英特尔 VTune 性能分析器 Linux 版

• 平台灵活性 – 支持的英特尔处理器，包括英特尔® 安腾® 2、64 位英特尔® 至强®、英特尔® 奔腾® 4 处理器，以及采用英特尔® 迅驰® 移动计算技术§ 的英特尔® 奔腾® M 处理器组件。

• 这套*基于 Linux 的解决方案在 IA-32 与“英特尔 EM64T”上提供 Eclipse† 图形用户界面，在 IA-32、含“英特尔 EM64T”技术的处理器以及基于安腾的系统上提供强大的命令行界面，具有*的灵活性。基于安腾的平台上自带的 Eclipse† 图形用户界面将于 2006 年早些时候推出。

• 系统范围内基于事件的采样使开发人员用几乎可以忽略不计的开销换来对软件实际性能的最准确表示。

• 调用图评测用图形表示程序流程，帮助快速确定关键的函数与调用序列，从高层次的算法角度了解程序的执行情况。

• 命令行功能给开发人员提供了一种灵活的手段，通过多种可选择的脚本语言（包括 PERL*）来采集采样与调用图数据。

• 查找瓶颈变得地轻松。 在一个简单的对话框中，使用列表列出五个耗时的函数。单击函数名称即可显示源代码以及占用最多处理器时间的代码。

• VTune 现支持 Linux 2.6 内核与 MontaVista Carrier Grade Linux。 安装过程已经大大简化，并能自动下载新内核的驱动程序。

• 对大型系统的支持比之以往更为强大，这是通过新增多用户“调用图”与每 CPU 采样缓冲技术实现的。最多可支持 4096 颗处理器。

• 远程分析现在更为安全。 没有相应访问权限的用户将不被允许访问或测评远程系统。

• 在安全的环境中进行性能调整变得异常轻松。“调用图”现在可以分析采用只读文件系统的安全系统。

• VTune 环境的显示布局已基于用户反馈进行了改善。 查找所需的信息现在变得更为轻松快捷。此外，新增的“入门”文档提供了一段简短的“手把手”培训教程，帮助用户更快提高工作效率。

• VTune 与 Eclipse† 3.1 主流开发环境集成，集成环境中包含上下文相关的提示窗格，利用重要的提示和说明帮助用户更快提高工作效率。

• VTune 支持的英特尔处理器，包括含“英特尔 EM64T”技术的处理器。

• 系统范围内基于事件的采样

• 调用图评测

• 功能丰富、易于使用的“图形用户界面”

• 热点分析

• 内存密集型应用支持

• 静态分析与反汇编

• VTune 性能分析器驱动程序包

• *基于 Linux 的解决方案，在 IA-32 上可以同 Eclipse 集成

• 支持的英特尔® 处理器，包括含“英特尔 EM64T”技术的处理器

• 多种编程语言

• Java 应用程序评测

• 命令行功能

• 对 Linux 2.6 内核与 MontaVista Carrier Grade Linux 的支持

All Architectures:

• Low Overhead
  Accurately identify where the program spends time. Sampling is system wide with negligible overhead (typically less than 5 percent).
• Find the Critical Path
  Determine function calling sequences and find the critical path using Call Graph.
• No Recompile Required
  Unlike traditional instrumented profilers that make you recompile or modify your build script, just use your production executables.
• Compatibility
  VTune™ Performance Analyzer supports the latest Intel® processors (Intel® 64 architecture-based processors, Intel® Itanium® processors, multi-core processors...) and a wide variety of Linux* distributions.
• Programming Language and Compiler Independent
  VTune analyzer supports all compilers that follow industry standards (ELF, STABS, DWARF).
• Mixed Java* and Native Code
  Unlike Java*-only analyzers, VTune analyzer tunes mixed Java and native code1.
• Minimal Memory Footprint
  Remote profiling minimizes the performance impact on the target system by running the user interface on a separate system.
• Commａnd Line Capability
  Automate batch operations.
• Large Applications Welcome
  VTune analyzer is a robust solution even with large executables2. If you have a large application with hundreds of thousands of functions, bring it to VTune analyzer.
• Listen to the Compiler's Advice
  An optimizing compiler can do a lot better with just a few tips from you. We've integrated the Intel®Compilers with VTune analyzer to make this easy and very effective.

Large Enterprise and HPC Systems:

• Minimize Traffic in Non-uniform Memory Architecture (NuMA) Systems by storing sampling data in local CPU memory. This is critical to avoid saturating the interconnect fabric and slowing the system under test.
• Designed for High Performance Computing
  Large High Performance Computing (HPC) systems have unique requirements supported by VTune analyzer.

Intel® Itanium® Architecture:

• Eclipse* Based Graphical User Interface
  The easy-to-use Eclipse* based graphical user interface in VTune analyzer is now native on Itanium® architecture.
• Instruction Filtered Events Pinpoint Bottleneck Locations
  Isolate problems like poor pre-fetch and poor memory alignment. Sometimes just choosing an event is not selective enough, because the event can occur both at critical and non-critical times. On Intel Itanium architecture, instruction filtering allows you to collect events only when they occur with a specified op-code.
• Minimize Data Collection with CPU Selection
  Collect only the data you need. CPU selection lets you control exactly where data is collected, from all the processors, only those in your allocation or only the processors you specify. This greatly reduces the amount of data you need to collect.