On generating set index functions for randomized caches

Caches hide the growing latency of accesses to the main memory from the processor by storing the most recently used data on-chip. To limit the search time through the caches, they are organized in a direct mapped or set-associative way. Such an organization introduces many conflict misses that hamper performance. This paper studies randomizing set index functions, a technique to place the data in the cache in such a way that conflict misses are avoided. The performance of such a randomized cache strongly depends on the randomization function. This paper discusses a methodology to generate randomization functions that perform well over a broad range of benchmarks. The methodology uses profiling information to predict the conflict miss rate of randomization functions. Then, using this information, a search algorithm finds the best randomization function. Due to implementation issues, it is preferable to use a randomization function that is extremely simple and can be evaluated in little time. For these reasons, we use randomization functions where each randomized address bit is computed as the XOR of a subset of the original address bits. These functions are chosen such that they operate on as few address bits as possible and have few inputs to each XOR. This paper shows that to index a 2(m)-set cache, it suffices to randomize m+2 or m+3 address bits and to limit the number of inputs to each XOR to 2 bits to obtain the full potential of randomization. Furthermore, it is shown that the randomization function that we generate for one set of benchmarks also works well for an entirely different set of benchmarks. Using the described methodology, it is possible to reduce the implementation cost of randomization functions with only an insignificant loss in conflict reduction.

Highly accurate and efficient evaluation of randomising set index functions

Randomising set index functions can reduce the number of conflict misses in data caches by spreading the cache blocks uniformly over all sets. Typically, the randomisation functions compute the exclusive ors of several address bits. Not all randomising set index functions perform equally well, which calls for the evaluation of many set index functions. This paper discusses and improves a technique that tackles this problem by predicting the miss rate incurred by a randomisation function, based on profiling information. A new way of looking at randomisation functions is used, namely the null space of the randomisation function. The members of the null space describe pairs of cache blocks that are mapped to the same set. This paper presents an analytical model of the error made by the technique and uses this to propose several optimisations to the technique. The technique is then applied to generate a conflict-free randomisation function for the SPEC benchmarks. (C) 2003 Elsevier Science B.V. All rights reserved.

Selective extracellular vesicle-mediated export of an overlapping set of microRNAs from multiple cell types

Background: MicroRNAs (miRNAs) are a class of small RNA molecules that regulate expression of specific mRNA targets. They can be released from cells, often encapsulated within extracellular vesicles (EVs), and therefore have the potential to mediate intercellular communication. It has been suggested that certain miRNAs may be selectively exported, although the mechanism has yet to be identified. Manipulation of the miRNA content of EVs will be important for future therapeutic applications. We therefore wished to assess which endogenous miRNAs are enriched in EVs and how effectively an overexpressed miRNA would be exported.

Results: Small RNA libraries from HEK293T cells and vesicles before or after transfection with a vector for miR-146a overexpression were analysed by deep sequencing. A subset of miRNAs was found to be enriched in EVs; pathway analysis of their predicted target genes suggests a potential role in regulation of endocytosis. RT-qPCR in additional cell types and analysis of publicly available data revealed that many of these miRNAs tend to be widely preferentially exported. Whilst overexpressed miR-146a was highly enriched both in transfected cells and their EVs, the cellular:EV ratios of endogenous miRNAs were not grossly altered. MiR-451 was consistently the most highly exported miRNA in many different cell types. Intriguingly, Argonaute2 (Ago2) is required for miR-451 maturation and knock out of Ago2 has been shown to decrease expression of other preferentially exported miRNAs (eg miR-150 and miR-142-3p).

Conclusion: The global expression data provided by deep sequencing confirms that specific miRNAs are enriched in EVs released by HEK293T cells. Observation of similar patterns in a range of cell types suggests that a common mechanism for selective miRNA export may exist.

"Variations on Psycho Killer" - Musical Score. Commissioned by Pauline Kim Harris. A virtuosic set of variations on the famous Talking Heads song for solo violin. General information State: Unpublished Organisations: School of Creative Arts, Sonic Arts, UOA35 - Music, Drama, Dance and Performing Arts Authors: Lyon, E. Keywords: (Talking Heads, Violin Music) Publication date: 2012

Musical Score. Commissioned by Pauline Kim Harris. A virtuosic set of variations on the famous Talking Heads song for solo violin.

Ordering effects and choice set awareness in repeat-response stated preference studies

We present an experiment designed to investigate the presence and nature of ordering effects within repeat-response stated preference (SP) studies. Our experiment takes the form of a large sample, full-factorial, discrete choice SP exercise investigating preferences for tap water quality improvements. Our study simultaneously investigates a variety of different forms of position-dependent and precedent-dependent ordering effect in preferences for attributes and options and in response randomness. We also examine whether advanced disclosure of the choice tasks impacts on the probability of exhibiting ordering effects of those different types. We analyze our data both non-parametrically and parametrically and find robust evidence for ordering effects. We also find that the patterns of order effect in respondents' preferences are significantly changed but not eradicated by the advanced disclosure of choice tasks a finding that offers insights into the choice behaviors underpinning order effects. © 2011 Elsevier Inc.