Intel® Software Development Products – Customer Comments

Intel® Advisor XE

Vickery Research Alliance: "Intel got the product name correct... 'Advisor' not 'Wizard'. Parallelization is a complex problem – especially when dealing with existing codes – and Intel® Advisor XE can be invaluable in developing the understanding required. Intel® Advisor XE assists with identifying opportunities, designing tests, modeling scenarios and revealing flaws."

"Intel got the product name correct... Advisor not "Wizard". Parallelization is a complex problem – especially when dealing with existing codes – and Intel® Advisor XE can be invaluable in developing the understanding required."

"Intel got the product name correct... Advisor not "Wizard". Intel® Advisor XE can be invaluable in developing the understanding required to parallelize existing code."

"Intel® Advisor XE can be invaluable in developing the understanding required to parallelize existing code."

"Parallelization is a complex problem. Intel® Advisor XE assists with identifying opportunities, designing tests, modeling scenarios and revealing flaws."

"Intel® Advisor XE assists with identifying opportunities, designing tests, modeling scenarios and revealing flaws."

"Insight is the essential ingredient for a successful parallelization effort. Intel® Advisor XE is a highly effective exploration tool that aids in developing the understanding required from identifying opportunities to evaluating the correctness of various implementations."

"Intel® Advisor XE is a highly effective exploration tool that aids in developing the understanding required for a successful parallelization effort."

"Intel® Advisor XE is a highly effective exploration tool for a successful parallelization effort."

"'Correctness' remains the fundamental value in software and validation is essential. By its nature, moving from serial to concurrent algorithms can lead to valid but 'different' results. Different breeds of errors can be introduced. Intel® Advisor XE can help you identity and repair flaws and subtle side effects in a given parallelization strategy before moving to implementation and testing."

"Intel® Advisor XE can help you identity and repair flaws and subtle side effects in a given parallelization strategy before moving to implementation and testing."

"Intel® Advisor XE can help you identity and repair flaws in a parallelization strategy before moving to implementation and testing."

Matt Osterberg
Vickery Research Alliance

Ufa State Aviation Technical University: "Experimenting with different parallelization techniques becomes much faster and more convenient with Intel Advisor XE. Before we started using it, many students spent the whole lab hour, unsuccessfully trying to parallelize "hopeless" locations. But with Intel Advisor XE we completely solved this problem. Suitability and Correctness analysis allowed students to quickly cut off wrong directions and concentrate on locations worth parallelization."

"Experimenting with different parallelization techniques becomes much faster and more convenient with Intel Advisor XE."

"Before we started using Intel Advisor XE, many students spent the whole lab hour, unsuccessfully trying to parallelize "hopeless" locations. But with Intel Advisor XE we completely solved this problem."

"Experimenting with different parallelization techniques becomes much faster and more convenient with Intel Advisor XE. Before we started using it, many students spent the whole lab hour, unsuccessfully trying to parallelize "hopeless" locations. But with Intel Advisor XE we completely solved this problem."

"Intel Advisor XEs Suitability and Correctness analysis allows students to quickly cut off wrong directions and concentrate on locations worth parallelization. Experimenting with different parallelization techniques becomes much faster and more convenient."

"It's powerful to have the performance and correctness analysis combined together in Intel Advisor XE. Experimenting with different parallelization techniques is much faster and more convenient."

"At our labs we work with tasks that can be parallelized rather well . As usual students are achieving speedup about 3.5x on 4 cores. It is important that students can estimate potential speedup using Suitability analysis in Intel Advisor XE and get corresponding results after manual parallelization."

"Students can estimate potential speedup using Intel Advisor XE's Suitability analysis and get corresponding results after parallelization."

Arthur Yuldashev
Senior Lecturer
Ufa State Aviation Technical University

Harmonic Inc.: a) The video processing and encoding libraries we develop need to be fast, error free, and take advantage of a variety of hardware platforms. The Intel® Amplifier XE and Intel® Inspector XE applications have dramatically sped up our ability to identify bottlenecks, find/fix memory problems and track down difficult to isolate threading errors before our packages are released to the field.

b) The Intel® Amplifier XE and Intel® Inspector XE applications have dramatically sped up our ability to identify bottlenecks, find/fix memory problems and track down difficult to isolate threading errors before our packages are released to the field.

c) The video processing and encoding libraries we develop need to be fast. Intel® Amplifier XE has dramatically sped up our ability to identify bottlenecks before our packages are released to the field.

d) Intel® Amplifier XE has dramatically sped up our ability to identify bottlenecks before our packages are released to the field.

e) The video processing and encoding libraries we develop need to be error free and take advantage of a variety of hardware platforms. Intel® Inspector XE has dramatically sped up our ability to find/fix memory problems and track down difficult to isolate threading errors before our packages are released to the field.

f) The video processing and encoding libraries we develop need to be error free. Intel® Inspector XE has dramatically sped up our ability to find/fix memory problems and track down difficult to isolate threading errors before our packages are released to the field.

g) Intel® Inspector XE has dramatically sped up our ability to find/fix memory problems and track down difficult to isolate threading errors before our packages are released to the field.

Peter von Kaenel
Director, Software Development
Harmonic Inc.

back to top

Intel® Cluster Studio XE

RIKEN: "Intel Trace Analyzer and Collector for Linux helped to drastically improve the performance of RIKEN's molecular dynamics cluster software. We were able to shorten MPI communication time by half by finding and removing bottlenecks with non-blocking communication patterns. Since Intel Trace Analyzer and Collector can embed instrumentation into the program, we can tell the execution time of each function and its load balance, which enabled us to very easily understand where to optimize. Intel's MPI library and Cluster tools provide us the best cluster development environment."

Dr. Takhiro Koishi
RIKEN

Univ of Seoul: "We evaluated the cluster tools and concluded that we would get much better performance on the Intel platform with the Intel software tools. For each application, Intel's compiler gave us better performance than before and using the Intel cluster tools let us know how the cluster system ran and how to improve it."

Jinwoo Park
Senior System Manager
Univ of Seoul

EM Software & Systems: "Our FEKO electromagnetics code runs on various parallel environments (e.g., clusters of different sizes, shared memory machines) and utilizes complex numerical algorithms that are non-trivial to parallelize. Intel Trace Analyzer and Collector have proven to be very valuable tools to help understand FEKO parallel communication patterns and consequently in optimizing the message passing call that result in an extremely well performing electromagnetics ISV cluster application."

Dr. Ing Ulrich Jakobus
Technical Director
EM Software & Systems

Flow Science, Incorporated: "Flow-3D's unique advantage is its ability for modeling complex fluid flows. As such, it is difficult to enable the parallel performance demanded by our customers. To meet the demand, we actively use the full functionality of Intel Cluster Studio XE to reduce and find previously undetectable shared and distributed memory errors, and improve the overall performance and scaling of our software on the different multicore architecture systems used by our customers. In addition to the development benefits, the Cluster Studio XE tools help us to solve un reproducible issues that occur at customer sites."

Dr. Anup Gokarn
Senior Developer
Flow Science, Incorporated

S and I Engineering: "The ability to simulate complex geometries with relative ease and the use of a matrix-free implicit procedure resulting in rapid convergence to steady state makes the solver both efficient and robust."

Dr. Nikhil V Shende
Director
S and I Engineering

Altair: "CFD simulations are notoriously computationally intensive. Along with robustness and accuracy, great parallel performance and scalability are keys for solving large CFD problems. These capabilities can be further realized with well-architected, high performance computing hardware such as Intel's— contributing to faster time-to-market."

Farzin Shakib
Vice President, CFD Technology
Altair

Flow Science: "Intel has provided us with hardware and software for testing purposes. We've had visits from Intel personnel to demonstrate he use of tools that would help our software. Intel technical support has been great in responding to questions in a timely ashion."

Anup Gokarn
Senior developer
Flow Science

Altair RADIOSS: "For the first time, we are able to port RADIOSS explicit on a coprocessor. This would have been far too difficult on a GPU, but is now possible given the capabilities of the Intel Xeon Phi coprocessor."

Eric Lequiniou
Director of High Performance Computing
Altair RADIOSS

S and I Engineering: "Fast and accurate state of the art general purpose CFD solvers is the focus at S and I Engineering Solutions Pvt, Ltd. Scalability and effeciency are key to us when it comes to our choice and use of MPI Libraries. The Intel MPI Library has enabled us to scale to over 10k cores with a high efficiency and performance."

Dr. Nikhil V Shende
Director
S and I Engineering

back to top

Intel® Compilers

Schneider Electric: "I was very impressed with the seamless integration of the Intel® C++ Compiler, a component of Emberson, into the Yocto Project* cross-build environment and the Eclipse* CDT. I was productive within minutes."

Rajendra Alluri
Technical Lead
Schneider Electric

River Silica Technologies: "Our software build time was significantly reduced when using Intel® C++ Compiler compared to my GCC* compiler."

"I'm a C++ developer and have high praise for the Intel® C++ Compiler. I got a nice performance boost compared to my GCC* compiler right out of the box – nice to see."

"The automated Eclipse* CDT integration of Intel® C++ Compiler worked flawlessly and made me productive fast."

Jagadish Kamath
Co-founder and Software Architect
River Silica Technologies

Macraigor Systems: "The integration of Intel® C++ Compiler with Eclipse* CDT worked very well and allowed me to be productive fast."

James MacGregor
Managing Partner
Macraigor Systems

NYU: "I especially like the C++ x support in the updated Intel® C++ compilers that ship in Intel® C++ Composer XE. These are working extremely well and I am impressed by the number of new features you were able to bring to the Intel C++ Compiler. I tested variadic templates, function template defaults, alternative function return syntax, named/typed enumerations, and more. One feature that I was happy to see implemented was 'SFINAE for expressions'. For library writers, this is a very powerful feature and can greatly simplify the mess of template meta-programming that is currently needed for cutting edge libraries. Keep up the good work with your excellent, and stable, compilers!"

Jesse Perla
PhD Candidate
NYU

Univ of Texas at Austin: "I recently tried the updated Intel® Cilk™ Plus in the Intel® C++ Compiler (part of Intel® Composer XE). I liked the lower overhead from using Cilk Plus spawning compared to that of OpenMP* task. I'm looking forward to using Cilk Plus Array Notations . The concepts behind Cilk Plus – simplification of adding parallelism – is really great. Thanks for offering this easy-to-use capability!"

David Carver
Univ of Texas at Austin

Sugars Int'l LLC: "I use the Intel® Fortran compiler in Intel® Fortran Composer XE, and I very much like the new Object Browser. It makes it much easier to navigate in modules that have many routines. The compiler is stable and offers outstanding performance. We are a small company but big proponents of Intel Fortran!"

Warner Weiss
Manager
Sugars Int'l LLC

Multipath Corporation: "Intel makes great C++ and Fortran compilers as well as math libraries. As currently offered in the Intel® C++ Composer XE 2011 and Intel® Fortran Composer XE 2011, they are stable and performance-oriented. I installed them and was successful 'out of the box,' so to speak, on both Windows and Linux. I recommend these feature-rich compilers to anyone who is serious about serial or parallel performance in their application software."

Dr. Ronald C. Young
President
Multipath Corporation

University of Pennsylvania: "Intel® Fortran performs beautifully – works like a charm on my iMac. The performance is great and the compiler is very stable. Thank you for this compiler!"

Peter Hahn
Adjunct Assoc. Professor
University of Pennsylvania

University of California Berkeley: "I have never seen a compiler so good, reliable and robust as your Intel Fortran compiler. It's so good that I'd never want to switch to any other compiler out there."

Louis Bouchard
Materials Sciences Division,
Lawrence Berkeley National Labs &
University of California Berkeley

Philips Medical: "At Philips Medical, we develop software for some of our medical diagnostic products using the Intel C++ Compiler. We found the performance of our applications to be excellent and were pleased to find that we could use the Intel compiler on both Intel and non-Intel 32 and 64-bit processors in combination with different Microsoft compilers. The main features of the compiler that we rely on are the support and optimized code generation for intrinsics (MMX, SSE, SSE2), automatic vectorization, and OpenMP support. We also appreciated the responsiveness to questions and issues related to the compiler. The online support makes it easy to post issues and we were kept informed about when issues would be resolved. Regardless of the processor, we found that the Intel compiler offered the best all-around performance."

Jonas Olsson
Design Engineer
Philips Medical

Paracel: "The Intel compiler generated faster code than other compilers for most of our tests on both ia32 and x86_64 platforms, which helps us deliver the performance our customers demand. It is the only compiler that we've seen that is able to perform multi-file interprocedural analysis (-ipo) with library archive (.a) files. Intel's web-based Premier Support is the best customer support interface that we've used."

Marc Rieffel
Senior Manager
Research and Development
Paracel

CCLRC Rutherford Appleton Laboratory: "We tried an early copy of the Intel Visual Fortran Compiler for Windows and our application built right out of the box. We noticed immediately that Intel improved compile-time and we really like the command- and source-compatibility with Compaq Visual Fortran*. The compiler also has strong Fortran 95 support. The new Intel Visual Fortran compiler did all that I expected of it. It is an impressive accomplishment."

Dr Stuart Campbell
ISIS Facility
CCLRC Rutherford Appleton Laboratory

Alias|Wavefront:: "We recently switched to the Intel C++ Compiler from a leading compiler for use on Maya 5.0's dynamics code. The output ran so much faster - up to 90% faster - than before, that our engineer didn't actually believe it and rechecked his data. To say the least, we are impressed with the Intel product."
Kevin Tureski
General Manager of Maya Engineering
Alias|Wavefront

CERN: "The Intel compilers have performed excellently on our ROOT code. On average, the Intel C++ Compiler for Linux produces executables that run 30% faster than ones produced by gcc 3.2. Due to the excellent compatibility with the GNU compilers, the porting effort was reduced to a minimum. Furthermore, the strong technical support that is provided with the Intel Compilers enables very quick turnaround on fixes for customer issues."

Dr. Fons Rademakers
Senior Scientist
CERN
Geneva

Fluent Inc.: "When we ran our standard benchmarks on GNU C and Intel C++ compilers for Linux, the Intel compiler gave us up to 37% performance improvement."
Dr. Dipankar Choudhury
Chief Technology Officer
Fluent Inc.

MUSICMATCH, Inc.: "After optimizing our code, (with the Intel compiler) MP3 encoding speeds dramatically increased, and MUSICMATCH Jukebox was able to rip MP3 files from audio CDs at a blistering 35x speed for the first time"

Jim McLaughlin
Director Hardware Alliances
MUSICMATCH, Inc.

back to top

Intel® Composer XE

Neubrex. Co., Ltd: "Intel continues to impress with me with its compilers for Windows* platform. The new Intel® Fortran Composer XE 2011 for Windows* and Intel® C++ Composer XE 2011 for Windows* offer developers great performance improvement features, such as the new Guided Auto Parallelism (GAP). Both products are nicely integrated into the newest editions of Visual Studio*. Enhancements to Intel® Math Kernel Library (Intel® MKL) and Intel® Integrated Performance Primitives (Intel® IPP) provide additional performance gains. These libraries not only provide speed, but also cover a wide range of applications and are great building-blocks for engineers and scientists. If you need performance for your C++ and/or Fortran applications, you need to try Intel software development products."

Dr. Artur Guzik
Senior Engineer
Neubrex. Co., Ltd

back to top

GameDev XDK, ImpactJS, App Dev Center, App Tester

Second Fiction: "Working with Intel's tools allows our team to shift our focus from "will it work" to "is it fun". Cutting out that hassle directly translates to more enjoyment for the development team and end user."

Joseph Krzemienski
Creative Director
Second Fiction

back to top

Intel® Inspector XE

River Silica Technologies: "Intel® Inspector's rich functionality has saved us a lot of testing time in our embedded project compared to traditional memory debugging techniques."

"Intel® Inspector allowed me to get in a single click down to the source and assembly code that was causing issues, a real time saver."

"Intel® Inspector was the only tool that was able to analyze our large binary file and provide meaningful information to help optimize our software."

"I was very happy to see how easy it was, with Intel® Inspector, to identify and fix memory and threading issues in our code."

"Intel® Inspector does a much better job at providing meaningful and human readable data compared to Valgrind*."

Jagadish Kamath
Co-founder and Software Architect
River Silica Technologies

Altech Multimedia: "Within a 1/2 hour after installing Intel® Inspector, it produced genuinely useful information that helped eliminate memory and threading issues"

"Intel® Inspector, a component of Emberson, allowed me to spot issues quickly that I was unaware off"

"Intel® Inspector found both memory and thread errors that could take days to find without it"

"The ability to use Intel® Inspector command lines to run analysis on a target system is extremely useful for debugging"

"Intel® Inspector has a very intuitive user interface that helped me find both memory errors like leaks and data access errors along with threading errors like data races in one package"

Ashley Driver
Solutions and Application Architect
Altech Multimedia

Envivio: a) "We struggled for a week with a crash situation, the corruption was identified but the source was really hard to find. Then we ran Intel® Inspector XE and immediately found the array out of bounds that occurred long before the actual crash. We could have saved a week!"

b) "We struggled for a week with a crash situation. Then we ran Intel® Inspector XE and immediately found the array out of bounds that occurred long before the actual crash. We could have saved a week!"

c) "We struggled for a week with a crash situation. Then we ran Intel® Inspector XE and immediately found the array out of bounds that occurred long before the actual crash."

d) "One week is a common delay to fix a memory corruption. With Intel® Inspector XE that is cut to less than a day."

e) "Intel® Inspector XE cuts the average time to fix a memory corruption from a week to less than a day."

f) "Intel® Inspector XE is reliable and easy to use. There are no special recompiles, the code inspected is exactly the same as the production code."

g) "Intel® Inspector XE is reliable and easy to use."

h) "Intel® Inspector XE has no special recompiles, the code inspected is exactly the same as the production code."

i) "I recommend Intel® Inspector XE for its ease of use and perfect integration in Visual Studio, plus its command line which allows us to use it for regression testing."

j) "I recommend Intel® Inspector XE for its ease of use and perfect integration in Visual Studio."

k) "One of Intel® Inspector XE's most useful features is the customizable analysis (from just memory leaks to full scan of memory usage).

i) The first level (memory leaks) runs almost as fast as the code goes normally. This level is usually used during development as it costs no time to run and secure memory expansion and allocation.

ii) The second level (detect memory issues) is used at the end of a development, before delivery to pre-integration to determine if there is a memory issue in the new feature.

iii) The third level (locate memory issues) is used in regression tests and when looking for a bug. This is slow to run, but the time saved vs. "manual" search is worth it.

iv) The thread checker is only used when working on the threading engine, but the process is the same, mandatory validation runs and (if needed) bug hunt runs."

l) "One of Intel® Inspector XE's most useful features is the customizable analysis (from just memory leaks to full scan of memory usage).

m) "One of Intel® Inspector XE's most useful features is the customizable analysis.

n) "One of Intel® Inspector XE's most useful features is the customizable analysis. We run fast checks frequently during development and more complete checks during regression testing."

o) "When writing multi-threaded code it is easy to introduce deadlocks and race conditions. Debugging them can be very difficult and time consuming because their occurrence is extremely random. Intel® Inspector XE saves valuable engineering time by detecting threading problems and locating them in the code."

p) "Debugging deadlocks and race conditions can be very difficult and time consuming because their occurrence is extremely random. Intel® Inspector XE saves valuable engineering time by detecting threading problems and locating them in the code."

q) "Intel® Inspector XE saves valuable engineering time by detecting threading problems and locating them in the code. Without it, debugging deadlocks and race conditions can be very difficult and time consuming because their occurrence is extremely random."

r) "Intel® Inspector XE saves valuable engineering time by detecting threading problems and locating them in the code."

Mikael Le Guerroue
Senior Codec Architecture Engineer
Envivio

Golaem: "Intel Inspector XE saves us weeks of time tracking down subtle errors in Golaem's parallel design implementation and , as an added bonus, pinpointed minor memory errors in legacy code." " Intel Inspector XE saved us weeks of time tracking down subtle errors in Golaem's parallel design implementation."

"Michaël Rouillé
CTO
Golaem

Siemens AG, Healthcare Sector: "a) "Intel® Inspector XE is quite fast and intuitive compared to products we have used in the past. We can now run our entire batch of test cases (~750) which was not feasible previously."

b) "Intel® Inspector XE is quite fast and intuitive compared to products we have used in the past."

c) "Unlike other products we have used, Intel® Inspector XE easily instruments DLLs and statically linked libraries."

d) "Intel® Inspector XE easily completed tests that failed due to lack of virtual memory on another product."

e) "Intel® Inspector XE reports call stacks for each error. The call stacks helped convince skeptical developers that their code might not be 100% correct..."

f) "Intel® Inspector XE is intuitive, 'fast' (for what it does) and stable. It is an easy way to improve code quality."

g) "Intel® Inspector XE works without a special recompile that some tools require. This speeds up the workflow and makes analysis more accessible to developers. They are much more likely to experiment with additional test cases and find more errors."

h) "Intel® Inspector XE works without a special recompile that some tools require. As a result, developers are much more likely to experiment with additional test cases and find more errors."

i) "Intel® Inspector XE works without a special recompile that some tools require. This speeds up the workflow and makes analysis more accessible to developers."

Gerald Mattauch
Senior Software Developer
Siemens AG, Healthcare Sector

back to top

Intel® Parallel Studio

Envivio: "Intel Parallel Studio, such as Intel Parallel Inspector, Intel Parallel Amplifier, and Intel Parallel Composer (which consists of the C++ Compiler, Intel IPP, and Intel TBB shortens our overall software development time by increasing the code's reliability and its performance in a multicore multithreaded environment. At the qualification stage, the number of dysfunctions is reduced due to a safer implementation, and the bug tracking becomes easier too. Intel Parallel Studio globally speeds up our software products time-to-market."

Eric Rosier
VP Engineering
Envivio

RWTH Aachen Univ: "Intel Parallel Studio extends Microsoft Visual Studio to provide an end-to-end integrated parallelism development environment."

Christian Terboven
Technical Engineer
RWTH Aachen Univ

MGH: "By utilizing technology that we developed in close collaboration with Intel, Microsoft, and Vectorform, we were able to significantly reduce the processing time and make the virtual colonoscopy a practical and efficient, and potentially lifesaving colon cancer screening technique."

Hiro Yoshida
Director, 3D Imaging Research, Massachusetts General Hospital and Associate Professor of Radiology, Harvard Medical School
MGH

ESI Group: "The Intel tools provided an excellent return on investment. Intel® Parallel Inspector allowed the code to be validated as 'data race-free' on our validation suite, and Intel® Parallel Amplifier allows us to focus efforts on the hotspots."

Andrew Cunningham
Technical Staff Member
ESI Group

Ionix: "I was pleasantly surprised by all the analysis functionality in Intel® Parallel Studio XE 2011 for Windows, and particularly impressed with the memory and threading analysis capabilities in Intel® Inspector XE 2011, part of Parallel Studio XE. I thought it was pretty cool that I could make changes, do another round of analysis and easily compare the 'before' and 'after effect on my app. I found the compiler in Intel® C++ Composer XE 2011 – also part of Parallel Studio XE – to be very solid. Its Guided Auto Parallelism (GAP) and Array Notation features were very helpful in adding performance. Parallel Studio XE 2011 adds a lot to Visual Studio*-based developers."

Erik Van Grunderbeeck
Owner- Ionix
Ionix

Golaem: "Using Intel Parallel Studio XE, we achieved a huge performance boost for Golaem Crowd 2 - a 15x improvement over Goalem Crowd 1. When run on the 3rd generation Intel Core processor family, performance will improve even more."

"Intel Parallel Studio Xe is driving innovation at Golaem, enabling us to achieve higher playback frame rates and faster render times - all which gives VFX and animation studios the ability to save both time and money while delivering impressive results."

Michaël Rouillé
CTO
Golaem

Harmonic Inc.: "a) The video processing and encoding libraries we develop need to be fast, error free, and take advantage of a variety of hardware platforms. The analysis applications in Intel® Parallel Studio XE have dramatically sped up our ability to identify bottlenecks, find/fix memory problems and track down difficult to isolate threading errors before our packages are released to the field.

b) The analysis applications in Intel® Parallel Studio XE have dramatically sped up our ability to identify bottlenecks, find/fix memory problems and track down difficult to isolate threading errors before our packages are released to the field.

c) The video processing and encoding libraries we develop need to be fast, error free, and take advantage of a variety of hardware platforms. The analysis applications in Intel® C++ Studio XE have dramatically sped up our ability to identify bottlenecks, find/fix memory problems and track down difficult to isolate threading errors before our packages are released to the field.

d) The analysis applications in Intel® C++ Studio XE have dramatically sped up our ability to identify bottlenecks, find/fix memory problems and track down difficult to isolate threading errors before our packages are released to the field."

Peter von Kaenel
Director, Software Development
Harmonic Inc.

Nik Software: "The new interface is a joy to use. Intel® VTune™ Amplifier XE gives us precise, down-to-the-metal performance data that's invaluable for pinpointing hotspots and evaluating the effect of optimizations."

Daniel Schwarz
Performance Engineer
Nik Software

Envivio: "Intel's static analysis allows us to close issues before the product ships. It locates "hard to find" errors (e.g., potential null pointers on conditional paths) that are unfortunately quite typical."

"Intel's static analysis locates "hard to find" errors (e.g., potential null pointers on conditional paths) that are unfortunately quite typical."

"Intel's static analysis allows us to close issues before the product ships."

"Intel's static analysis allows us to solve problems that have not yet showed up. Issues from atypical usage of our products that usually come long after the product is shipped and cost so much to solve at all levels of our company."

"Intel's static analysis allows us to solve problems from atypical usage of our products that usually come long after the product is shipped and cost so much to solve at all levels of our company."

"The best thing about Intel's static analysis is its ease of use. It is not intrusive in the process. Before any deliveries or while developing a feature, the tool can be launched with a single click."

"The best thing about Intel's static analysis is its ease of use. It is not intrusive in the process."

"The best thing about Intel's static analysis is its ease of use."

"I recommend Intel's static analysis. It is so easy to use. Finding bugs very early in the development process saves a considerable amount of time!"

"I recommend Intel's static analysis. Finding bugs very early in the development process saves a considerable amount of time!"

"I recommend Intel's static analysis. It is so easy to use."

"While not a replacement for regular testing, combining Intel's static analysis with Intel® Inspector XE creates a 'safe to deliver' sensation for nearly no effort."

Mikael Le Guerroue
Senior Codec Architecture Engineer
Envivio

Neubrex. Co., Ltd: "Our simulation and optimization software code base contains well over a 100,000 lines of Fortran, C, and C++ written over several years. Like most high performance simulation codes, it contains very complex execution flow, parallel sections, portions of legacy code, pointers, etc. Finding a bug in that type of application can be, and usually is, a real nightmare. Intel's static code analysis has been a very productive use of engineering time. We've found errors with OpenMP, pointers, arrays and legacy code that would have been difficult and time consuming to find with normal debugging techniques. Intel static analysis has become a regular part of our development workflow."

"Our simulation and optimization software code base contains well over a 100,000 lines of Fortran, C, and C++ written over several years. Finding a bug in that type of application can be, and usually is, a real nightmare. Intel's static code analysis has found errors with OpenMP, pointers, arrays and legacy code that would have been difficult and time consuming to find with normal debugging techniques. Intel static analysis has become a regular part of our development workflow."

"Finding a bug in well over a 100,000 lines of Fortran, C, and C++ written over several years can be, and usually is, a real nightmare. Intel's static code analysis has found errors with OpenMP, pointers, arrays and legacy code that would have been difficult and time consuming to find with normal debugging techniques."

"Intel's static code analysis has found errors with OpenMP, pointers, arrays and legacy code that would have been difficult and time consuming to find with normal debugging techniques."

"Intel's static code analysis has found errors that would have been difficult and time consuming to find with normal debugging techniques."

"Intel static analysis has become a regular part of our development workflow."

"I inherited 20 - 30,000 lines of mostly uncommented Fortran 77 code and had to add it to our simulation and optimization code already containing well over a 100,000 lines of Fortran, C, and C++. The inherited code was filled with error-prone "tricks" and contained a bug that was showing up "down-the-road", deep in the solving process. After more than a week of pure frustration, I gave up on finding it. Later, I tried Intel's Static Analysis and found the bug, along with at least 10 other error prone portions of code. Time saved? No idea, maybe I'd still be fighting it."

"I inherited 20 - 30,000 lines of mostly uncommented Fortran 77 code filled with error-prone "tricks". It contained a bug that was showing up "down-the-road", deep in the solving process. After more than a week of pure frustration, I gave up on finding it. Later, I tried Intel's Static Analysis and found the bug, along with at least 10 other error prone portions of code. Time saved? No idea, maybe I'd still be fighting it."

"I inherited 20 - 30,000 lines of mostly uncommented Fortran 77 code that contained a bug that showed up "down-the-road". After more than a week of pure frustration, I gave up on finding it. Later, I tried Intel's Static Analysis and found the bug, along with at least 10 other error prone portions of code. Time saved? No idea, maybe I'd still be fighting it."

"I inherited 20 - 30,000 lines of mostly uncommented Fortran 77 code that contained a bug that showed up "down-the-road". After more than a week of pure frustration, I gave up on finding it. Later, I tried Intel's Static Analysis and found the bug. Time saved? No idea, maybe I'd still be fighting it."

"I inherited 20 - 30,000 lines of code that contained a bug. After more than a week of pure frustration, I gave up on finding it. Later, I tried Intel's Static Analysis and found it. Time saved? No idea, maybe I'd still be fighting it." "

Dr. Artur Guzik
Senior Engineer, R&D
Neubrex. Co., Ltd

Golaem: "Intel Parallel Studio XE 2011 is a great software development tool for performance-oriented Windows-based C++ software developers. I achieved an astonishing boost in performance by sing Intel Cilk Plus and Array Notation features in my code. If you need performance, try Intel Parallel Studio XE 2011."

Michaël Rouillé
CTO
Golaem

Fujian Centerm: "Intel's Static Security Analysis finds more issues is more accurate than other tools we have tried.

We use Intel's Static Security Analysis before we do real debugging because it reduces our overall debugging time. It makes us more efficient.

We use Intel's Static Security Analysis before we do real debugging. It reduces our overall debugging time and makes us more efficient.

Intel's Static Security Analysis reduces our overall debugging time and makes us more efficient.

Intel's Static Security Analysis locates difficult to find errors in rarely executed code branches that are not tested by our regular functional test.

Intel's Static Security Analysis locates difficult to find errors in rarely executed code branches.

Intel's Static Security Analysis locates difficult to find errors that are not tested by our regular functional test.

Intel's Static Security Analysis locates difficult to find errors.

Intel's Static Security Analysis improves our code quality because it finds errors in code branches that are not executed by our test suite.

Intel's Static Security Analysis improves our code quality.

Intel's Static Analysis finds more issues is more accurate than other tools we have tried.

We use Intel's Static Analysis before we do real debugging because it reduces our overall debugging time. It makes us more efficient.

We use Intel's Static Analysis before we do real debugging. It reduces our overall debugging time and makes us more efficient.

Intel's Static Analysis reduces our overall debugging time and makes us more efficient.

Intel's Static Analysis locates difficult to find errors in rarely executed code branches that are not tested by our regular functional test.

Intel's Static Analysis locates difficult to find errors in rarely executed code branches.

Intel's Static Analysis locates difficult to find errors that are not tested by our regular functional test.

Intel's Static Analysis locates difficult to find errors.

Intel's Static Analysis improves our code quality because it finds errors in code branches that are not executed by our test suite.

Intel's Static Analysis improves our code quality."

Wang Hui
Vice Director of Software & Service
Fujian Centerm

back to top

Intel® System Studio

Schneider Electric: "The integration of Emberson into Yocto Project* and Eclipse* CDT streamlined our development as it is very easy to build and debug optimized code, customized Yocto Project images and application components for our Intel® Core™ and Intel® Atom™-based embedded solutions."

"Emberson allowed us to speed-up develop and testing of our compact integrated programmable logic controller designs."

"The integration to Yocto Project* was very easy in Emberson compared to other Linux* development tools."

Rajendra Alluri
Technical Lead
Schneider Electric

River Silica Technologies: "I was productive within minutes after installing Emberson."

"I was very pleased to see that Intel provides Emberson as it provides key capabilities that I need for my embedded projects."

"Emberson allowed us to speed-up development and testing time of our Video transcoding solution through its deep hardware and software insights."

Jagadish Kamath
Co-founder and Software Architect
River Silica Technologies

back to top

Intel® Threading Building Blocks

Pacific Northwest National Laboratory: "Using Intel TBB's new flow graph feature, we accomplished what was previously not possible, parallelize a very sizable task graph with thousands of interrelationships - all in about a week."

Robert Link
GCAM Project Scientist
Pacific Northwest National Laboratory

Golaem: "Intel TBB provided us with optimized code that we did not have to develop or maintain for critical system services. I could assign my developers to code what we bring to the software table - crowd simulation software."

"We achieved great scaling with a minimum of effort on our part. Intel TBB gave us cross-platform atomic operations, thread-safe containers, threads, and thread-locking mechanisms as well as multi-thread allocators in one package."

"Intel TGG gave us cross-platform atomic operations, thread-safe containers, threads, and thread-locking mechanisms as well as multi-thread allocators in a solid package."

Michaël Rouillé
CTO
Golaem

back to top

Intel® Integrated Performance Primitives

Logitech Inc.: "The Intel Integrated Performance Primitives library is to multimedia programming what the C language and its libraries were to assembly language. The abstraction provided by this library is a welcome addition in the realm of multimedia processing on PC platforms; it provides rapid implementation and integration, optimization for multimedia extension instruction sets, operating system portability, user and kernel mode support, and more -- a 'must have' in the toolkit of multimedia developers."

Remy Zimmermann
Video Software Engineering Manager
Logitech Inc.

Intecam, Inc.: "Intel IPP provided a 300 percent improvement in the number of users who can simultaneously participate in a web cast."

Leo Vovlfson
President and Chief Technology Officer
Intecam, Inc.

River Silica Technologies: "I have extensively used Intel® IPP functions in my code to accelerate development of multi format encoding and decoding, and realized a significant performance boost for our video transcoder which resulted in improved user experiences for media playback."

"Using Intel® IPP allowed us to accelerate time to market of our Video transcoding solution. We will continue to use Intel® IPP in our solution to speed up development and boost performance."

"Single Intel® IPP library takes care of multiple media extensions such as SSSE3, SSE and AVX, thus making porting life easier across multiple Intel platforms."

Jagadish Kamath
Co-founder and Software Architect
River Silica Technologies

back to top

Intel® JTAG Debugger

Macraigor Systems: "Intel® JTAG Debugger, a component of Emberson, allowed me to quickly examine memory and data structures and fix complex system errors to enhance system reliability."

"Intel® JTAG Debugger's bitfield editor allows for easy access and manipulation of processor and peripheral registers to speed-up error detection."

"Being able to access the entire processor and peripheral registers documentation within Intel® JTAG debugger's bitfield editor is a big time saver when debugging system software. Once you have used this feature you don't want to go back to manually browsing through architecture manuals."

James MacGregor
Managing Partner
Macraigor Systems

back to top

Intel® MPI Library

S&I Engineering Solutions: "Fast and accurate state of the art general purpose CFD solvers is the focus at S and I Engineering Solutions Pvt, Ltd. Scalability and effeciency are key to us when it comes to our choice and use of MPI libraries. The Intel MPI Library has enabled us to scale to over 10k cores with a high effeciency and performance."

Nikhil Vijay Shende
Director
S&I Engineering Solutions

University of Toronto: "We're delighted by the efforts of the Intel cluster tools team in helping us scale our applications to 10s of thousands of cores with Intel(r) MPI Library 4.0. Intel is again raising the performance bar in providing us with the most scalable commercial MPI library for Intel architecture based processors."

Dr. Daniel Gruner
Chief Technical Officer - Software
University of Toronto

University of Toronto: "As we scale our astrophysics codes and our clusters to handle larger and larger simulations, the Intel® MPI Library continues to deliver on scaling, performance, and stability."

Ue-Li Pen
Associate Director at Canadian Institute for Theoretical Astrophysics
University of Toronto

back to top

Intel® Performance Libraries

Univ of Tennessee, Innovative Computing Laboratory: "University of Tennessee, Knoxville: "The Intel® Math Kernel Library is indispensable for any high performance computer user on x86 platforms. It provides a rich, highly optimized collection of math routines, including our own BLAS, LAPACK, and ScaLAPACK in addition to other basic functions such as sparse matrix operations and FFTs. Outstanding performance is achieved on both multicore and multiprocessor systems."

Prof. Jack Dongarra
Univ of Tennessee, Innovative Computing Laboratory

ESI Computational Structural Mechanics Group: "ESI Computational Structural Mechanics Group: "PAM-CRASH and PAM-STAMP rely on the performance of the Intel® Math Kernel Library 10.0. We are delighted with the results in terms of both memory use and SMP performance."

Dr Antoine Petitet, Ph.D
HPC Lead
ESI Computational Structural Mechanics Group

CALYON: "CALYON counter party risk is computed using complex numerical algorithms, distributed on a cluster of hundreds of Intel processors. We are constantly facing performance issues. The Intel® MKL includes vectors and matrixes calculation, FFT, random generators, and many other algorithms which allowed substantial performance improvements, sometimes up to a hundred times, and reduced development costs."

Gilles Artaud
Head of Quantitative Research, Risk and Permanent Control
CALYON

ANSYS, Inc.: "Intel MKL helps ANSYS achieve excellent performance on Intel processors and has powered our engineering simulation software for more than 10 years. Intel multi-core processors coupled with the Intel MKL library help us deliver high performance, with engineered scalability from workstations to server systems. The continued optimization of Intel MKL ensures the best performance for ANSYS software users on the latest generation Intel processors."

Chris Reid
Vice President of Marketing
ANSYS, Inc.

Paradigm Geophysical: "Paradigm use Intel® MKL in our technical solutions for our oil and gas customers because it provides outstanding performance. Our HPC algorithms rely on Intel MKL being tuned, thread safe and multicore aware across Intel's processor families. Using Intel MKL enables us to focus our time and effort on our solutions, yet still deliver optimal performance across a range of systems, from lap tops and workstations to HPC clusters."

Alistair Downie
Paradigm Geophysical

Schneider Electric: "Intel® Math Kernel Library's optimized math functions are very important for our industrial automation solution as it helped boost system performance and reduced development time."

"For our Intel® Core™ processor targeted designs it is very valuable to have the performance optimized Intel® Math Kernel Library available at your fingertips through the integrated Emberson build environment."

Rajendra Alluri
Technical Lead
Schneider Electric

ESI Computational Structural Mechanics Group: "ESI Computational Structural Mechanics Group: "PAM-CRASH and PAM-STAMP rely on the performance of the Intel® Math Kernel Library 10.0. We are delighted with the results in terms of both memory use and SMP performance."

Dr Antoine Petitet, Ph.D
HPC Lead
ESI Computational Structural Mechanics Group

ANSYS, Inc.: "Intel MKL helps ANSYS achieve excellent performance on Intel processors and has powered our engineering simulation software for more than 10 years. Intel multi-core processors coupled with the Intel MKL library help us deliver high performance, with engineered scalability from workstations to server systems. The continued optimization of Intel MKL ensures the best performance for ANSYS software users on the latest generation Intel processors."

Chris Reid
Vice President of Marketing
ANSYS, Inc.

University of Tennessee, Knoxville: "The Intel® Math Kernel Library is indispensable for any high performance computer user on x86 platforms. It provides a rich, highly optimized collection of math routines, including our own BLAS, LAPACK, and ScaLAPACK in addition to other basic functions such as sparse matrix operations and FFTs. Outstanding performance is achieved on both multicore and multiprocessor systems."

Prof. Jack Dongarra
University of Tennessee, Knoxville
Innovative Computing Laboratory

CALYON: "CALYON counterparty risk is computed using complex numerical algorithms, distributed on a cluster of hundreds of Intel processors. We are constantly facing performance issues. The Intel® MKL includes vectors and matrixes calculation, FFT, random generators, and many other algorithms which allowed substantial performance improvements, sometimes up to a hundred times, and reduced development costs."

Gilles Artaud
Head of Quantitative Research, Risk and Permanent Control
CALYON

ANSYS, Inc.: "Intel® MKL helps ANSYS achieve excellent performance on Intel processors and has powered our engineering simulation software for more than 10 years. Intel multi-core processors coupled with the Intel MKL library help us deliver high performance, with engineered scalability from workstations to server systems. The continued optimization of Intel MKL ensures the best performance for ANSYS software users on the latest generation Intel processors."
Chris Reid
Vice President of Marketing
ANSYS, Inc.

Paradigm Geophysical: "Paradigm use Intel® MKL in our technical solutions for our oil and gas customers because it provides outstanding performance. Our HPC algorithms rely on Intel MKL being tuned, thread safe and multicore aware across Intel's processor families. Using Intel MKL enables us to focus our time and effort on our solutions, yet still deliver optimal performance across a range of systems, from laptops and workstations to HPC clusters."
Alistair Downie
Paradigm Geophysical

DALCO AG Switzerland: "DALCO is happy to recommend the Intel Math Kernel Library (MKL) as a valuable software accessory to customers who develop their own applications. Linking with MKL provides a significant performance boost over home-grown code for many commonly-used mathematical functions (as well as some that are not so common). It is cost-effective, easy to use, and works as a natural extension to several different compilers."
Dr. Enda O'Brien
Senior Software Engineer
DALCO AG Switzerland

University of Illinois: "I used to hand-code image processing routines in assembler for fast image recognition and detection frame rate. I 'discovered' that the Intel Integrated Performance Primitives (IPP) library offers comparable speed to assembler with significantly less programming effort. Also IPP provides a valuable but very complex computer vision algorithm which I wouldn't even dare to program myself in assembler. Kudos to Intel!"

Javier I. Girado
University of Illinois at Chicago

NPCube: "Intel IPP will help you out solve or discover your current optimizations issues in game development. If you need accurate and fast data computation, this is the way to go!"
Laurent Paret
VP/Business Development
NPCube

Logitech: "The Intel Integrated Performance Primitives Library has proven to be a valuable asset to Logitech. Developing products with IPP means we don't worry about optimizing for each Intel processor. IPP does that for us. Intel IPP clearly helps Logitech deliver superior products."

Ken Erbes
Director, Software Applications and Services
Logitech

ABAQUS, Inc.: "By adopting the [Intel® Math Kernel Library] (Intel® MKL) DGEMM libraries, our standard benchmarks timing improved between 43% and 71%, which is very impressive."
Matt Dunbar
Software Developer
ABAQUS, Inc.

ComVu: "Intelligent integration of the [Intel® Integrated Performance Primitive] (Intel® IPP) functions has saved us hundreds of hours of development time and greatly sped up delivery of our products to market."

William Mutual
CEO
ComVU

Inetcam, Inc.: "The Intel Integrated Performance Primitives ([Intel] IPP) has enhanced the iVISTA* application to be more in line with customers' expectations. For example, it enables us to dynamically rescale, in real-time, a video stream without loss of performance. This capability is more robust with Intel IPP."

Leo Volfson
President and Chief Technology Officer
Inetcam, Inc.

Inetcam, Inc.: "The Intel IPP provided a 300% improvement in the number of users who can simultaneously participate in a Webcast. In addition, the migration from the Intel® Pentium® III processor to the Intel Pentium 4 processor took only a day."
Leo Volfson
President and Chief Technology Officer
Inetcam, Inc.

University of Oxford: "One thing I particularly like about the Intel Math Kernel Library is the option for block-splitting in the random number generation. This is very useful for parallel applications."

Mike Giles
Professor
University of Oxford

NAG Ltd.: "It's possible that the next time you use the Internet to calculate your optimal financial portfolio, the calculation may depend upon the NAG optimization routines and beneath them, the Intel MKL Library."
"Intel MKL plays a key role in the multiplicative effect of the software developer stack."
Dr. David Sayers
Principal Consultant
NAG Ltd.

NAG Ltd.: "The Intel MKL routines perfectly complement the quality and coverage of the NAG libraries and they together deliver a wide-ranging set of high-performance components to NAG's users."
Dr. Brian Ford
Managing Director
NAG Ltd.

back to top


Intel® VTune™ Performance Analyzer

River Silica Technologies: "When optimizing its code and performance, Intel® VTune™ Amplifier eliminated the need for River Silica to use other profiling tools or guess which functions to optimize."

"The ability for Intel® VTune™ Amplifier to exactly pinpoint performance bottlenecks in our code was a big time saver and made it a far better choice compared to other analysis tools that we used."

"Using hotspot analysis in Intel® VTune™ Amplifier allowed us to significantly improve the performance of our video transcoder solution."

"Jagadish Kamath
Co-founder and Software Architect
River Silica Technologies

SAS Institute Inc.: "Last week, Intel® VTune™ Amplifier XE helped us find almost 3 fold performance improvement. This week it helped us improve the performance another 3 fold."

"I would definitely recommend Intel® VTune™ Amplifier XE and Intel® Inspector XE to anyone."

"Performance tools like Intel® VTune™ Amplifier XE are essential for finding application bottlenecks."

"I've worked with many third party tools. Integrating our application with Intel® VTune™ Amplifier XE and Intel® Inspector XE was the easiest integration I've even performed."

Claire Cates
Principal Developer
SAS Institute Inc.

OPEN CASCADE S.A.S.: "Intel® VTune™ Amplifier XE has a friendly interface. It is better than other profilers I have tried."

"One of Intel® VTune™ Amplifier XE's best features is that it is easy to use. I did not need to read the documentation."

"Intel® VTune™ Amplifier XE's timeline is very information intensive. It organizes the data I need to tune threaded applications."

Sergey Zaritchny
Software Development Manager
OPEN CASCADE S.A.S.

ESRI-UK: "My team develop an optimisation tool that is continually having more options and strategies added to it. It is only one part of a much larger product. It stands out because its use can take minutes or even hours, rather than seconds. It is not always easy to see how combinations of the features will interact on clients' data, and in particular how much new functionality will slow down their productivity.We've been tuning the tool for 15 years and there's not much easy slack in it now. An improvement of 10% is a major victory and it's easy for a bug fix to cause a 100% slow down if not implemented carefully.

We recently had to carry out some significant re-engineering and the resultant extra-level-of-indirection initially caused 50-1000% slow down when we removed some of the old caching. Our new caching mechanism involved several levels of interrelated object life-cycles and it took weeks to refine the mechanism. But it was having vtune's analysis that reassured us that the work was under control and progressing. The new Hot-Spot analysis analogous to the old call-tree profiling (but much faster to run)) was instrumental in the key breakthrough that brought our times back down to close to where we'd started, but with the re-engineered infrastructure enabling the next set of functional enhancements.

My previous experience with rival products had not been impressive. For anything other than toy projects I've rarely been able to get a good build. When they did build, they either crashed in unrelated areas of code or took so long that the results were too late to use.

VTune has always been more successful managing all the dozens of dlls that make up our application (of which I only want to examine one or two in detail). But recently, VTune 9.1's call stack instrumentation had started to fail too. The build just seemed to be too complex.

I'd resigned myself to hand instrumenting the application with simple logging and counting classes (still invaluable for some tasks) when I discovered a reference to the new XE beta while I was checking for a newer version of VTune.

Installing it was surprisingly painless and although the new interface was a little different, with a bit of experimentation, I was able to get simple sample statistics out within minutes. XE seems to be a major reimplementation of the profiling tools, not just a new UI on the same tools. Pushing my luck, I was delighted to discover that I could get very comprehensive call stack information and statistics from the new VTune XE Hotspot analysis and it had no trouble running and ran an order of magnitude faster! The sampling approach that it appears to use gives me everything I want and none of the instrumentation issues.

Being able to look at statistics for functions from the bottom up and from top down was invaluable in getting a feel for where to concentrate my efforts and for checking that changes had the effect I expected.

I'll be pushing hard to get a copy when XE is released and I'd strongly recommend VTune to anyone wanting to understand better were their time is going in heightweight scientific calculations. Tuning without it feels like compiling without warning or error messages!"

Richard Shepherd
Software Engineer
ESRI-UK

Nik Software: "The new interface is a joy to use. Intel® VTune Amplifier XE gives us precise, down-to-the-metal performance data that's invaluable for pinpointing hotspots and evaluating the effect of optimizations"

Daniel Schwarz
Performance Engineer
Nik Software

Adobe Systems, Inc.: "VTune™ analyzer has been a key tool in our development process for years. The new VTune™ Amplifier XE brings even more capability to an already indispensible tool. The sampling based call stack hotspots is excellent and alone is worthy of the upgrade. We have also been impressed by how the concurrency and Locks and Waits analysis can even provide useful data on complex applications such as Premiere Pro.

The new VTune™ Amplifier XE brings even more capability to an already indispensible tool. The sampling based call stack hotspots is excellent and alone is worthy of the upgrade. We have also been impressed by how the concurrency and Locks and Waits analysis can even provide useful data on complex applications such as Premiere Pro.

The new VTune™ Amplifier XE brings even more capability to an already indispensible tool. The sampling based call stack hotspots is excellent and alone is worthy of the upgrade.

VTune™ Amplifier XE's sampling based call stack hotspots is excellent and alone is worthy of the upgrade.

The new VTune™ Amplifier XE brings even more capability to an already indispensible tool. We have been impressed by how the concurrency and Locks and Waits analysis can even provide useful data on complex applications such as Premiere Pro.

VTune™ Amplifier XE's concurrency and Locks and Waits analysis can even provide useful data on complex applications such as Premiere Pro.

The new VTune™ Amplifier XE brings even more capability to an already indispensible tool."

Rich Gerber
Engineering Manager
Adobe Systems, Inc.

Golaem: "When optimizing its code and performance, Intel VTune Amplifier XE eliminated the need for Golaem to incldue profiling code in each function or guess which functions to optimize. The tool is fast enough for us to use on a daily basis without relying on our own performance analysis layer."

"Intel VTune Amplifier XE is fast enough for us to use on a daily basis without relying on our own performance analysis layer."

"Intel VTune Amplifer XE helped identify performance bottlenecks. The tool gives reliable performance reports through hardware-based sampling that we could compare with blueprints at each improvement and thus measure the impact of any modifications. Furthermore, system-wide, event-based sampling gave us precious information on cache misses in our code. Ite was an eye-opener for us."

"Intel VTune Amplifier XE gives reliable performance reports through hardware-based sampling that we could compare with blueprints at each improvement and thus measure the impact of any modifications. System-wide, event-based sampling gave us precious information on cache misses in our code. It was an eye-opener for us."

"Intel VTune Amplifier XE's system-wide, event-based sampling gave us precious information on cache misses in our code. It was an eye-opener for us."

Michaël Rouillé
CTO
Golaem

Microsoft Corporation: "We used Intel's VTune analyzer to give customers a better performance experience with our new SQL Server 2008 release, which has earned great reviews for performance and scalability. VTune gives us key performance insights into how our code can scale better, and how we can make more effective use of hardware. And the combination of performance event counters in Intel's new Xeon 5500 series processors, coupled with VTune's sampling capabilities, is just killer for performance professionals."

David Powell
Director, SQL Server Performance Engineering Team
Microsoft Corporation

back to top


Intel® Cluster Tools

Computational Astrophysics Laboratory: "Intel Trace Analyzer and Collector for Linux helped to drastically improve the performance of RIKEN's molecular dynamics cluster software. We were able to shorten MPI communication time by half by finding and removing bottlenecks with non-blocking communication patterns. Since Intel Trace Analyzer and Collector can embed instrumentation into the program, we can tell the execution time of each function and its load balance, which enabled us to very easily understand where to optimize. Intel's MPI and Cluster tools provide us the best cluster development environment."

Dr. Takahiro Koishi
Computational Astrophysics Laboratory
RIDEN, Japan

EM Software & Systems: "Our FEKO electromagnetics code runs on various parallel environments (e.g., clusters of different sizes, shared memory machines) and utilizes complex numerical algorithms that are non-trivial to parallelize. Intel Trace Analyzer and Collector have proven to be very valuable tools to help understand FEKO parallel communication patterns and consequently in optimizing the message passing call that result in an extremely well performing electromagnetics ISV cluster application."

Dr. Ing. Ulrich Jakobus
Technical Director
EM Software & Systems

back to top

Пожалуйста, обратитесь к странице Уведомление об оптимизации для более подробной информации относительно производительности и оптимизации в программных продуктах компании Intel.