Hyesoon Kim – författare

Singular poetry made through censorship, elusion, and language renewal_x000D__x000D_The astonishing poetry collection The Hell of That Star enlivens the horror of Korean life under U.S.-backed authoritarianism. Poems of blows and vomit, births and coffins alternate blithe confidence and trembling terror. When slapped seven times by a government censor, Kim responded with defiant poems. The death of language becomes a death of the writer; within death, Kim finds new life in fragmentation and reorientation. This singular volume provides a wild and rigorous study of the words of the nation-state and the self, as well as the deprivations, detainments, and surprises in between. In evading censorship, Kim's poems question, twist, and transmute; language is a site where the personal and political meet to escape containment, emptiness, and domestication. The book includes an essay by the author, with an introduction and notes by the translator._x000D__x000D_[sample poem]_x000D__x000D_The tough after all_x000D_we still remain_x000D_and just in gathering it is lovingly_x000D_even while building each other's tombs_x000D_while patting each other's backs_x000D__x000D_But when each bird turns around_x000D_their arms flung! open_x000D_embracing tightly what_x000D_they do not even recognize as their grave_x000D_and they hug and hold harder and harder_x000D_stretching four limbs out over the laid sleeping mat and blanket_x000D_saying I love you I love you even in their sleep_x000D_In this world from which crying birds have disappeared_x000D_only I am left

General-purpose graphics processing units (GPGPU) have emerged as an important class of shared memory parallel processing architectures, with widespread deployment in every computer class from high-end supercomputers to embedded mobile platforms. Relative to more traditional multicore systems of today, GPGPUs have distinctly higher degrees of hardware multithreading (hundreds of hardware thread contexts vs. tens), a return to wide vector units (several tens vs. 1-10), memory architectures that deliver higher peak memory bandwidth (hundreds of gigabytes per second vs. tens), and smaller caches/scratchpad memories (less than 1 megabyte vs. 1-10 megabytes). In this book, we provide a high-level overview of current GPGPU architectures and programming models. We review the principles that are used in previous shared memory parallel platforms, focusing on recent results in both the theory and practice of parallel algorithms, and suggest a connection to GPGPU platforms. We aim to provide hints to architects about understanding algorithm aspect to GPGPU. We also provide detailed performance analysis and guide optimizations from high-level algorithms to low-level instruction level optimizations. As a case study, we use n-body particle simulations known as the fast multipole method (FMM) as an example. We also briefly survey the state-of-the-art in GPU performance analysis tools and techniques. Table of Contents: GPU Design, Programming, and Trends / Performance Principles / From Principles to Practice: Analysis and Tuning / Using Detailed Performance Analysis to Guide Optimization

General-purpose graphics processing units (GPGPU) have emerged as an important class of shared memory parallel processing architectures, with widespread deployment in every computer class from high-end supercomputers to embedded mobile platforms. Relative to more traditional multicore systems of today, GPGPUs have distinctly higher degrees of hardware multithreading (hundreds of hardware thread contexts vs. tens), a return to wide vector units (several tens vs. 1-10), memory architectures that deliver higher peak memory bandwidth (hundreds of gigabytes per second vs. tens), and smaller caches/scratchpad memories (less than 1 megabyte vs. 1-10 megabytes). In this book, we provide a high-level overview of current GPGPU architectures and programming models. We review the principles that are used in previous shared memory parallel platforms, focusing on recent results in both the theory and practice of parallel algorithms, and suggest a connection to GPGPU platforms. We aim to provide hints to architects about understanding algorithm aspect to GPGPU. We also provide detailed performance analysis and guide optimizations from high-level algorithms to low-level instruction level optimizations. As a case study, we use n-body particle simulations known as the fast multipole method (FMM) as an example. We also briefly survey the state-of-the-art in GPU performance analysis tools and techniques. Table of Contents: GPU Design, Programming, and Trends / Performance Principles / From Principles to Practice: Analysis and Tuning / Using Detailed Performance Analysis to Guide Optimization

Die Berliner Rede zur Poesie wird am 11. Juni 2023 von der südkoreanischen Lyrikerin Kim Hyesoon gehalten.Die Berliner Rede zur Poesie 2023 wird die 1955 in Uljin (Provinz Nord-Kyŏngsang) geborene Kim Hyesoon halten. Sie studierte koreanische Literatur und erlangte in den späten 90er Jahren große Bekanntheit. Ihr literarisches Debüt "Poet`Smoking a Cigarette" erschien 1979. Kims Virtuosität als Dichterin liegt in ihrer Fähigkeit, mit experimenteller Sprache einzigartige poetische Bilder zu schaffen und gleichzeitig ein tiefes Interesse am so genannten weiblichen Schreiben zu bewahren, das in weiblichen Erfahrungen und Seinsweisen wurzelt. Ihre zentralen, in den unkonventionellen Arbeiten verarbeiteten Themen sind sowohl die Emanzipation als auch das Streben nach Freiheit in der koreanischen Gesellschaft. Sie gehört zu den bekanntesten modernen Lyrikerinnen Koreas. Derzeit lehrt sie Kreatives Schreiben am Seoul Institute of the Arts.