High-performance parallelisation of real-time applications
نویسندگان
چکیده
This paper presents an overview of theP-SOCRATES methodology and tools, instantiated in theUpScale SDK (Software Development Kit) for the development oftime-predictable high-performance applications. The proposedmethodology was designed to provide an integrated SDK to fullyexploit the huge performance opportunities brought by the mostadvanced many-core processors, whilst ensuring a predictableperformance and maintaining (or even reducing) developmentcosts of applications. The paper also provides the performanceresults of the application of the SDK in relevant embedded usecases.
منابع مشابه
Large drilling machine control code - Parallelisation and WCET speedup
Hard real-time applications in safety-critical domains – namely avionics, automotive, and machinery – require high-performance and timing analysability. We present research results of the parallelisation and WCET analysis of an industrial hard real-time application, i.e. the control code of a large drilling machine from BAUER Maschinen GmbH. We reached a quadcore speedup of 2.62 for the maximum...
متن کاملA Mobile and Fog-based Computing Method to Execute Smart Device Applications in a Secure Environment
With the rapid growth of smart device and Internet of things applications, the volume of communication and data in networks have increased. Due to the network lag and massive demands, centralized and traditional cloud computing architecture are not accountable to the high users' demands and not proper for execution of delay-sensitive and real time applications. To resolve these challenges, we p...
متن کاملTowards Automatic Parallelisation for Multi-Processor DSPs
This paper describes a preliminary compiler based approach to achieving high performance DSP applications by automatically mapping C programs to multi-processor DSP systems. DSP programs typically contain pointer based memory accesses making automatic parallelisation difficult. This paper presents a new method to convert a restricted class of pointer-based memory accesses into array accesses wi...
متن کاملAn interactive environment to assist in the parallelisation of Fortran application codes
Introduction The cost in porting applications to high performance parallel computers still remains a very expensive effort. The shared memory and distributed memory programming models are two of the most popular models used to transform existing serial application codes to a parallel form. Despite the error-prone and costly effort involved in the parallelisation process, the use of message pass...
متن کاملDiscovering and exploiting parallelism in DOACROSS loops
Although multicore processors have been the norm for a decade, programmers still struggle to write parallel general-purpose applications, resulting in underutilised on-chip resources. Automatic parallelisation is a promising approach to improving the performance of such applications without burdening the programmer. I explore various techniques for automatically extracting parallelism which spa...
متن کامل