Main Conference Program (Preliminary)


Day 1: September 23rd, Monday
8:00am-5:00pm Registration (Room: Foyer 3)
8:00-9:30am Keynote – Dr. Kathryn McKinley (Room: Emerald 1)
10:00am-12:00pm Session 1: Best-Papers Session (Room: Emerald 1)
  MASR: A Modular Accelerator for Sparse RNNs
  Gluon-Async: A Bulk-Asynchronous System for Distributed and Heterogeneous Graph Analytics
  Optimizing OpenMP Nested Parallel Regions with User-Level Threads
  SMT-COP: Defeating Side-Channel Attacks on Execution Units in SMT Processors
12:00-1:00pm Lunch Break (Room: Frol!k Buyout)
1:30-3:00pm Session 2A: Compiler Optimization and Code Generation 1 (Room: Emerald 1)
  Exploiting Accelerator Libraries Without Polluting Your Code
  Deepframe: A Profile-driven Compiler for Spatial Hardware Accelerators
  Fast Parallel Equivalence Relations in a Datalog Compiler
  Session 2B: Memory/Storage Systems 1 (Room: Emerald 2&3)
  Enforcing Last-level Cache Partitioning through Memory Virtual Channels
  To Stack or Not to Stack
  Enforcing Crash Consistency of Evolving Network Analytics in Non-Volatile Main Memory Systems
3:00-3:30pm Coffee Break
3:30-4:30pm Session 3A: Hardware/Software for Security (Room: Emerald 1)
  Fooling the Sense of Cross-core Last-level Cache Eviction based Attacker by Prefetching Common Sense
  SpecShield: Shielding Speculative Data from Microarchitectural Covert Channels
  Session 3B: Hardware/Software for Machine Learning (Room: Emerald 2&3)
  MOSAIC: Heterogeneity-, Communication-, and Constraint-Aware Model Slicing and Execution for Accurate and Efficient Inference
  Acorns: A Framework for Accelerating Deep Neural Networks with Input Sparsity
6:00-8:00pm Poster Session & Reception
Day 2: September 24th, Tuesday
8:00am-5:00pm Registration (Room: Foyer 3)
8:00-9:30am Keynote – Dr. Luis Ceze (Room: Emerald 1)
10:00-11:30am Session 4A: Concurrency Management (Room: Emerald 1)
  Forgive-TM: Supporting Lazy Conflict Detection In Eager Hardware Transactional Memory
  Unfair Scheduling Patterns in NUMA Architectures
  Optimizing Persistent Memory Transactions
  Session 4B: Heterogeneous Systems and Accelerators 1 (Room: Emerald 2&3)
  HeTM: Transactional Memory for Heterogeneous Systems
  Achieving scalability in a k-NN multi-GPU network service with Centaur
  Analyzing and Leveraging Remote-core Bandwidth for Enhanced Performance in GPUs
12:00am-1:00pm Lunch Break (Room: Frol!k Buyout)
1:30-3:00pm Session 5A: Domain/Application-Specific Hardware/Software (Room: Emerald 1)
  Specialization Opportunities in Graphical Workloads
  FindeR: Accelerating FM-Index-based Exact Pattern Matching in Genomic Sequences through ReRAM technology
  SLAMBooster: An Application-aware Online Controller for Approximation in Dense SLAM
  Session 5B: Heterogeneous Systems and Accelerators 2 (Room: Emerald 2&3)
  Exploring Memory Persistency Models for GPUs
  Adaptive Task Aggregation for High-Performance Sparse Solvers on GPUs
  EDGE: Event-Driven GPU Execution
3:00-3:30pm Coffee Break
3:30-4:30pm Session 6A: Compiler Optimization and Code Generation 2 (Room: Emerald 1)
  Generating Portable High-Performance Code via Multi-Dimensional Homomorphisms
  Absinthe: Learning an Analytical Performance Model to Fuse and Tile Stencil Codes in One Shot
  Session 6B: Memory/Storage Systems 2 (Room: Emerald 2&3)
  Reducing Data Movement and Energy in Multilevel Cache Hierarchies without Losing Performance: Can you have it all?
  Multiversioned Page Overlays: Enabling Faster Serializable Hardware Transactional Memory
Day 3: September 25th, Wednesday
8:00am-12:00pm Registration (Room: Foyer 3)
8:00-9:30am Keynote – Dr. Peter Kogge (Room: Emerald 1)
10:00am-12:00pm Session 7: Parallel Algorithms and Applications (Room: Emerald 1)
  Computing Three-dimensional Constrained Delaunay Refinement Using the GPU
  A Synchronization-Avoiding Distance-1 Grundy Coloring Algorithm for Power-Law Graphs
  Accelerating DCA++ (Dynamical Cluster Approximation) Scientific Application on the Summit supercomputer
  A Methodology for Characterizing Sparse Datasets and Its Application to SIMD Performance Prediction
10:00am-12:00pm Student Research Competition (SRC) Poster Session (Room: Belltown)