No abstract

Optimal multiprocessor real-time schedulers incur significant overhead for preemptions and migrations. We present RUN, an efficient scheduler that reduces the multiprocessor problem to a series of uniprocessor problems. RUN significantly outperforms existing optimal algorithms with an upper bound of O(log m) average preemptions per job on m processors (fewer than 3 per job in all of our simulated task sets) and reduces to Partitioned EDF whenever a proper partitioning is found.

No abstract

No abstract

SUMMARYWe present a low-supply voltage (2 V) low-power consumption (500 W) analogue phase-locked loop (PLL), working at two low frequencies (1 and 10 kHz), to be used in an integrated lock-in ampliÿer. An externally settable control bit allows the switching operation between the two di erent frequencies. The circuit has been designed in a standard 0.6 -m CMOS technology and di ers from the standard analogue PLL architectures for the current mode implementation of both the loop ÿlter and of the oscillator. Three di erent locked waveforms (sinusoidal, triangular, squared) can be obtained at the PLL output. Simulation results, obtained through the use of PSPICE and using accurate transistor models, will be proposed. The pull-in ranges are about ±250 Hz around 1 and ±1:3 kHz around 10 kHz, with pull-in times of about 10 and 4 ms, respectively.