Modern CPUs provide SIMD (single instruction, multiple data) vector instructions which simultaneously apply the same operation to multiple elements. Peak performance of single cores can often only be reached if vector instructions are used. Vector instructions are automatically generated by compilers, but this approach has its limits. To fully exploit the SIMD capabilities, vector instructions can directly be used in the C programming language through "intrinsics". The lecture explains how vector intrinsics available for the Intel vector extensions (SSE, AVX, AVX-512) can be used to parallelize C programs. In the exercises, this knowledge is applied in programming projects.
| Frequency | Weekday | Time | Format / Place | Period | |
|---|---|---|---|---|---|
| weekly | Do | 12-14 | D4-135 | 12.10.2026-05.02.2027 | |
| by appointment | n.V. | D4-135 | 12.10.2026-05.02.2027 | Die Übungen werden zum Semesterbeginn festgelegt! |
| Module | Course | Requirements | |
|---|---|---|---|
| 39-Inf-EGMI Addition Module in Informatics Ergänzungsmodul Informatik | vertiefende Informatikvorlesung 3.6 | Ungraded examination
|
Student information |
| vertiefende Übung 1.1 zu einer Vorlesung | Ungraded examination
|
Student information |
The binding module descriptions contain further information, including specifications on the "types of assignments" students need to complete. In cases where a module description mentions more than one kind of assignment, the respective member of the teaching staff will decide which task(s) they assign the students.