Top 5 reasons to use the Intel oneAPI HPC Toolkit
News|by Leanne Bevan|5 December 2022
With so many different developer solutions on the market, it can be hard to choose the right toolkit for your needs. Below we have listed the top 5 reasons why developers should use the Intel oneAPI HPC Toolkit to build, analyse, optimise and scale HPC applications.
1. A wide choice of compilers
The toolkit includes not one, but four compilers.
The first is the Intel C++ Compiler Classic (known as ICC). This is a standard-based C++ compiler that includes support for OpenMP. Developers can use this compiler to take advantage of more cores and built-in technologies in platforms based on Intel Xeon and Intel Core processors.
The second is the Intel oneAPI DPC++/C++ Compiler (ICX). DPC++ is the oneAPI implementation of Khronos SYCL. This compiler targets CPUs and accelerators through single-source code while permitting custom tuning.
- Allows code reuse across hardware targets, while permitting custom tuning for a specific accelerator.
- Delivers C++ productivity benefits, using common and familiar C and C++ constructs. SYCL supports data parallelism and heterogeneous programming such as GPUs and FPGAs.
- It boasts enhanced OpenMP* 5.1 compliance and improves the performance of OpenMP reductions for compute offload.
- The FPGA emulation and compilation enables experienced FPGA developers to target acceleration workloads to field-programmable platforms.
The third is Intel® Fortran Compiler Classic (known as ifort). The Intel® Fortran Compiler is built on a long history of generating optimised code that supports industry standards while taking advantage of built-in technology for Intel® Xeon® Scalable processors and Intel® Core™ processors. The best-in-class Fortran compiler. Use for production with existing F77, 90, 03, 08 codes; provides continuity with existing CPU-focused workflows.
The fourth is the new Intel® Fortran Compiler (IFX), based on modern LLVM technology, which adds support for parameterised-derived types, F2018 IEEE Compare, and VAX structures support and expands support for OpenMP 5.0 with Declare Mapper for scalars support. Provides CPU and GPU offload support.
2. Use the latest techniques to optimise your HPC apps
The Intel oneAPI HPC Toolkit enables you to take advantage of the following latest techniques to optimise your applications:
Vectorisation
As Intel’s Evgueny Khartchenko notes in his blog, vectorisation is the process of converting an algorithm from operating on a single value at a time to operating on a set of values (vector) at one time. Modern CPUs provide direct support for vector operations where a single instruction is applied to multiple data (SIMD).
For example, a CPU with a 512-bit register could hold 16 32- bit single precision doubles and do a single calculation.
16 times faster than executing a single instruction at a time. Combining this with threading and multi-core CPUs leads to orders of magnitude performance gains.
Multithreading
Intel explains that multithreading is a form of parallelisation or dividing up work for simultaneous processing. Instead of giving a large workload to a single core, threaded programs split the work into multiple software threads. These threads are processed in parallel by different CPU cores to save time.
Memory optimisation
As noted by an article on TotalView, memory optimisation is “a range of techniques related to improving computer memory, such as identifying memory leaks and corruption, to optimise memory usage and increase performance and application usability... With memory optimisation, memory resources are made more performant by resolving leaks and other difficult issues such as memory block overwrite and improper memory-API use.”
3. Analysis and Debug Tools
Intel® DPC++ Compatibility Tool
Minimises Code Migration Time.
Assists developers migrating code written in CUDA to SYCL once, generating human readable code wherever possible. ~90-95% of code typically migrates automatically. Inline comments are provided to help developers finish porting the application.
Intel® VTune™ Profiler
Profiling¾Tune for CPU, GPU & FPGA
Analyse SYCL code. See the lines of SYCL that consume the most time. Tune for Intel CPUs, GPUs & FPGAs. Optimise for any supported hardware accelerator. Optimise Offload. Tune OpenMP offload performance. Wide range of performance profiles. CPU, GPU, FPGA, threading, memory, cache, storage…
Flame graph display improves visualization of hot spots. Supports Popular Languages: SYCL, C, C++, Fortran, Python, Go, Java, or a mix.
Intel® Advisor
Design Assistant¾Design for Modern Hardware
- Offload Advisor: Estimate performance of offloading to an accelerator
- Roofline Analysis: Optimise CPU/GPU code for memory and compute
- Vectorisation Advisor: Add and optimise vectorisation
- Threading Advisor: Add effective threading to unthreaded applications
- Flow Graph Analyser: Create and analyse efficient flow graphs
Intel® Distribution of GDB
Data Parallel C++ Debug¾Heterogeneous Application Debug
- High-level language debug support
- Multiple accelerator support: Intel CPU, GPU, FPGA emulation
- Auto-detect accelerator architecture during application runtime
- Non-proprietary open-source solution based on GDB
4. The Base Toolkit is included
The Intel oneAPI Base Toolkit is a core set of tools and libraries for developing high-performance, data-centric applications across diverse architectures. It features an industry-leading C++ compiler that implements SYCL, an evolution of C++ for heterogeneous computing.
Domain-specific libraries and the Intel Distribution for Python provide drop-in acceleration across relevant architectures. Enhanced profiling, design assistance, and debug tools complete the kit.
Watch the overview video: What is the Intel oneAPI Base Toolkit?
5. Benefit from Priority Support
You can purchase the Intel oneAPI HPC (and other toolkits) with Priority Support.
Priority Support provides developers with the ability to submit questions, problems, and other technical support issues through direct and private interactions with Intel’s engineers. Rather than public forums where there could be delays in responses, false fixes, or perhaps no information available with the answer to the question you have.
Priority Support can be bought through Grey Matter, as we are an Intel Software Elite Reseller.
Watch the overview video: Intel Priority Support.
Want an Intel oneAPI HPC Toolkit free trial or quote?
Watch the overview video to learn more about the Intel oneAPI HPC Toolkit: What is the Intel oneAPI HPC Toolkit?
As an Intel Software Elite Reseller, Grey Matter can provide you with free trials, exclusive pricing, and licensing information about the Intel oneAPI Toolkits with Priority Support.
Fill out the form below and a member of our team will be in touch with more information.
Contact Grey Matter
If you have any questions or want some extra information, complete the form below and one of the team will be in touch ASAP. If you have a specific use case, please let us know and we'll help you find the right solution faster.
By submitting this form you are agreeing to our Privacy Policy and Website Terms of Use.
