Custom Search

nCore Design LLC

nCore Design LLC
nCore Design LLC develops high performance software for multicore processors, GPUs and embedded systems. We have significant experience on a wide variety of computing platforms and specialized knowledge of applied software optimization techniques. We also deliver in-depth, instructor-led, on-site and scheduled training classes focused on programming multicore processors and GPUs. Founded in 2007, nCore is a privately held company with offices in Japan and the US. nCore works with local partners to deliver state-of-the-art solutions to multinational technology companies.
nCore Design LLC is based in Beaverton, OR, USA

High Performance Software Architecture and Development

  • Conceptualization, design and development of bespoke, high-performance, scalable ANSI C/ C software systems on Unix/ Linux based platforms
  • Specialization in networked, multicore, multithreaded and parallel software systems and architectures
  • Adaptive, agile development process minimizes risk and engages directly with the customer to produce a superior solution
  • Multi-platform and cross-platform configuration management and build systems using Jam and gmake
  • Experienced with the development of online games, distributed simulations, transaction execution engines, search frameworks, in-memory databases, expert systems, and special purpose Internet content servers
  • Object Databases (Versant, ObjectStore), CORBA, MPI, Myrinet MX/ GM middleware
  • Linux, FreeBSD, OS X, OpenSolaris, HP-UX and HP Tru64 operating systems
  • IA-32, IA-64, PowerPC, SPARC and MIPS microprocessors

Multicore Software

  • Software design and development for multicore systems including pipelined processing engines, video decoding and encoding, transaction execution engines and concurrent networking systems
  • Technical analysis of existing software, hardware and threading APIs for future multicore implementations
  • Migration and porting of existing software to multicore-aware versions using specialized tools and techniques
  • Performance analysis and software instrumentation for multicore systems

Embedded Systems

  • Software and solution development for digital microprocessor controlled systems, system-on-chip, digital signal processors and microcontrollers
  • Embedded operating systems and embedded web application programming with a Linux specialization
  • Specific experience with Broadcom BCM74XX HD AVG system processors
  • Finished systems with integrated GUI, USB peripherals, LCD panels, keyboards and hardware encryption modules
  • ARM, XScale, x86, MIPS, PowerPC, TMS320, ADSP/ Blackfin, AVR, MSP430 platforms

Software Optimization

  • Correction and optimization of performance issues using advanced code profiling and compiler optimization techniques
  • Integrated system, application and microarchitecture optimization strategies
  • Optimization solutions for SIMD vectorization, branch mis-prediction, memory problems, cache problems and processor-specific microarchitecture problems
  • Optimization using Intel IPP and Math Kernel Library on Linux and OS X

Technical Consulting

  • Technical consulting for software development projects, including systems architecture, system feasibility, requirements review and development, design review, test plan development, coding standards and guidelines development, and independent evaluations of existing projects
  • Training and mentoring programs in advanced C tools and techniques
  • Training in multicore software development using Intel development tools
Multicore Training
NCT-100 Programming Multicore Processors instructor led trainingcomputer lab : This course covers concepts and approaches related to programming multicore processors in C/ C++. From recognizing parallelism opportunities, to multithreaded programming and designing multithreaded algorithms, this course teaches you how to deal with the many aspects of synchronization, multithreading libraries, processor caches, shared memory and specialized methods for multithreaded programming using modern approaches such as OpenMP and Intel Threading Building Blocks.
Head Office
14525 SW Millikan Way
#29870
Beaverton
OR
USA
Training and Seminar Locations
Do you teach concurrency ?
tcw11-gfc-v396M-11/20/09-02:10:52-()[A]-[B]-[A] -13:43:42