No description available.

Saarbrücken, Univ., Diss., 1992. / Computerdatei im Fernzugriff.

Universiẗat, Diss., 1992--Saarbrücken.

Stuttgart, Univ., Fakultät Informatik, Diplomarb., 1999.

Zugl.: Paderborn, University-Gesamthochsch., Diss., 1995.

REPLICA is a family of novel scalable chip multiprocessors with configurable emulated shared memory architecture, whose computation model is based on the PRAM (Parallel Random Access Machine) model. The purpose of this thesis is to, by benchmarking different types of computation problems on REPLICA, similar parallel architectures (SB-PRAM and XMT) and more diverse ones (Xeon X5660 and Tesla M2050), evaluate how REPLICA is positioned among other existing architectures, both in performance and programming effort. But it should also examine if REPLICA is more suited for any special kinds of computational problems. By using some of the well known Berkeley dwarfs, and input from unbiased sources, such as The University of Florida Sparse Matrix Collection and Rodinia benchmark suite, we have made sure that the benchmarks measure relevant computation problems. We show that today’s parallel architectures have some performance issues for applications with irregular memory access patterns, which the REPLICA architecture can solve. For example, REPLICA only need to be clocked with a few MHz to match both Xeon X5660 and Tesla M2050 for the irregular memory access benchmark breadth first search. By comparing the efficiency of REPLICA to a CPU (Xeon X5660), we show that it is easier to program REPLICA efficiently than today’s multiprocessors. / REPLICA är en grupp av konfigurerbara multiprocessorer som med hjälp utav ett emulerat delat minne realiserar PRAM modellen. Syftet med denna avhandling är att genom benchmarking av olika beräkningsproblem på REPLICA, liknande (SB-PRAM och XMT) och mindre lika (Xeon X5660 och Tesla M2050) parallella arkitekturer, utvärdera hur REPLICA står sig mot andra befintliga arkitekturer. Både prestandamässigt och hur enkel arkitekturen är att programmera effektiv, men även försöka ta reda på om REPLICA är speciellt lämpad för några särskilda typer av beräkningsproblem. Genom att använda välkända Berkeley dwarfs applikationer och opartisk indata från bland annat The University of Florida Sparse Matrix Collection och Rodinia benchmark suite, säkerställer vi att det är relevanta beräkningsproblem som utförs och mäts. Vi visar att dagens parallella arkitekturer har problem med prestandan för applikationer med oregelbundna minnesaccessmönster, vilken REPLICA arkitekturen kan vara en lösning på. Till exempel, så behöver REPLICA endast vara klockad med några få MHz för att matcha både Xeon X5660 och Tesla M2050 för algoritmen breadth first search, vilken lider av just oregelbunden minnesåtkomst. Genom att jämföra effektiviteten för REPLICA gentemot en CPU (Xeon X5660), visar vi att det är lättare att programmera REPLICA effektivt än dagens multiprocessorer.

Pram is more complicated product, than you may think. Regular first opinion puts pram into some sort of toy but it is exactly contrary. This product contains user’s demands from two point of view - parents and baby. These users have different needs and - as we are get used today - we want to have not just working stuffs but modern, nice and make it naturally like part of our lifes. We can’t prefer design and leave functionality behind or make it really complicated to produce just with efford create something new. In this project I tried find compromise and come out with design with new features, ideas and experimental look but with same function ability. Main point of designing the pram was create universal pram with easy use during whole first 3 baby periods (from 0 to 2 years old baby) and decrease the final number of needed parts. I focused on original design and carriage parts which can be formed according to baby needs. Create universall pram but also easy to transport and fold. There are two main features designed completelly from scratch: modifiable carriage part and design attractive no center axis wheels.

This master's thesis discusses the design and implementation of a simulator for the REPLICA architecture, a many-core PRAM-like machine. REPLICA provides a programming model that seemingly cannot be provided by mainstream hardware without significant slowdown compared to traditional models. This also implies that it is difficult to simulate REPLICA's programming model on mainstream hardware. Simulator design decisions are described and the resulting simulator is evaluated and compared to existing simulators, where we see that the simulator presented in this thesis is the fastest of them. As seen from the discussion focus in the thesis, most efforts were directed towards simulator execution speed rather than user-facing features.

REPLICA is a PRAM-NUMA hybrid architecture, with support for instructionlevel parallelism as a VLIW architecture. REPLICA can also chain instructionsso that the output from an earlier instruction can be used as input to a laterinstruction in the same execution step. There are plans in the REPLICA project to develop a new C-based program-ming language, compilers and libraries to speed up development of parallel pro-grams. We have developed a LLVM back-end as a part of the REPLICA projectthat can be used to generate code for the REPLICA architecture. We have alsocreated a simple optimization algorithm to make better use of REPLICAs supportfor instruction level parallelism. Some changes to Clang, LLVMs front-end forC/C++/Objective-C, was also necessary so that we could use assembler in-liningin our REPLICA programs. Using Clang to compile C-code to LLVMs internal representation and LLVMwith our REPLICA back-end to transform LLVMs internal representation intoMBTAC1 assembler. / REPLICA är en VLIW liknande PRAM-NUMA arkitektur, med möjlighet för attkedja ihop instruktioner så att resultat från tidigare instruktioner kan användassom indata till nästa instruktion i samma exekveringssteg. Inom REPLICA projetet finns planer på att utecklar ett nytt C-baserat pro-grammeringsspråk, kompilatorer och bibliotek för att snabbba upp utvecklingen avparallella program. Som en del av REPLICA projektet har vi utvecklat ett kompi-lator back-end för LLVM som kan användas för att generera kod till REPLICA. Vihar även utvecklat en enklare optimerings algoritm för att bättre utnyttja REPLI-CAs förmåga för instruktions parallelisering. Vi har även gjort ändringar i Clang,LLVMs front-end för C/C++/Objective-C, så att vi kan använda inline assembleri REPLICA program. Med Clang kan man kompilera C-kod till LLVMs interna representation somi sin tur genom LLVM och REPLICA back-end kan omvandlas till MBTAC3 as-sembler.

April 6, 2007 I submitted the first draft of this final project to Sabina Murray’s workshop in the fall of 2004. I was mainly concerned with narrative then. I wanted the story to move. I wasn’t interested in writing as much as providing a framework. I told myself, there’ll be time enough for writing, the creative kind—keep the flourishes to a minimum for now. The result was an outline with flimsy characters sifted in. Workshop feedback suggested the same. (Sabina also caught a malapropism I’ve not been allowed to forget: I described a character’s long legs as like those of a gelding. I meant yearling, I think. We all had a laugh. In a draft before this final project version Sabina snagged another one: ‘braying’ used when ‘rearing’ was intended. For a lobster.) I didn’t work much on that first draft for the next year, focused as I was on short stories. (During that time I planned on submitting a collection as my final project. I had six or seven that I thought were pretty fit and worth more work.) In the spring of 2006 I went at the novel more consistently. By the summer I was working on it daily, which has more or less been the case since. The basic narrative of the second attempt was the same: guy leaves corporate job to work on lobster boat on Cape Cod, meets many characters in that insular world, one of whom may be hauling other lobstermen’s pots. Guy also has love interest. Through that narrative I meant to explore the development of D.L., who saw himself as a lost, alienated 30-year-old with potential but unable to fit in anywhere. His journey from New York, the site of his two most recent failures, to Orleans would yield new and different perceptions. I like the idea that a protagonist continues to do what he/she’s always done until circumstances of the narrative prompt change. D.L. was meant to see that many of his problems existed only in his head. I read Dostoyevsky’s The Double in David Lenson’s Individualism and its Discontents course (spring 2005). It got me thinking again about self consciousness. I didn’t intend for D.L. to descend into paranoid madness like Golyadkin but The Double served as an example of how I might explore D.L. more than I had on the first go round. Changing perception was the broad theme to work through the narrative. I would fill D.L. out more, lend him more introspection. The challenge seemed to be in gauging when and how much the story should be in his head. I started doing an end around. I thought I could introduce new perceptions through other characters. Hence Jack’s early talk of cliffs moving. In previous iterations, D.L. ruminated more on that non sequitur. He was talking to himself and it seemed too obvious: Was Jack saying my problems were in my head? Another concern is that D.L.’s thoughts often felt and still do at times feel like commentary. The scene in Wellfleet, which is mostly dialogue between D.L. and Suzie, is an example. Dreamlike moments—the cliffs, high tide flowing as a river down Jack’s street, D.L.’s faux dream, Jack’s real dream, even H. erectus walking among us—are attempts to create more of D.L.’s world, to create more of a feeling. To whatever extent those moments are effective, I think there should be more. My biggest concern with this effort is that there’s still too much dialogue. I don’t know if I’ve lost some creative vigor or if the paucity of literary prose is a result of the process, particularly the initial narrative concern of wanting to keep the story moving. Or something else. But I’ve felt less natural, less like myself as a writer throughout much of the process of writing this novel. I handed over a new draft to committee this past winter. Feedback included concerns over the dialogue itself, that the characters weren’t moving while talking, weren’t described enough; missing connective tissue between chapters; problems with moving around in time rather than taking a more linear approach; and too much in the way of lobstering details. These are the areas I’ve focused on since. Switching to linear time has made a big difference, I think. I’ve worked on movement and dialogue but there are still long chunks in which characters are talking more than doing (as related to above creative concerns). Lobstering details have been a concern from the start as I didn’t want this to be Lobstering 101, but also because it can clog the narrative. I’ve often struggled with feeling the need to include information that isn’t necessary and gets in the way. It’s a tough habit to break for me but I think I’ve gotten better in that regard in my years in the program, particularly with short stories, in which more concision is a requirement because of space. The pages that follow are a considerable improvement from that first draft of 2004. I see this as the first half of the novel. I imagine the second half will be about as long. I’m not crazy about the title but I have a scene in mind at or near the end that makes sense of it. ‘Thin water’ is clear water, not turbid. - Jeremy Church