All Stories

  1. Scalable and Adaptive Parallel Training of Graph Transformer on Large Graphs
  2. MLAM: A Machine Learning-Aided Architectural BottleneckAnalysis Model for x86 Architectures
  3. AnyKey: A Key-Value SSD for All Workload Types
  4. Load and MLP-Aware Thread Orchestration for Recommendation Systems Inference on CPUs
  5. Pirate: No Compromise Low-Bandwidth VR Streaming for Edge Devices
  6. Towards SLO-Compliant and Cost-Effective Serverless Computing on Emerging GPU Architectures
  7. SpotVerse: Optimizing Bioinformatics Workflows with Multi-Region Spot Instances in Galaxy and Beyond
  8. SmartGraph: A Framework for Graph Processing in Computational Storage
  9. FAAStloop: Optimizing Loop-Based Applications for Serverless Computing
  10. Foveated HDR: Efficient HDR Content Generation on Edge Devices Leveraging User's Visual Attention
  11. GameStreamSR: Enabling Neural-Augmented Game Streaming on Commodity Mobile Platforms
  12. Thorough Characterization and Analysis of Large Transformer Model Training At-Scale
  13. Thorough Characterization and Analysis of Large Transformer Model Training At-Scale
  14. Minimizing Coherence Errors via Dynamic Decoupling
  15. Thorough Characterization and Analysis of Large Transformer Model Training At-Scale
  16. Impact of Write-Allocate Elimination on Fujitsu A64FX
  17. MBFGraph: An SSD-based External Graph System for Evolving Graphs
  18. Hardware Support for Constant-Time Programming
  19. Architecture-Aware Currying
  20. Quantifying and Mitigating Cache Side Channel Leakage with Differential Set
  21. Optimizing CPU Performance for Recommendation Systems At-Scale
  22. EdgePC: Efficient Deep Learning Analytics for Point Clouds on Edge Devices
  23. Cypress
  24. Multi-resource fair allocation for consolidated flash-based caching systems
  25. Fine-Granular Computation and Data Layout Reorganization for Improving Locality
  26. An architecture interface and offload model for low-overhead, near-data, distributed accelerators
  27. Skipper: Enabling efficient SNN training through activation-checkpointing and time-skipping
  28. Pushing Point Cloud Compression to the Edge
  29. End-to-end Characterization of Game Streaming Applications on Mobile Platforms
  30. Memory Space Recycling
  31. Data Convection
  32. Data Convection
  33. End-to-end Characterization of Game Streaming Applications on Mobile Platforms
  34. Memory Space Recycling
  35. A Scheduling Framework for Decomposable Kernels on Energy Harvesting IoT Edge Nodes
  36. End-to-end Characterization of Game Streaming Applications on Mobile Platforms
  37. Memory Space Recycling
  38. Data Convection
  39. Kraken
  40. SpecSafe: detecting cache side channels in a speculative world
  41. Mix and Match: Reorganizing Tasks for Enhancing Data Locality
  42. Distance-in-time versus distance-in-space
  43. Fluid: a framework for approximate concurrency via controlled dependency relaxation
  44. Mix and Match: Reorganizing Tasks for Enhancing Data Locality
  45. Mix and Match: Reorganizing Tasks for Enhancing Data Locality
  46. Ghost Thread
  47. SplitServe
  48. Fifer
  49. Implications of Public Cloud Resource Heterogeneity for Inference Serving
  50. Minimal Variance Sampling with Provable Guarantees for Fast Training of Graph Neural Networks
  51. Déjà View: Spatio-Temporal Compute Reuse for‘ Energy-Efficient 360° VR Video Streaming
  52. Computing with Near Data
  53. Quantifying Data Locality in Dynamic Parallelism in GPUs
  54. Architecture-Aware Approximate Computing
  55. Distilling the Essence of Raw Video to Reduce Memory Usage and Energy at Edge Devices
  56. Architecture-Aware Approximate Computing
  57. Co-optimizing memory-level parallelism and cache-level parallelism
  58. Quantifying Data Locality in Dynamic Parallelism in GPUs
  59. NEOFog
  60. ReveNAND
  61. Enhancing computation-to-core assignment with physical location information
  62. Enhancing computation-to-core assignment with physical location information
  63. Exploiting Data Longevity for Enhancing the Lifetime of Flash-based Storage Class Memory
  64. Exploiting Data Longevity for Enhancing the Lifetime of Flash-based Storage Class Memory
  65. A Study on Performance and Power Efficiency of Dense Non-Volatile Caches in Multi-Core Systems
  66. Hardware-Software Co-design to Mitigate DRAM Refresh Overheads
  67. Exploiting Intra-Request Slack to Improve SSD Performance
  68. Exploiting Intra-Request Slack to Improve SSD Performance
  69. Hardware-Software Co-design to Mitigate DRAM Refresh Overheads
  70. VIP
  71. A case for core-assisted bottleneck acceleration in GPUs
  72. Anatomy of GPU Memory System for Multi-Application Execution
  73. Optimizing off-chip accesses in multicores
  74. EECache
  75. Memory Row Reuse Distance and its Role in Optimizing Application Performance
  76. Memory Row Reuse Distance and its Role in Optimizing Application Performance
  77. VIP
  78. A case for core-assisted bottleneck acceleration in GPUs
  79. Network footprint reduction through data access and computation placement in NoC-based manycores
  80. Optimizing off-chip accesses in multicores
  81. TaPEr
  82. Volatile STT-RAM Scratchpad Design and Data Allocation for Low Energy
  83. Application-aware Memory System for Fair and Efficient Execution of Concurrent GPGPU Applications
  84. Trading cache hit rate for memory performance
  85. Orchestrated scheduling and prefetching for GPGPUs
  86. Performance enhancement under power constraints using heterogeneous CMOS-TFET multicores
  87. Physically addressed queueing (PAQ)
  88. A compiler framework for extracting superword level parallelism
  89. A compiler framework for extracting superword level parallelism