Technology
TMT Unique Value
Only TMT delivers the technology to completely overcome the barriers to finding and implementing parallel processing performance on multicore systems.
- Transparency - The SequenceL™ compiler enables easy programming, optimization and debug of complex algorithms. The compiler simplifies the implementation of multi-core applications by eliminating the need to identify and specify parallelism in the code. The use of a compact general purpose language allows users to easily cut through code complexity with fewer lines of code, reducing test burden, improving development productivity, and more easily finding opportunities for optimization.
- Transportability - The SequenceL™ compiler automatically extracts parallelism from the application code, enabling code compilation to any parallel architecture from common source code. TMT multi-core programming solutions also include technology libraries that optimize system performance and synchronization. For example, the SequenceL compiler can evenly distribute processor loads, remove bottlenecks from core communications and optimize shared memory access. As a result of these solutions, the SequenceL code remains independent of computing system architecture, eliminating the need to re-write application code for new multi-core processor releases and future heterogeneous many-core architectures.
Key Benefits of TMT Solutions
TMT multicore solutions allow software programmers to recognize these key benefits for their application development projects.
- Increased Operational Performance - Using the SequenceL™ compiler, parallel or threaded programs are automatically produced and optimized for a given implementation. When run on multicore machines, the program will automatically identify how many cores are available, allocate the work load appropriately to each core, and adjust those allocations as the program runs to ensure all cores are being fully utilized. This will result in dramatically faster application operation and more efficient computing utilization. In fact, the more cores available, the more effective the SequenceL™ compiler methodology is opposed to sequential languages, which become dramatically more difficult to program as the number of cores increases.
- Improved software quality - Because the SequenceL™ compiler code is written in "Engineering Terms", it is dramatically simplified and much easier to read and debug. With functionality being represented in just a few lines, as opposed to 100's or 1000's of lines required in a sequential language, it becomes much easier to understand code intent, find potential problems and optimize the application functionality. The end result is higher program quality and reduced testing costs. This explains how applications developed using the SequenceL™ compiler often achieve acceleration results exceeding the number of additional cores in the target system.
- Reduced Development Costs - Programs are written in "Engineering Terms" averting the need for detailed knowledge of parallel programming or multicore architectures. This reduces the development schedule and costs by enabling algorithm developers to produce optimized, threaded code from simple code. Engineers and scientists can then concentrate their efforts on application creativity rather than on the obscure intricacies of the program implementation.