Thesis (M.S.)--George Mason University, 2007. / Title from PDF t.p. (viewed Jan. 18, 2008). Thesis director: Elizabeth White. Submitted in partial fulfillment of the requirements for the degree of Master of Science in Computer Science. Vita: p. 152. Includes bibliographical references (p. 150-151). Also available in print.

Thesis (Ph. D.)--Washington State University, August 2007. / Includes bibliographical references (p. 168-179).

Title Garbage Collection in Software Performance Engineering Author Peter Libič peter.libic@d3s.mff.cuni.cz Advisor doc. Ing. Petr Tůma, Dr. petr.tuma@d3s.mff.cuni.cz Department Department of Distributed and Dependable Systems Faculty of Mathematics and Physics Charles University Malostranské nám. 25, 118 00 Prague, Czech Republic Abstract The increasing popularity of languages with automatic memory management makes the garbage collector (GC) performance key to effective application execution. Unfortunately, performance behavior of contemporary GC is not well understood by the application developers and often ignored by the per- formance model designers. In this thesis, we (1) evaluate nature of GC overhead with respect to its effect on accuracy of performance models. We assess the possibility to model GC overhead as a black-box and identify workload characteristics that contribute to GC performance. Then we (2) design an analytical model of one-generation collector and a simulation model of both one-generation and two-generation collectors. These models rely on application characteristics. We evaluate the accuracy of such models and perform an analysis of their sensitivity to the inputs. Using the model we expose the gap between under- standing the GC overhead based on knowing the algorithm...

CARVALHO, Everjane de. Representações sociais e práticas cotidianas de trabalhadores e trabalhadoras do lixo - a experiência do centro comunitário Santa Terezinha. 2004. 84 f. : Dissertação (mestrado) - Universidade Federal do Ceará, Mestrado em Desenvolvimento e Meio Ambiente, Fortaleza-CE, 2004. / Submitted by demia Maia (demiamlm@gmail.com) on 2016-05-13T14:02:47Z No. of bitstreams: 1 2004_dis_ecarvalho.pdf: 325080 bytes, checksum: 085476bac044ae68f1643004e4d48407 (MD5) / Approved for entry into archive by demia Maia (demiamlm@gmail.com) on 2016-05-13T14:03:41Z (GMT) No. of bitstreams: 1 2004_dis_ecarvalho.pdf: 325080 bytes, checksum: 085476bac044ae68f1643004e4d48407 (MD5) / Made available in DSpace on 2016-05-13T14:03:41Z (GMT). No. of bitstreams: 1 2004_dis_ecarvalho.pdf: 325080 bytes, checksum: 085476bac044ae68f1643004e4d48407 (MD5) Previous issue date: 2004 / O presente trabalho busca apreender as representações sociais de trabalhadores e trabalhadoras do lixo e suas práticas cotidianas, a partir da experiência do Centro Comunitário Santa Terezinha, com o Programa Reciclando - Rede Solidária de Coleta Seletiva. Parte-se do pressuposto de que as representações sociais se constroem no cotidiano e orientam as ações das pessoas. Nesse processo de interação entre a ação e as reflexões, as representações sociais modificam as práticas sociais e são modificadas pelas mesmas. Na reflexão acerca do trabalho realizado pelos agentes analisa-se a forma de ocupação espacial que se deu no morro Santa Terezinha bem como os novos sentidos atribuído às vias públicas como locus de trabalho e não como via de acesso que se dá nos bairros nobres ao redor do Morro. A pesquisa foi realizada com agentes recicladores do referido Centro. Os agentes buscam resignificar seu trabalho atribuindo um conjunto de regras que os diferenciam dos outros profissionais que se ocupam da mesma atividade na área da pesquisa. Dessa forma estes procuram superar os preconceitos existentes em relação à atividade da coleta. Observa-se que a coleta seletiva não se dá em virtude da ausência de um investimento em educação ambiental que possibilitaria a separação do lixo na própria fonte geradora. Os poucos clientes que separam é resultado da iniciativa pontual dos agentes que negociam em alguns condomínios essa separação. O cotidiano dos agentes consiste em percorrer as ruas e remexendo o lixo deixado nas calçadas recolher o material reciclável para posterior venda dos mesmos. Transitar nas ruas competindo com os carros representa um risco para os agentes. Outro risco a que estão submetidos é o de adquirir doenças em decorrência do contato direto com materiais contaminados. A falta de seguranças trabalhistas preocupa os agentes. Acredita-se que se impostos são pagos para assegurar a limpeza urbana esse trabalho deveria ser subsidiado pelo poder público de forma a garantir uma melhor condição de vida e trabalho aos agentes. Os problemas ambientais apontam para a urgência de um desenvolvimento sustentável para a sociedade. Educação ambiental e coleta seletiva do lixo são instrumentos fundamentais na luta pela sustentabilidade.

Made available in DSpace on 2014-06-11T19:27:52Z (GMT). No. of bitstreams: 0 Previous issue date: 2007-07-26Bitstream added on 2014-06-13T19:36:07Z : No. of bitstreams: 1 mota_rc_me_rcla.pdf: 864792 bytes, checksum: b5dc444d25c637c8318c55c35af6ecd9 (MD5) / A produção de lixo vem aumentando de forma preocupante em todo o mundo, estando este fato intrinsecamente ligado aos hábitos de consumo e ao estilo de vida das pessoas. Este é um dos graves problemas da atualidade e qualquer ação que vise sua solução, ou pelo menos sua minimização, é bem vinda. Assim, é dever de toda instituição pública, dentre elas as Universidades, desenvolver relacionados à questão da produção e destinação final dos resíduos sólidos. O presente trabalho tem o intuito de diagnosticar a atual situação de Programas de Coleta Seletiva desenvolvidos por algumas das principais instituições públicas de ensino superior do Estado de São Paulo, dando destaque ao Programa de Coleta Seletiva já em andamento na UNESP - Campus Rio Claro. Também é objetivo do trabalho entender a dinâmica deste Programa e de suas relações com a comunidade local, procurando formular ações que propiciem sua manutenção e continuidade, assim como a criação de novos Programas de minimização de resíduos. / The production of garbage is increasing in a dangerous way all over the world, and this fact is inherent to the consumption habits and the people's lifestyle. Nowadays, this is one of the greatest problems and any kind of action to solve or lessen it, is welcome. Thus, it is a duty for all the public institutions, including universities, to develop programs related to garbage production and destination. This research intends to diagnose the current situation of the Recycling Programs developed by some of the main public universities in the State of São Paulo, mainly the program that is happening in UNESP - Campus Rio Claro. It is also the objective of this research to know this program's dynamic and its relations with the local community, in order to formulate actions that assure its maintenance and continuation, as well as the creation of new recycling programs.

Orientador: João Antonio Galbiatti / Banca: Carlos Eduardo Angeli Furlani / Banca: Emerson Fachini / Resumo: O desenvolvimento dos centros urbanos tem conduzido a um aumento desordenado na geração de resíduos e conseqüente poluição ambiental. Tendo em vista o potencial agronômico dos resíduos gerados pelas cidades, o presente trabalho teve por objetivo verificar a influência do composto de lixo (CL), composto de poda de árvore (CP) e substrato comercial (SC), associados a diferentes lâminas de irrigação no desenvolvimento de porta-enxerto (limão-cravo) de citros. O delineamento experimental adotado foi em blocos casualizados em um esquema fatorial 6x3, sendo seis diferentes combinações de substratos: S1 - 100% SC; S2 - 80% SC + 20% CP; S3 - 60% SC + 30% CP + 10% CL; S4 - 40% SC + 40% CP + 20% CL; S5 - 20% SC + 50% CP + 30% CL e S6 - 60% CP + 40% CL associados a três lâminas de irrigação: 50% (I1), 100% (I2) e 150% (I3) da evapotranspiração diária medida pelo atmômetro. Os parâmetros avaliados das plantas foram: altura, número de folhas, diâmetro do caule, matéria fresca aérea e matéria seca aérea. Os resultados permitiram concluir que a irrigação abaixo da evaporação medida pelo atmômetro causaram efeito negativo no desenvolvimento das plantas e que a reposição de água com 150% (I3) foi a que favoreceu o desenvolvimento. Os substratos S2, o qual se utilizou 80% de SC associado à 20% CP foi os que proporcionou melhor desenvolvimento dos porta-enxertos. Os substratos com porcentagem de composto de lixo superior a 30% mostraram-se inadequados para o desenvolvimento dos porta-enxertos / Abstract: The development of urban centers has led to an inordinate increase in the generation of waste and consequent environmental pollution. Considering the agronomic potential of the waste generated by cities, the present work objective was to check the influence of garbage compound (GC), pruning of tree (PT) and commercial substrate (CS), associated to different irrigation blades in the rootstocks (Rangpur Lime) citros development. The experimental delineation adopted was in random blocks in a fatorial scheme 6x3, being six different substrates combinations: S1 - 100% CS; S2 - 80% CS + 20% PT; S3 - 60% CS + 30% PT + 10% GC; S4 - 40% CS + 40% PT + 20% GC; S5 - 20% CS + 50% PT + 30% GC e S6 - 60% PT + 40% GC associated to different irrigation blades: 50% (I1), 100% (I2) e 150% (I3) of the daily evapotranspiration measured by the atmometer. The evaluated plants parameters were: height, leaves number, stem diameter, fresh matter and dry matter of the air part. The results allowed to conclude that the irrigation below the evaporation measured by the atmometer caused negative effect in the plants development and the water replacement with 150 % (I3) was the one that favored the development. The substrates S2, which is used 80% of SC associated with 20% CP was that the enhanced growth of the rootstocks. The substrates with garbage compound percentage superior to 30 % appeared unsuitable for the rootstocks development / Mestre

I’m interested in writing stories that make me as author disappear. A little. I’d like for my stories to unravel themselves while I sit just barely visible, maybe on a porch across the street. I’m also interested in playing with unusual phrases and syntax to achieve authentic voice in my stories. This sets up a conflict because while I want to develop small, fairly simple stories, I also value some language trickery, which might come off as authorial. I want nuanced voices that don’t feel editorialized. I want the stories to be authentic in an off-putting way. My biggest challenge has been monkeying with language in ways I find interesting while still maintaining a cool distance. It feels like training a service dog without getting sentimental. I like these problems though. I like the tiptoe-ing. My goal is to be able to drop readers in the middle of a situation: childhood, a factory, the grieving process, and carry them through it, without them knowing I’m there, without having to rely on explanations of characters’ thoughts, their motives. I am drawn to stories with little exposition. As a reader, I like making my discoveries through characters, how they navigate the world. I like to read stories that are revelatory in an interesting way – without having to be told outright how a life got so raw, or why lying can be the greatest relief, or how come it’s heartbreaking to see up close how much makeup a woman wears. I’ve heard this advice over the years: “Write what you know.” I’ve tried this with dull results. I’ve decided that I disagree. I’m working to write more stories about lives, jobs, concepts, illnesses, joys and sadnesses that I don’t know. I like trying on the other: a housewife, a man, a teen, a liar, someone forgotten. By writing what I don’t know, I want to stir up the reader, deliver something familiar yet jarring.

Garbage collector performance in LISP systems on custom hardware has been substantially improved by the adoption of lifetime-based garbage collection techniques. To date, however, successful lifetime-based garbage collectors have required special-purpose hardware, or at least privileged access to data structures maintained by the virtual memory system. I present here a lifetime-based garbage collector requiring no special-purpose hardware or virtual memory system support, and discuss its performance.

Modern Garbage Collectors provide performance improvements by increasing program locality to utilize the faster CPU cache. A common approach is to move objects together according to the mutators’ access order, which brings more relocations during GC. In most cases, more relocations would not impact performance when using concurrent Garbage Collectors such as ZGC. However, in constrained environments with fewer CPU cores or less memory, bad relocation performance will cause overall performance degradation. In this thesis, we investigated why larger objects do not benefit from better program locality, then proposed a new design to reduce the number of relocations by efficiently identifying and ignoring larger objects. As a result, the relocation performance can be improved. In constrained environments, this can lead to an increase in overall throughput. In the new design, we introduce an extra page type, the tiny page. If an object is considerably small that it could benefit from relocation, it will be placed on the tiny page when allocating. As a result, we could replace the time‐consuming size check of objects with a faster page type check. Memory fragmentation also can be reduced by this design. To evaluate this design, we add the size identification procedure into a locality improvement implementation named HCSGC. The results of benchmarks show a slight improvement in constrained environments. In the JGraphT benchmark, we see a 3‐5% speedup in different configurations with memory limitations. In the SPECjbb2015 benchmark, we see a ~1% increase in performance on average, but with overlapping confidence intervals. In the DaCapo benchmark suite, we see a 1% improvement in the sunflow benchmark with CPU constraint. For other benchmarks in DaCapo, no significant difference is discovered. The results suggest that the proposed new design is a feasible way of filtering out larger objects, and doing so can further improve the relocation and overall performance. / Modern Garbage Collector ger prestandaförbättringar genom att öka programplatsen för att använda den snabbare CPU‐cachen. En vanlig metod är att flytta fler objekt baserat på mutators åtkomstorder. I de flesta fall skulle fler omplaceringar inte påverka prestanda vid användning av samtidiga Garbage Collector som ZGC. Men i begränsade miljöer med färre CPU‐kärnor eller mindre minne kommer dålig flyttningsprestanda att leda till övergripande prestandaförsämring. I denna avhandling undersökte vi varför större objekt inte gynnas av bättre programplats, och föreslog sedan en ny design för att minska antalet flyttningar genom att effektivt identifiera och ignorera större objekt. Som ett resultat kan flyttningsprestandan förbättras. I begränsade miljöer kan detta leda till en ökning av den totala genomströmningen. I den nya designen introducerade vi en extra sidtyp, den lilla sidan. Om ett objekt är avsevärt litet som kan ha nytta av omplacering, kommer det att placeras på den lilla sidan vid allokeringen. Som ett resultat kan vi ersätta den tidskrävande storlekskontrollen av objekt med en snabbare sidtypskontroll. Minnesfragmentering kan också reduceras med denna design. För att utvärdera denna design lägger vi till storleksidentifieringsproceduren i en implementering av lokaliseringsförbättring som heter HCSGC. Resultaten av riktmärken visar en liten förbättring i begränsade miljöer. I JGraphT‐riktmärket ser vi en hastighet på 3‐ 5% i olika konfigurationer med minnesbegränsningar. I riktmärket SPECjbb2015 ser vi i genomsnitt en ~1% prestationsökning, men med överlappande konfidensintervall. I DaCapo ‐riktmärket ser vi en förbättring på 1% i solflödesriktmärket med CPU‐begränsning. För andra riktmärken i DaCapo upptäcks ingen signifikant skillnad. Resultaten tyder på att den föreslagna nya designen är ett genomförbart sätt att filtrera bort större objekt, och det kan ytterligare förbättra flytten och den övergripande prestandan.

Pointers in 64-bit architectures are unlikely to exhaust their vast address range, and are as such needlessly big. Reducing the amount of memory a pointer occupies leads to reduced memory demands, better usage of memory, and better locality. Pointer compression is a term that encompasses techniques that aim to make pointers occupy less memory, often to 32-bit for the sake of word alignment. Pointers that are 32-bit embody the opposite problem of having too restricted of an address range, being able to address only 4 GB. Z is a garbage collector in the HotSpot JVM which does not support pointer compression. Partly because the aforementioned address range restriction, and partly because the implementation of compressed pointers which exist in HotSpot would clash with the goals of the garbage collector. This project explores ways of implementing pointer compression for Z that isn't detrimental to the goals of the garbage collector, and aims to find where problems may occur. The outset was to explore compressing speculatively during runtime. The result is a design that relies on a custom bit layout for compressed pointers, inspecting bit layouts of the pointers on each read and write to detect the compression status. This seems to be the most promising in terms of code maintainability and ease of implementation.