AMPLab Publications

• SparkR: Scaling R Programs with Spark
• Time-Evolving Graph Processing at Scale
• Ernest: Efficient Performance Prediction for Large-Scale Advanced Analytics
• BlowFish: Dynamic Storage-Performance Tradeoff in Data Stores
• FairRide: Near-Optimal, Fair Cache Sharing
• SparkNet: Training Deep Networks on Spark
• Succinct: Enabling Queries on Compressed Data
• FastLane: Making Short Flows Shorter with Agile Drop Notification
• Low Latency, Geo-distributed Data Analytics
• Efficient Coflow Scheduling Without Prior Knowledge
• Feral Concurrency Control: An Empirical Investigation of Modern Application Integrity
• G-OLA: Generalized Online Aggregation for Interactive Analysis on Big Data
• CellIQ: Real-Time Cellular Network Analytics at Scale
• Tachyon: Reliable, Memory Speed Storage for Cluster Computing Frameworks
• Coordination Avoidance in Database Systems
• Discretized Streams: Fault-Tolerant Streaming Computation at Scale
• The Power of Choice in Data-Aware Cluster Scheduling
• GraphX: Graph Processing in a Distributed Dataflow Framework
• Efficient Coflow Scheduling with Varys
• Knowing When You’re Wrong: Building Fast and Reliable Approximate Query Processing Systems
• Scalable Atomic Visibility with RAMP Transactions
• Tachyon: Memory Throughput I/O for Cluster Computing Frameworks
• GraphX: Unifying Data-Parallel and Graph-Parallel Analytics
• Sparrow: Distributed, Low Latency Scheduling
• Carat: Collaborative Energy Diagnosis for Mobile Devices
• Highly Available Transactions: Virtues and Limitations
• A General Bootstrap Performance Diagnostic
• Leveraging Endpoint Flexibility in Data-Intensive Clusters
• GraphX: A Resilient Distributed Graph System on Spark
• HAT, not CAP: Towards Highly Available Transactions
• PBS at Work: Advancing Data Management with Consistency Metrics (Demo)
• Bolt-on Causal Consistency
• The Case for Tiny Tasks in Compute Clusters
• Choosy: Proportional Sharing for Datacenter Jobs with Constraints
• BlinkDB: Queries with Bounded Errors and Bounded Response Times on Very Large Data (Best Paper Award)
• Blink and It’s Done: Interactive Queries on Very Large Data
• Fast and Interactive Analytics over Hadoop Data with Spark
• Sweet Storage SLOs with Frosting
• Cake: Enabling High-level SLOs on Shared Storage Systems
• Shark: SQL and Rich Analytics at Scale
• Tradeoffs in CDN Designs for Throughput Oriented Traffic
• The Potential Dangers of Causal Consistency and an Explicit Solution
• Discretized Streams: An Efficient and Fault-Tolerant Model for Stream Processing on Large Clusters
• Coflow: A Networking Abstraction for Cluster Applications
• Carat: Collaborative Energy Debugging for Mobile Devices
• FairCloud: Sharing The Network in Cloud Computing
• Surviving Failures in Bandwidth-Constrained Datacenters
• Probabilistically Bounded Staleness for Practical Partial Quorums
• A Case for Performance-Centric Network Allocation
• Faster and More Accurate Sequence Alignment with SNAP
• Blue-Fi: Enhancing Wi-Fi Performance using Bluetooth Signals
• PACMan: Coordinated Memory Caching for Parallel Jobs
• Reining in the Outliers in MapReduce Clusters using Mantri
• Re-optimizing Data Parallel Computing
• Shark: Fast Data Analysis Using Coarse-grained Distributed Memory (Best Demo Award)
• A Common Substrate for Cluster Computing
• Improving MapReduce Performance in Heterogeneous Environments
• Resilient Distributed Datasets: A Fault-Tolerant Abstraction for In-Memory Cluster Computing (Best Paper Award)
• Scarlett: Coping with Skewed Popularity Content in MapReduce Clusters
• The Datacenter Needs an Operating System
• Mesos: A Platform for Fine-Grained Resource Sharing in the Data Center
• Dominant Resource Fairness: Fair Allocation of Multiple Resources Types
• Spark: Cluster Computing with Working Sets
• Managing Data Transfers in Computer Clusters with Orchestra
• Disk-Locality in Datacenter Computing Considered Irrelevant