Content deleted Content added
Matt Britt (talk | contribs) bah... that section sucked all around; there are plenty of prior examples of multiple core microprocessors |
→Licensing: new section for software development |
||
Line 41:
*Scaling efficiency is largely dependent on the application or problem set. For example, applications that require processing large amounts of data with low computer-overhead algorithms may find this architecture has an I/O bottleneck, underutilizing the device.
==software impact ==
Most existing software is not ready to directly utilize the power of multicore processors since they are written in traditional sequential [[programming language]]s like C, C++ and [[FORTRAN]], all of which have the limited scope of only one processor in mind.
[[Parallel programming]] is a must option for a single software to exploit multiple computation units(cores) simultaneous, often by multithread or multitask programming. Two long-existing [[parallel programming model]]s ([[OpenMP]] and [[MPI]] ) can be directly used on multi-core platforms. Other research efforts have been seen also, like Cray’s Chapel, Sun’s Fortress, and IBM’s X10.
==Licensing==
| |||