Gjesteforelesning: " cphVB: A forward compatible bytecode (mostly) for scientific applications " - Professor Brian Vinter
Professor Brian Vinter, Leader of the e-Science Centre at the Faculty of Science at the University of Copenhagen, vil holde en gjesteforelesning fredag 25. mai 2012 kl 11:15. Sted: Lille auditorium, Realfagsbygget, UiT.
Kart over Campus:
( http://www2.uit.no/ikbViewer/page/campustromso)
Alle interesserte er hjertelig velkomne.
Påmelding er ikke nødvendig.
Vel møtt!
----------------------------------------------------------------------
Date: Friday May 25th 2012, 11:15
Place: Lille Auditorium, Realfagsbygget, UiT
Title: " cphVB: A forward compatible bytecode (mostly) for scientific applications "
Lecturer: Professor Brian Vinter, Leader of the e-Science Centre at the Faculty of Science at the University of Copenhagen
Abstract
Recent years have provided a wealth of projects showing that using Python for scientific applications outperforms even popular choices such as Matlab. A major factor driving these successes is the efficient utilization of multi-cores, GPUs for general-purpose computation and scaling computations to clusters.
However, often these advances sacrifice some of the high-productivity features of Python by introducing new language constructs, enforcing new language semantics and/or enforcing explicit data types. The result is that the users will have to rewrite existing Python applications to utilize the Python extension.
In order to utilize GPGPUs in Python, a popular approach is to embed CUDA/OpenCL code kernels directly in the Python application. The programming productively of the approach is better and more readable than C/C++ applications but it is still inferior to native Python code. Furthermore, the approach enforces hardware specific programming and thus requires intimate knowledge of the underlying hardware and the CUDA/OpenCL programming model.
Copenhagen Vector Byte Code (cphVB) strives to provide a high-performance back-end for Numerical Python (NumPy) without reducing the high-productivity of Python/NumPy. Without any involvement of the user, cphVB will transforms regular sequential Python/NumPy applications into high-performance applications. The cphVB runtime system is capable of utilizing a broad range of computing platforms efficiently, e.g. Multi-core CPUs, GPGPUs and clusters of such machines.
The implementation of cphVB consists of a bridge that translates NumPy array operations into cphVB vector operations. The bridge will send these vector operations to a Vector Engine that performs the actual execution of the operations. cphVB comes with a broad range of Vector Engines that are optimized to specific hardware architectures, such as multi-core CPUs, GPGPU and clusters of said architectures. Thus, cphVB provides a high-productivity, high-performance framework that support legacy NumPy applications without changing a single line of code.
------------
Professor Brian Vinter is leader of the e-Science Centre at the Faculty of Science at the University of Copenhagen and an expert in high performance computing with experience from CERN and several universities in the use of large amounts of data in physics, business, and the health industry. Brian often asks the question to the financial sector - What would you do if you had a 1000 times more processing power?
