We present a novel taxonomy that characterizes in a structured way the software and hardware tradeoffs for I/O data passing between applications and operating system. This work contributes new techniques, input-disabled pageout, transient output copy-on-write, and input alignment, that are used

nv c, neti me 8, U Sparse representation Non-convex optimization s si the ng W proposed. It has better performance in preserving image edges than conventional sparsifying transforms. are presented in the smooth region when the data are highly undersampled. In RI) is ical str

/or additions to the approach described as the DSL [i/o] v 4.1 algorithm • avoiding loudness discomfort during hearing instrument use, • recommending a hearing instrument frequency response that ensures audibility of important acoustic cues in a conversational speech signal as much as possible, • supporting

Assuming the existence of iO for P/poly and one-way functions, we show how to succinctly garble bounded-space computations (BSC) M: the size of the garbled program (as well as the time needed to generate the garbling) only depends on the size and space (including the input and output) complexity

neural network from I/O measurements, can be modified with an NL q stability constraint in order to ensure globally asymptotically stable identified models. An example illustrates how system identification of an internally stable model, corrupted by process noise, may lead to unwanted limit cycle

this asymmetry in FD-Buffer. Specifically, we use the average I/O cost per page access as op-posed to the traditional miss rate as the performance metric for a buffer. We develop a new replacement policy in which we sepa-rate clean and dirty pages into two pools. The size ratio of the two pools is automatically

n S o u t h A f r i c a

FD-buffer: a buffer manager for databases on flash disks.

This thesis must be used in accordance with the provisions of the Copyright Act 1968. Reproduction of material protected by copyright may be an infringement of copyright and copyright owners may be entitled to take legal action against persons who infringe their copyright. Section 51 (2

of the Problem and the Explanation of O What we really want. What we want is to estimate how fast the algorithm is, i.e., how much time T (x) it takes on dierent inputs x. Worst-case and average computation time. In some situations, e.g., in auto-matic control, we need to make decisions in real time, i