Build Performance-Optimized Code


Implement efficient, elegant code for system and IoT device applications from the start.

Every edition of Intel® System Studio includes these foundational tools:

benchmark of system performance
Complete specification for the test configuration.1

Intel® C++ Compiler

Plugging right into popular development environments like Eclipse* and Microsoft Visual Studio*, Intel® C++ Compilers are source- and binary-compatible with other compilers, such as Visual C++* for Windows* and GNU Compiler Collection (GCC) for Linux*, macOS*, and Android*.

benchmark of scalable performance on an Intel Xeon processor
Complete specification for the test configuration.2

Intel® Threading Building Blocks (Intel® TBB)

This is a widely used C++ library for shared-memory parallel programming and heterogeneous computing (intra-node distributed memory programming). It's ideal for optimizing IoT solutions to take advantage of multicore architectures.

Training

benchmark comparing Intel Integrated Performance Primitives to L Z 4 and L Z O libraries
Complete specification for the test configuration.3

Intel® Integrated Performance Primitives

This library is a developer's one-stop shop for image, signal, and data processing applications. Ready-to-use and royalty-free, it provides high-quality, production-ready, low-level building blocks to help you tune applications and get the best performance on a given platform.

Training

benchmark of fast Fourier transforms on an Intel Core processor
Configuration information. 4

Intel® Math Kernel Library

Accelerate math processing and neural network routines that increase application performance and reduce development time. This ready-to-use, royalty-free library includes highly vectorized and threaded linear algebra and fast Fourier transforms (FFT), as well as deep neural network, vector math, and statistics functions.

Training

benchmark comparing Intel Data Analytics Library to Apache Spark
Complete specification for the test configuration.5

Intel® Data Analytics Acceleration Library

Purpose-built for machine learning and analytics on your edge devices, this library addresses all stages of the data analytics pipeline: preprocessing, transforming, analyzing, modeling, validating, and decision-making.

diagram showing the relationship of O T gateways to edge and cloud devices

IoT Connection Tools

Remove the complexities of connecting sensor-to-device and device-to-cloud with two open-source connection tools. Includes cloud connectors and Eclipse* plugin modules. Each one includes:

  • MRAA: This low-level library simplifies the logic for connecting to different sensor pins, and allows developers to program in C++, Python*, JavaScript*, or Java*.
  • UPM: This high-level library makes it easy to connect and use sensors and actuators in IoT solutions. With over 400 supported sensors and modules, developers can quickly build and deploy custom IoT solutions.

Discover More Capabilities

Analyze

Find out more about robust tools that help you analyze and optimize what you built.

Debug

Learn more about Intel® System Studio 2018 Ultimate Edition—powerful tools to help ensure system and application stability, performance, and market success.

Ready to Purchase?

Take advantage of Priority Support. For one year, connect directly with Intel's engineers for answers to technical questions and to access older versions of the products.

Buy Now

Get a Free 90-Day Renewable Commercial License

Use Intel® System Studio with a free commercial license backed by community forum support. This license allows usage for 90 days, and can be refreshed an unlimited number of times to use the latest version. You can convert any time to a paid license, which provides Priority Support for one year.

Choose & Download

Benchmark results were obtained prior to the implementation of recent software patches and firmware updates intended to address exploits referred to as "Spectre" and "Meltdown". Implementation of these updates may make these results inapplicable to your device or system.

Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems, components, software, operations and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other products. For more information, see Performance Benchmark Test Disclosure.

1 Configuration: Intel® Core™ i5-7600 processor at 3.50 GHz, 16 GB RAM, Intel® Hyper-Threading Technology is not supported. Software: Intel® compilers 18.0.1, GCC 7.2.0. Clang with the LLVM compiler 5.0. Linux*: Red Hat Enterprise Linux* server release 7.2 (Maipo), kernel 3.10.0-514.el7.x86_64. Standard Performance Evaluation Corporation (SPEC)* benchmark suite. SmartHeap 10 was used for CXX tests when measuring SPECint* benchmarks. SPECint_rate_base_2006 (32 bit) compiler switches: SmartHeap* library 10 was used for C++ tests. Intel® C and Intel® C++ Compilers 18.0: '-m32 -xCORE-AVX2 -ipo -O3 -no-prec-div -qopt-prefetch -qopt-mem-layout-trans=3. C code adds option – static. GCC 7.2.0: '-m32 -Ofast -flto -march=core-avx2 -mfpmath=sse -funroll-loops. Clang 5.0: '-m32 -Ofast -march=core-avx2 -flto -mfpmath=sse -funroll-loops. C++ code adds the option: –fno-fast-math. SPECfp_rate_base_2006 (64 bit) compiler switches: Intel C and Intel C++ Compilers 18.0: '-m64 -xCORE-AVX2 -ipo -O3 -no-prec-div -qopt-prefetch -qopt-mem-layout-trans=3 -auto-p32. C code adds the option: –static. GCC 7.2.0: '-m64 -Ofast -flto -march=core-avx2 -mfpmath=sse -funroll-loops. Clang 5.0: '-m64 -Ofast -march=core-avx2 -flto -mfpmath=sse -funroll-loops. SPECint_speed_base_2006 (64 bit) compiler switches: SmartHeap library 10 was used for C++ tests. Intel C and Intel C++ Compilers 18.0: '-m64 -xCORE-AVX2 -ipo -O3 -no-prec-div -qopt-prefetch -auto-p32. C code adds options: –static –parallel. GCC 7.2.0: '-m64 -Ofast -flto -march=core-avx2 -mfpmath=sse -funroll-loops. C code adds the option: -ftree-parallelize-loops=4. Clang 5.0: '-m64 -Ofast -march=core-avx2 -flto -mfpmath=sse -funroll-loops. C++ code adds the option: -fno-fast-math -fno-fast-math. SPECfp_speed_base_2006 (64 bit) compiler switches: Intel C and Intel C++ Compilers 18.0, '-m64 -xCORE-AVX2 -ipo -O3 -no-prec-div -qopt-prefetch -static -auto-p32. C code adds the option: –parallel. GCC 7.2.0: '-m64 -Ofast -flto -march=core-avx2 -mfpmath=sse -funroll-loops. C code adds the option: -ftree-parallelize-loops=4. Clang 5.0: '-m64 -Ofast -march=core-avx2 -flto -mfpmath=sse -funroll-loops. Benchmark source: Intel Corporation

2 Configuration: Intel® Core™ i5-7600 processor, 1 x 4 core, 3.5 GHz, 6 MB, Intel® Smart Cache L3, 64 GB RAM, Operating System: Red Hat Enterprise Linux 7.2 FA; Intel® Math Kernel Library (Intel® MKL) 2018. Benchmark source: Intel Corporation

3 Configuration: Intel® Xeon® Platinum 81xx processor, Intel® Xeon® Platinum processor 8168 at 2.70 GHz, L3=33 MB, 2 x 24 cores with Intel Hyper-Threading Technology, Ubuntu* 64, 109 GB, Intel® compilers 18, Intel Performance Primitives 2018. Benchmark source: Intel Corporation

4 Configuration: Software versions: Intel® C++ Compiler and Intel® 64, version 17.4, Intel® Threading Building Blocks 2018 (Intel® TBB); Hardware: Intel® Xeon® CPU E5-2699 version 4 at 2.20 GHz 44/T, 128 GB main memory; Operating system: Red Hat Enterprise Linux Server* 7.2 (Maipo), kernel 3.10.0-327.4.5.el7.x86_64; Note: Sudoku, primes and tachyon are included with Intel TBB. Benchmark source: Intel Corporation

5 Configuration: Intel® Xeon® CPU E5-2660 at 2.60 GHz, 128 GB, Intel® Data Analytics Library (Intel® DAAL) 2018; Alternating least squares: – Users=1M Products=1M Ratings=10M Factors=100 Iterations=1 MLLib time=165.9 sec DAAL time=40.5 sec Gain=4.1x; Correlation – N=1M P=2000 size=37 GB Mllib time=169.2 sec DAAL=12.9 sec Gain=13.1x; PCA – n=10M p=1000 Partitions=360 Size=75 GB Mllib=246.6 sec DAAL (seq)=17.4 sec Gain=14.2x. Benchmark Source: Intel Corporation

Optimization Notice: Intel® compilers may or may not optimize to the same degree for non-Intel microprocessors for optimizations that are not unique to Intel microprocessors. These optimizations include SSE2, SSE3, and SSSE3 instruction sets and other optimization. Intel does not guarantee the availability, functionality, or effectiveness of any optimization on microprocessors not manufactured by Intel. Microprocessors-dependent optimizations in this product are intended for use with Intel® microprocessors. Certain optimizations not specific to Intel microarchitecture are reserved for Intel microprocessors. Please refer to the applicable product user and reference guides for more information regarding the specific instruction sets covered by this notice. Notice revision #20110804