VEESC: Virtual Execution Environments for Scientific Computing Workshop

计算机程序设计语言在科学生产力中起着核心作用。它们是领域专家和机器之间的接口。第一代科学语言是静态编译的计算机程序语言，如Fortran、C和c++。为了优化效率，它们以牺牲使用成本为代价与硬件保持密切联系。科学家们越来越多地选择动态语言，比如Perl、Python、R、Matlab、Mathematica、Maple和javascript。例如，生物信息学的教材包括许多Perl、Python和R的教材，却没有C、c++或Fortran的教材。但是，前沿科学方法的应用受到实施该方法的系统能力的限制。虽然动态语言一开始是本机库的粘合剂，但随着时间的推移，越来越多的实际计算是用这些语言表示的。但是，用户很快体验到这些语言的局限性：性能不足以处理计算密集型任务，空间开销妨碍操作大型数据集，信息安全和结果的可再现性必须得到解决。本次研讨会探讨了建立一个以下一代科学编程语言的虚拟执行基础设施为中心的科学软件创新研究所（S2I2）的必要性。有必要开发专门知识和动力，以鼓励开发面向生产力的科学编程语言的开源高性能实现。为此，VEESC研讨会汇集了虚拟机和编译器方面的专家，高级动态语言和相关系统的设计人员和实现者，以及科学领域最终用户社区的主要代表。

Computer programming languages play a central role in scientificproductivity. They are the interface between domain experts and themachine. The first generation of scientific languages were staticallycompiled computer programming languages such as Fortran, C and C++.Optimized for efficiency they stayed close to the hardware at the cost ofease of use. The languages of choice for scientists are increasinglydynamic ones like Perl, Python, R, Matlab, Mathematica, Maple, andJavaScript. Educational material in Bioinformatics, for example, includesmany textbooks for Perl, Python and R and none for C, C++ or Fortran. Butthe application of cutting-edge scientific methodology is limited by thecapabilities of the system in which it is implemented. While dynamiclanguages start out as glue for native libraries, with time, more and moreof the actual computations is expressed in those languages. But quicklyusers experience limitations of these languages: performance is not adequatefor compute intensive tasks, space overheads preclude manipulating largedatasets, and information security and reproducibility of results must beaddressed. This workshop explores the need for a Scientific Software Innovation Institute (S2I2) centered around Virtual Execution Infrastructures for Next Generation Scientific Programming Languages. There is a needto develop expertise and momentum to encourage development of open source high-performance implementations of productivity-oriented scientificprogramming languages. To this end, the VEESC workshop brings togetherexperts in virtual machines and compilers, designers and implementors ofhigh-level dynamic languages and related systems, and key representativesfrom end-user communities in the sciences.