Overview
The VTune(TM) Performance Analyzer 8.0 for Linux* product provides
graphical user interface and command-line capabilities that assist in
finding performance bottlenecks and hotspots in application code. It
enables you to collect, analyze, and display performance data for Linux*
applications running on IA-32, Intel(R) EM64T or Itanium(R) processor
based systems.
The VTune analyzer in the integrated Eclipse* platform environment
provides these data collection functionalities:
.Sampling: Basic, non-intrusive, system-wide profiling capability with
support for systems with IA-32, Intel(R) EM64T and Itanium(R)
processors, including multiprocessor systems.
.Call graph: Intrusive profiling capability for capturing the flow of
the application and the parent-child relationships with support for
systems with IA-32, Intel(R) EM64T or Itanium(R) processors.

What's New in Version 8.0 for Linux
The following are the main new features in the VTune(TM) Performance
Analyzer 8.0 for Linux*:
.The VTune analyzer is now integrated in the Eclipse 3.1 platform
environment for IA-32 architecture and systems with Intel(R) EM64T
processors.
.Authentication for remote sessions. Users without appropriate access
rights will not be allowed to access or profile the remote system.
.VTune analyzer's views enable viewing Activity results for IA-32
architectures, Itanium architectures and systems with Intel(R)
EM64Tprocessors. The following views are available: Call Graph,
Sampling and Source View. Source View can be invoked from Call Graph
or Sampling views.
.Support for the latest Intel(R) processors including Intel(R) EM64T.
.Profiling mixed architecture applications. For example: a user may
profile 32-bit applications on Intel(R) EM64T. Profiling applications
that contain both Intel(R) EM64T and IA-32 binaries is not supported
and each part of these applications should be profiled separately.
.Java* application profiling on systems with all supported Itanium(R)
processor-based Linux* distributions.
.Support for viewing and analyzing samples collected for up to 4096
processors.
.Support for CPU masking in sampling collection.
.Multi-User Call Graph functionality, enabling multiple users to work
on the same remote machine simultaneously.
.Call Graph on read-only file systems.
.Support for per CPU buffering in sampling collection, allowing users
of large scale systems to get higher sampling resolution. This feature
is useful for users who have experienced missing samples due to a
large number of CPUs at a desired sample after value.
.Automatic filtering of functions Activities with call graph data
collection, which have large results.
.Support for Event Ratios for sampling and call graph data collection.

Download File Size:191.63 MB

---