The proceedings contain 8 papers. The topics discussed include: integration of burst buffer in high-level parallel I/O library for exa-scale computing era;using a robust metadata management system to accelerate scient...
ISBN:
(纸本)9781728101927
The proceedings contain 8 papers. The topics discussed include: integration of burst buffer in high-level parallel I/O library for exa-scale computing era;using a robust metadata management system to accelerate scientific discovery at extreme scales;evaluation of HPC application I/O on object storage systems;pufferbench: evaluating and optimizing malleability of distributed storage;understanding SSD reliability in large-scale cloud systems;characterizing deep-learning I/O workloads in TensorFlow;toward understanding I/O behavior in HPC workflows;and methodology for the rapid development of scalable HPC data services.
The proceedings contain 10 papers. The topics discussed include: Par-BF: a parallel partitioned bloom filter for dynamic data sets;dispel4py: a python framework for data-intensive scientific computing;efficient, failu...
ISBN:
(纸本)9781479970384
The proceedings contain 10 papers. The topics discussed include: Par-BF: a parallel partitioned bloom filter for dynamic data sets;dispel4py: a python framework for data-intensive scientific computing;efficient, failure resilient transactions for parallel and distributed computing;BPAR: a bundle-based parallel aggregation framework for decoupled I/O execution;rethinking key-value store for parallel I/O optimization;PSA: a performance and space-aware data layout scheme for hybrid parallel file systems;distributed multipath routing algorithm for data center networks;CULZSS-Bit: a bit-vector algorithm for lossless data compression on GPGPUs;a caching approach to reduce communication in graph search algorithms;and mapping of RAID controller performance data to the job history on large computingsystems.
The proceedings contain 10 papers. The topics discussed include: SDAFT: a novel scalabledata access framework for parallel BLAST;large data and computation in a hazard map workflow using Hadoop and Neteeza architectu...
ISBN:
(纸本)9781450325066
The proceedings contain 10 papers. The topics discussed include: SDAFT: a novel scalabledata access framework for parallel BLAST;large data and computation in a hazard map workflow using Hadoop and Neteeza architectures;a framework for an in-depth comparison of scale-up and scale-out;BDMPI: conquering big data with small clusters using mpi;on the core affinity and file upload performance of Hadoop;enhancing both fairness and performance using rate-aware dynamic storage cache partitioning;design of an active storage cluster file system for dag workflows;and novel parallel method for mining frequent patterns on multi-core shared memory systems.
The proceedings contain 10 papers. The topics discussed include: scientific workflows at dataWarp-speed: accelerated data-intensive science using NERSC's burst buffer;parallel I/O characterisation based on server-...
ISBN:
(纸本)9781509052165
The proceedings contain 10 papers. The topics discussed include: scientific workflows at dataWarp-speed: accelerated data-intensive science using NERSC's burst buffer;parallel I/O characterisation based on server-side performance counters;replicating HPC I/O workloads with proxy applications;can non-volatile memory benefit MapReduce applications on HPC clusters?;FatMan vs. LittleBoy: scaling up linear algebraic operations in scale-out data platforms;Klimatic: a virtual data lake for harvesting and distribution of geospatial data;get out of the way! applying compression to internal data structures;towards energy efficient data management in HPC: the open Ethernet drive approach;a generic framework for testing parallel file systems;and a bloom filter based scalabledata integrity check tool for large-scale dataset.
The proceedings contain 9 papers. The topics discussed include: SJM: an SCM-based journaling mechanism with write reduction for file systems;efficient disk-to-disk sorting: a case study in the decoupled execution para...
ISBN:
(纸本)9781450339933
The proceedings contain 9 papers. The topics discussed include: SJM: an SCM-based journaling mechanism with write reduction for file systems;efficient disk-to-disk sorting: a case study in the decoupled execution paradigm;a low-cost adaptive data separation method for the flash translation layer of solid state drives;big data analytics on traditional HPC infrastructure using two-level storage;route-aware independent MPI I/O on the Blue Gene/Q;experimental evaluation of a flexible I/O architecture for accelerating workflow engines in cloud environments;a case study of MapReduce speculation for failure recovery;supporting online analytics with user-defined estimation and early termination in a MapReduce-like framework;and performance evaluation and tuning of BioPig for genomic analysis.
The proceedings contain 10 papers. The topics discussed include: optimized scatter/gather data operations for parallel storage;software-defined storage for fast trajectory queries using a DeltaFS indexed massive direc...
ISBN:
(纸本)9781450351348
The proceedings contain 10 papers. The topics discussed include: optimized scatter/gather data operations for parallel storage;software-defined storage for fast trajectory queries using a DeltaFS indexed massive directory;CoSS: proposing a contract-based storage system for HPC;EMPRESS extensible metadata provider for extreme-scale scientific simulations;taming metadata storms in parallel filesystems with MetaFS;architecting HBM as a high bandwidth, high capacity, self-managed last-level cache;diving into petascale production file systems through large scale profiling and analysis;performance analysis of emerging data analytics and HPC workloads;and toward scalable monitoring on large-scale storage for software defined cyberinfrastructure.
暂无评论