multimedia benchmarks (a subset of the EEMBC suite), as well as a simple vector addition microbenchmark. We demonstrate our approach for optimizing the performance of streaming applications using a...

متن کامل

Application of SCOR Model in an Oil- producing Company

Supply Chain Operations Reference (SCOR) model is developed and maintained by the Supply Chain Council (SCC). The model is a reference model which can be utilized to map benchmark and improve the supply chain operations. SCOR model provides companies with a basic process modeling tool, an extensive benchmark database by defining a set of supply chain metrics. This paper explains the process and...

متن کامل

Streaming data analytics via message passing with application to graph algorithms

The need to process streaming data, which arrives continuously at high-volume in real-time, arises in a variety of contexts including data produced by experiments, collections of environmental or network sensors, and running simulations. Streaming data can also be formulated as queries or transactions which operate on a large dynamic data store, e.g. a distributed database. We describe a lightw...

متن کامل

presentation of a two stages method to determine the suitable benchmark and return to scale (case study: girls high school of one zone shiraz city)

In this paper, a two stages method to determine suitable benchmark and return scale of the decision making units set is presented. At first, all of the efficient reference set in no radial data envelopment analysis (DEA) based on linear programming is found. first, RAM model is introduced and units is investigated using this model, then, to run the given algorithm below steps is performed. At t...

متن کامل

Fuzzy Data Envelopment Analysis for Classification of Streaming Data

The classification of fuzzy uncertain data is considered one of the most challenging issues in data analysis. In spite of the significance of fuzzy data in mathematical programming, the development of the analytical methods of fuzzy data is slow. Therefore, the current study proposes a new fuzzy data classification method based on fuzzy data envelopment analysis (DEA) which can handle strea...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2010