Carlos Maltzahn

Latest

• Extending Composable Data Services into SmartNICS (Best Paper Award)
• Mapping Out the HPC Dependency Chaos
• Rethinking basic primitives for store based systems
• Advancing Adoption of Reproducibility in HPC: A Preface to the Special Section
• Processing Particle Data Flows with SmartNICs
• Skyhook: Towards an Arrow-Native Storage System
• Skyhook: Bringing Computation to Storage with Apache Arrow
• Expanding the Scope of Artifact Evaluation at HPC Conferences: Experience of SC21
• Zero-Cost, Arrow-Enabled Data Interface for Apache Spark
• SkyhookDM: An Arrow-Native Storage System
• Zero-Cost, Arrow-Enabled Data Interface for Apache Spark
• Performance Characteristics of the BlueField-2 SmartNIC
• Towards an Arrow-native Storage System
• Enabling seamless execution of computational and data science workflows on HPC and cloud with the Popper container-native automation engine
• Mapping Scientific Datasets to Programmable Storage
• The CROSS Incubator: A Case Study for funding and training RSEs
• SkyhookDM: Storage and Management of Tabular Data in Ceph
• Scale-out Edge Storage Systems with Embedded Storage Nodes to Get Better Availability and Cost-Efficiency At the Same Time
• SkyhookDM: Data Processing in Ceph with Programmable Storage
• Is Big Data Performance Reproducible in Modern Cloud Networks?
• Scaling databases and file APIs with programmable Ceph object storage
• SkyhookDM: Programmable Storage for Datasets
• Popper 2.0: A Container-native Workflow Execution Engine For Testing Complex Applications and Validating Scientific Claims
• SkyhookDM: Mapping Scientific Datasets to Programmable Storage
• Towards Physical Design Management in Storage Systems
• MBWU: Benefit Quantification for Data Access Function Offloading
• Reproducible Computer Network Experiments: A Case Study Using Popper
• Quantifying benefits of offloading data management to storage devices
• MBWU (MibeeWu): Quantifying benefits of offloading data management to storage devices
• Skyhook: Programmable storage for databases
• Reproducible, Automated and Portable Computational and Data Science Experimentation Pipelines with Popper
• Spotting Black Swans With Ease: The Case for a Practical Reproducibility Platform
• Taming Performance Variability
• Should Storage Devices Stay Dumb or Become Smart?
• Tintenfisch: File System Namespace Schemas and Generators
• Popper Pitfalls: Experiences Following a Reproducibility Convention
• Cudele: An API and Framework for Programmable Consistency and Durability in a Global Namespace
• Programmable Caches with a Data Management Language & Policy Engine
• quiho: Automated Performance Regression Testing Using Inferred Resource Utilization Profiles
• Reproducible Computational and Data-Intensive Experimentation Pipelines with Popper
• Tintenfisch: File System Namespace Schemas and Generators
• Eusocial Storage Devices
• Eusocial Storage Devices - Offloading Data Management to Storage Devices that Can Act Collectively
• Integrating External Resources with a Task-Based Programming Model
• Optimized Scatter/Gather Data Operations for Parallel Storage
• DeclStore: Layering is for the Faint of Heart
• The Popper Convention: Making Reproducible Systems Evaluation Practical
• PopperCI: Automated Reproducibility Validation
• Malacology: A Programmable Storage System
• A Containerized Mesoscale Model and Analysis Toolkit to Accelerate Classroom Learning, Collaborative Research, and Uncertainty Quantification
• DAOS and Friends: A Proposal for an Exascale Storage System
• Exascale Storage Systems the SIRIUS Way
• Unum Arithmetic: Better Math with Clearer Tradeoffs
• Brados: Declarative,Programmable Object Storage
• TCP Inigo: Ambidextrous Congestion Control
• Collaborative WRF-based research and education with reproducible numerical weather prediction enabled by software containers
• The Case for Programmable Object Storage Systems
• ZEA, A Data Management Approach for SMR
• Characterizing and Reducing Cross-Platform Performance Variability Using OS-level Virtualization
• Popper: Making Reproducible Systems Performance Evaluation Practical
• Big Weather Web: A common and sustainable big data infrastructure in support of weather prediction research and education in universities
• Collaborative Research and Education with Numerical Weather Prediction Enabled by Software Containers
• I Aver: Providing Declarative Experiment Specifications Facilitates the Evaluation of Computer Systems Research
• Standing on the Shoulders of Giants by Managing Scientific Experiments Like Software
• Automatic and transparent I/O optimization with storage integrated application runtime support
• Mantle: A Programmable Metadata Load Balancer for the Ceph File System
• Tackling the Reproducibility Problem in Storage Systems Research with Declarative Experiment Specifications
• The Case for Programmable Object Storage Systems
• The Role of Container Technology in Reproducible Computer Systems Research
• Efficient, Failure Resilient Transactions for Parallel and Distributed Computing
• Erasure Coding & Read/Write Separation in Flash Storage
• An Innovative Storage Stack Addressing Extreme Scale Platforms and Big Data Applications
• Consistency and Fault Tolerance Considerations for the Next Iteration of the DOE Fast Forward Storage and IO Project
• Run, Fatboy, Run: Applying the Reduction to Uniprocessor Algorithm to Other Wide Resources
• Automatic Generation of Behavioral Hard Disk Drive Access Time Models
• Flash on Rails: Consistent Flash Performance through Redundancy
• SupMR: Circumventing Disk and Memory Bandwidth Bottlenecks for Scale-up MapReduce
• Gamification of Private Digital Data Archive Management
• Automatic Generation of Behavioral Hard Disk Drive Access Time Models
• Exploring Resource Migration using the CephFS Metadata cluster
• Big Weather - A workshop on overcoming barriers to distributed production, storage, and analysis of multi-model ensemble forecasts in support of weather prediction research and education in universities
• A Framework for an In-depth Comparison of Scale-up and Scale-out
• Efficient Transactions for Parallel Data Movement
• Exploring Trade-offs in Transactional Parallel Data Movement
• Fourier-Assisted Machine Learning of Hard Disk Drive Access Time Models
• High Performance & Low Latency in Solid-State Drives Through Redundancy
• SIDR: Structure-Aware Intelligent Data Routing in Hadoop
• In-Vivo Storage System Development
• Latency Minimization in SSD Clusters for Free
• I/O Acceleration with Pattern Detection
• DRepl: Optimizing Access to Application Data for Analysis and Visualization
• Ianus: Guaranteeing High Performance in Solid-State Drives
• In-Vivo Storage System Development
• High Performance & Low Latency in Solid-State Drives Through Redundancy
• Compressing Intermediate Keys between Mappers and Reducers in SciHadoop
• DataMods: Programmable File System Services
• Discovering Structure in Unstructured I/O
• SciHadoop Semantic Compression
• DataMods: Programmable File System Services
• Structure-Aware Intelligent Data Routing in SciHadoop
• Recommendation-based De-Identification | A Practical Systems Approach towards De-identification of Unstructured Text in Healthcare
• On the Role of Burst Buffers in Leadership-class Storage Systems
• Gdev: First-Class GPU Resource Management in the Operating System
• QMDS: a file system metadata management service supporting a graph data model-based query language
• FLAMBES: Evolving Fast Performance Models
• SciHadoop: Array-based Query Processing in Hadoop
• Modeling a Leadership-scale Storage System
• QMDS: A File System Metadata Management Service Supporting a Graph Data Model-based Query Language
• RAD-FLOWS: Buffering for Predictable Communication
• SciHadoop: Array-based Query Processing in Hadoop
• DRepl: Optimizing Access to Application Data for Analysis and Visualization
• PLFS and HDFS: Enabling Parallel Filesystem Semantics In The Cloud
• QMDS: A File System Metadata Service Supporting a Graph Data Model-Based Query Language
• Haceph: Scalable Metadata Management for Hadoop using Ceph
• RAID4S: Adding SSDs to RAID Arrays
• Design and Implementation of a Metadata-Rich File System
• Design and Implementation of a Metadata-Rich File System
• Enabling Scientific Application I/O on Cloud FileSystems
• InfoGarden: A Casual-Game Approach to Digital Archive Management
• RAID4S: Adding SSDs to RAID Arrays
• Ceph as a Scalable Alternative to the Hadoop Distributed File System
• Fusing Data Management Services with File Systems
• Mixing Hadoop and HPC Workloads on Parallel Filesystems
• JabberWocky: Crowd-Sourcing Metadata for Files
• Abstract Storage: Moving file format-specific abstractions into petabyte-scale storage systems
• Comparing the Performance of Different Parallel File system Placement Strategies
• Building a Parallel File System Simulator
• An Integrated Model for Performance Management in a Distributed System
• Efficient Guaranteed Disk Request Scheduling with Fahrrad
• Virtualizing Disk Performance
• Efficient Guaranteed Disk Request Scheduling with Fahrrad
• Adapting RAID Methods for Use in Object Storage Systems
• Dynamic Load Balancing in Ceph
• How Private are Home Directories?
• RADoN: QoS in Storage Networks
• Ringer: A Global-Scale Lightweight P2P File Service
• Virtualizing Disk Performance with Fahrrad
• RADOS: A Fast, Scalable, and Reliable Storage Service for Petabyte-scale Storage Clusters
• Searching and Navigating Petabyte Scale File Systems Based on Facets
• A File System Query Language
• Using Comprehensive Analysis for Performance Debugging in Distributed Storage Systems
• United States Patent 7,249,219: Method and Apparatus to Improve Buffer Cache Hit Rate
• Scaling Linux Storage to Petabytes
• End-to-end Performance Management for Scalable Distributed Storage
• Graffiti: A Framework for Testing Collaborative Distributed Metadata
• RADOS: A Reliable Autonomic Distributed Object Store
• Ceph: A Scalable, High-Performance Distributed File System
• CRUSH: Controlled, Scalable, Decentralized Placement of Replicated Data
• LiFS: An Attribute-Rich File System for Storage Class Memories
• Ceph: A Scalable Object-based Storage System
• CRUSH: Controlled, Scalable, Decentralized Placement of Replicated Data
• Graffiti: A Framework for Testing Collaborative Distributed Metadata
• POTSHARDS: Storing Data for the Long-term Without Encryption
• Richer File System Metadata Using Links and Attributes
• Reducing the Disk I/O of Web Proxy Server Caches
• On Bandwidth Smoothing
• A Feasibility Study of Bandwidth Smoothing on the World-Wide Web Using Machine Learning
• Improving Resource Utilization of Enterprise-Level World-Wide Web Proxy Servers
• Performance Issues of Enterprise Level Web Proxies
• Performance Issues of Enterprise Level Web Proxies
• The Chautauqua Workflow System
• Digital's Web Proxy Traces
• Community Help: Discovering Tools and Locating Experts in a Dynamic Environment
• Collaboration with Spreadsheets
• Integrating object and agent worlds
• Sharing Processes: Team Coordination in Design Repositories
• A Decision-Based Configuration Process Environment
• ConceptTalk: Kooperationsunterstützung in Softwareumgebungen