Intel Parallel Studio XE 2020 Cluster Studio
As the relentless pursuit of compute capacity of HPC systems have shifted to multicore systems, and soon to mixed multicore and many-core systems, the compute power of the hardware is scaling at faster than 2x Moore’s Law. Software development solutions and tools need to scale to meet the developers needs to utilize these systems.
Intel Parallel Studio Cluster Edition meets the challenges facing HPC developers by providing, for the first time, a comprehensive suite of tools that enables developers to boost HPC application performance and reliability. It combines Intel’s proven cluster tools with Intel’s advanced threading/memory correctness analysis and performance profiling tools to enable scaling application development for today’s and tomorrow’s HPC cluster systems.
Intel Parallel Studio XE Cluster Edition has everything you need to build, analyze, and scale your applications, then do it all again for continual optimization. Start compiling and optimizing your code today using all of the tools and libraries in this suite.
The Intel Parallel Studio suite includes 10+ software development tools to help you build fast, scalable, reliable applications for enterprise, cloud, HPC, and AI. It comes with a bevy of benchmarks to prove it. This comprehensive tool suite simplifies the creation and modernization of code, and accelerates workloads with the latest techniques in vectorization, multi-threading, multi-node, and memory optimization.
Intel Parallel Studio Cluster Edition includes the next-generation software development tools:
- Intel MPI Library - performance - optimised MPI Library with sustained scalability: low latencies, higher bandwidth, and increased processes.
- Intel Trace Analyzer and Collector - powerful MPI communications profiling and analysis with low overhead and effective visualization.
- Intel MPI Benchmarks - perform a set of MPI performance measurements for point-to-point and global communication operations for a range of message sizes.
- Intel Cluster Checker - Verify that cluster components work together seamlessly for improved uptime, productivity, and lower total cost of ownership.
- Intel Inspector - quickly find memory leaks and memory allocation errors, locate difficult-to-find threading errors like data races and deadlocks and detect out of bounds accesses and dangling pointers.
- Intel VTune Profiler - collect a rich set of data to tune CPU and GPU compute performance, multicore scalability, bandwidth, and more. Sort, filter, and visualise results for quick insight into performance bottlenecks and automate regression tests and collect data remotely using the powerful command line.
- Intel Advisor - analyse, design, tune, and check your threading design before implementation, explore and test threading options without disrupting normal development and predict thread errors and performance scaling on systems with more cores.
- Intel C++ Compiler - industry-leading C and C++ application performance with built-in advanced parallelism support with OpenMP support.
- Intel Fortran Compiler - industry-leading Fortran application performance with extensive support for Fortran standards, OpenMP, and more.
- Intel Distribution for Python - Improves Python applications and speed up core computational packages with this performance-oriented distribution.
- Intel Math Kernel Library (MKL) - extensive C, C++, and Fortran compatible math library that uses standard APIs for simple code integration.
- Intel Threading Building Blocks - widely used C++ template library with rich set of components to efficiently implement higher-level, task-based parallelism.
- Intel Integrated Performance Primitives (IPP) - optimised C++ library of software functions for multimedia processing, data processing, and communications applications.
- Intel-provided Debug Solutions - based on GNU GDB for debugging on CPU, Intel GUP and FPGA.
- Intel Data Analytics Acceleration Library (Intel DAAL) - a library which helps reduce the time it takes to develop high-performance data science applications.
Highlights of Intel Parallel Studio Cluster Edition
Superior shared, distributed, or hybrid application performance through industry leading Intel compilers, parallel models and libraries with advanced performance optimizations for today’s multicore and tomorrow’s many-core processors in HPC clusters.
- MPI Latency - Intel MPI Library is up to 6.5X as fast as alternative MPI libraries
- Compiler Performance –Industry leading Intel C, C++ & Fortran compilers
- Thread Profiling & Tuning – Intel VTune Amplifier XE now MPI enabled for every node
Intel Cluster Studio XE provides the tools, programming models, and performance libraries that enable developers to develop code that scales on Intel Xeon Processors today while easily extending to the Intel® Many Integrated Core (Intel MIC) architecture.
- MPI Capacity – Intel MPI Library scales beyond 90k processes
- Parallel Programming Models – Commercially supported Intel versions of open source Intel® Threading Building Blocks (Intel® TBB) and Intel Cilk Plus for threading parallelism
The impact of budget and schedule pressure makes it crucial to have the right tools and
programming models to rapidly develop and deploy reliable HPC applications. Intel Cluster Studio
XE delivers powerful threading and correctness tools for hybrid applications development and
parallel programming models that are simple to adopt.
- Thread & Memory Correctness– Intel Inspector XE now MPI enabled for every node
- MPI Correctness– Increased productivity in finding MPI errors.
- Rapid Performance Profiling – Intel VTune Amplifier finds hotspots faster
- Parallel Programming Models – Parallelize code using three keywords with Intel® Cilk™ Plus and Intel TBB templates for simple and robust scalable task based parallelism
What’s new in Intel Parallel Studio XE 2020
Modernize Code for Performance, Scalability, and Portability
- Write applications that scale with improved parallel performance on the latest Intel® Xeon® and Intel® Core™ processors using Intel® Advanced Vector Extensions 512 (Intel® AVX-512) instructions.
- Vectorize and thread your code using OpenMP* to take advantage of the latest SIMD-enabled hardware, including Intel AVX-512.
- Speed up AI inferencing with compilers from Intel, Intel® Performance Libraries, and analysis tools that support Intel® Deep Learning Boost with Vector Neural Network Instructions (VNNI) in 2nd generation Intel® Xeon® Scalable processors.
- Develop for large memories of up to 512 GB DIMMs with persistence. Identify, optimize, and tune Intel platforms for Intel® Optane™ DC persistent memory using Intel® VTune™ Profiler.
- Use extended coarse-grain profiling with platform-level collection and analysis in Intel VTune Profiler to understand and optimize platform configuration for applications.
- Perform cache simulation insights for vectorization with roofline analysis for L1, L2, L3, and DRAM in Intel® Advisor.
- Support for the HPC cloud helps you take advantage of the AWS* Parallel Cluster* and AWS Elastic Fabric Adapter for low-latency, high-bandwidth communications for MPI applications with the Intel® MPI Library.
- Supports an additional operating system: Amazon Linux 2*.
- Stay up to date with the latest standards and integrated development environments (IDE):
- Full C++ 2017 support with initial C++20 support
- Full Fortran 2008 and expanded Fortran 2018
- Python 3.7 support
- Full OpenMP 4.5 and initial OpenMP 5.0 draft
- Microsoft Visual Studio* 2019 integration
Flexibility for Your Needs
Access the latest Intel® Performance Libraries and Intel Distribution for Python via conda*, PIP*, Docker Hub*, APT GET, YUM, and NuGet.
Following matrix explains which products that are included in the different product editions
Systems based on IA-32 architecture are supported as target platforms on Windows* and Linux*. Systems based on Intel® 64 architectures below are supported both as host and target platforms.
Systems based on Intel® 64 architecture:
• Intel® Core™ processor family or higher
• Intel® Xeon® E5 v5 processor families recommended
• Intel® Xeon® E7 v5 processor families recommended
NOTE: It is assumed that the processors listed above are configured into homogeneous clusters.
Disk Space Requirements
12 GB of disk space (minimum) on a standard installation. Cluster installations require an
additional 4 GB of disk space.
NOTE: During the installation process, the installer may need up to 12 GB of additional temporary disk storage to manage the intermediate installation files.
Operating System Requirements
The operating systems listed below are supported by all tools on Intel® 64 Architecture. Individual tools may support additional operating systems and architecture configurations. See the individual tool release notes for full details.
• Amazon Linux 2
• Clear Linux
• Debian* 9.x, 10.x
• Fedora* 30.x
• Red Hat Enterprise Linux* 7.x, 8.x (equivalent CentOS versions supported, but not separately tested)
• SUSE Linux Enterprise Server* 12.x, 15.x
• Ubuntu* 16.04, 18.04, 19.04
• Microsoft* Windows* 10
• Microsoft* Windows* Server 2016, 2019
• macOS* 10.14, 10.15
The Intel MPI Library and Intel® Trace Analyzer and Collector are supported on Intel® Cluster Ready systems and HPC versions of the listed versions of Microsoft* Windows* Server. These tools are not supported on Ubuntu non-LTS systems.
Intel VTune Profiler, Intel Advisor, and Intel Inspector graphical user interfaces may require newer operating system versions. Please see their respective Release Notes documents for details.
Installation on IA-32 hosts is no longer supported by any tools.
2 GB RAM (minimum)
Additional Software Requirements
Development for a 32-bit target on a 64-bit host may require optional library components
(ia32-libs, lib32gcc1, lib32stdc++6, libc6-dev-i386, gcc-multilib, g++-multilib) to be installed from your Linux distribution.
On Microsoft Windows* OS, the Intel® C/C++ Compiler and Intel® Fortran Compiler require a version of Microsoft Visual Studio* to be installed. The following versions are currently supported:
• Microsoft Visual Studio* 2017, 2019
• Microsoft Visual Studio Express* (only for command line compilation)
On macOS*, the Intel® C/C++ Compiler and Intel® Fortran Compiler require a version of Xcode* to be installed. The following versions are currently supported:
• Xcode* 10.x, 11.x
A 64-bit operating system host is required to use the Intel® VTune™ Profiler graphical user interface to analyze collected profile data. Command line profiling and reporting is supported on a 32-bit operating system host. On Linux*, the following packages must be installed for the
Intel® VTune™ Profiler GUI:
• X.Org (v1.0 or higher, v1.7 or higher is recommended)
• X.Org X11 libXss runtime library
• Network Security Services library (v3.22 or higher)
• ALSA library
Intel Parallel Studio XE is offered as a perpetual license with no timeout. 1year support and maintenance is always included in the initial purchase. Two licensing models are available:
- Named user is a system-locked license, and products are priced per named user.
- Floating licenses can be shared by multiple users simultaneously on several systems, managed from a licensing server. Two- or five-seat licenses are available. When a license is released from one user, another user can request it.
Student and academic licenses are offered seperately.
Please contact us today for an offer!
All product updates and Intel Premier Support services are included for one year. Intel Support gives you secure, web-based, engineer-to-engineer support.
Use following links to get access to Intel Support
Alfasoft offers startup support to customers.