标题: 为未来的云端运算环境设计与制作高效能软体与系统整合---总计画
High Performance Software and System Integration for Future Cloud Computing
作者: 杨武
YANG WUU
国立交通大学资讯工程学系(所)
关键字: 云端计算;二元码转译系统;LLVM 中间码;OpenCL;TLP-aware 暂存器管理;cloud computing;binary translation;LLVM IR;OpenCL;GPU;TLP-aware register allocation
公开日期: 2012
摘要: 云端计算(cloud computing)是目前非常被看好的计算环境,它有许多项让人看好的经济优势,在云端计算环境里,平行处理是电脑运作的常态,目前多数的电脑系统都属于多核心(multicores)、多处理器(multiprocessors)、多电脑(multicomputing)系统,这类的电脑系统都能平行执行工作,只是他们内部的平行架构发展于不同层次。我们也需要使用不同的技术,来充分发展各自系统的特长。在子计画一里面,我们针对在新的指令集上缺少应用程式对于设计的问题,设计及实作了一个以LLVM为基础的新的静态二元码转译系统。本计画设计及实作了一个以LLVM为基础的新的静态二元码转译系统,除了移植性高,更可以利用LLVM现有的分析以及优化技术来转译二元码。实验结果证明我们的静态转译器所产生的二元码能够的执行速度是QEMU的动态转译速度的3至64倍。在子计画二里面,我们在高阶程式语言的层次加入自发性的 actor (autonomous actors) 的模型。我们对云端计算系统作效能的实验依据各种物件模型(onventional object model, active object model,hybrid object model),根据这些结果提出一个新的程式语言CLAOD,也开发一个CLAOD的编译器。在子计画三里面,我们将含有 annotation LLVM 中间码 (VMIR),根据注解(annotation)转换成OpenCL 的程式码,并加入动态优化技术。主要议题有资料/工作平行性(data/task parallelism)专用之 annotation 的设计和支援 data/task parallelism annotation 之 OpenCL 后端(backend)之设计与实作。在子计画四里面,将集中研究与精进 OpenCL 系统,将 OpenCL 的程式码编译成适当的 GPU机器码,并加上适当的辅助程式码。现针对 OpenCL runtime的实作运行于系统虚拟化架构下Guestruntime library,并设计了在hypervisor中可能的GPU抽象化daemon以利 resource management并进行初步的测定与验证,以及提出针对GPU核心的TLP-aware register allocation。
Cloud computing gradually becomes a hot topic in the research community because it reflects the capabilities of current computer and network systems and users‘ expectations of this technology. Cloud computing offers significant economic advantages compared to the traditional private computer centers and private network connections. This advantage comes mainly from the sharing of computer and network resources.One characteristic of the cloud computing is that it provides much parallelism and heterogeneity. Our integrated project focuses on the systems software that can harness this inherent parallelism. Our integrated project consists of four subprojects:1. Build a hybrid (including both dynamic and static) binary translator that translates existing binary code into LLVM IR. The code generated by our translator can run 3 to 64 times faster than the one by QEMU on the EEMBC benchmark suite.2. Design and implement an actor-based concurrent languages, called CLOAD.3. Extract loops from LLVM IR (with the help of programmer-supplied annotations) and translate them into OpenCL kernel functions so that they can execute concurrently.4. Develop scheduling and management tools for the OpenCL programming system.The four subprojects together will create a comprehensive program development environment that intends to fully utilize the heterogeneous computing and communicating resources.
官方说明文件#: NSC100-2218-E009-009-MY3
URI: http://hdl.handle.net/11536/98845
https://www.grb.gov.tw/search/planDetail?id=2392254&docId=380559
显示于类别:Research Plans