In this paper, we measure the memory performance throughout the Phoronix test "RAMspeed SMP". We decide to test this specific benchmark because we know how important is the memory for the system performance. This document shows how much the memory performance could change if we modify some variables in the linux kernel.
The Linux kernel controls the way tasks (or processes) are managed in the running system. The task scheduler, sometimes called process scheduler, is the part of the kernel that decides which task to run next. In this project its analyzed the behavior of scheduler by changing a default value from the runtime scheduling. The default value is 950000µs, or 0.95 seconds for the sched\_rt\_runtime\_us or scheduler realtime running variable. Meaning that 5% of the CPU time is reserved for processes that don't run under a real-time or deadline scheduling policy. This value in this file specifies how much of the "period" time can be used by all real-time and deadline scheduled processes on the system. The AIO-Stress which shows the obtained results in the different tests is an a-synchronous I/O benchmark created by SuSE which is is a German Linux distribution provider and business unit of Novell, Inc.
In this paper we will study an algorithm designed by Madgwick which is commonly used to determine the orientation of a quadcopter. The algorithm uses a group of accelerometers, gyroscopes and magnetometers integrated in what is called an IMU as input. Some differences have been found between the results obtained by the original paper and the implementation done by the author. Therefore, a thorough study has been made, finding a miscalculation in the equations. The results show a relative average error in the orientation of 1,44 ppm.