Ocean uses HyperLogLog estimators to skip the costly symbolic phase of GPU SpGEMM, pairs it with dynamic workflow choice and a shared-plus-global hash accumulator, and reports 1.4-2.8x speedups over prior GPU implementations.
Awad, Saman Ashkiani, Rob Johnson, Martín Farach-Colton, and John D
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
verdicts
UNVERDICTED 3representative citing papers
FliX flips the indexing paradigm by mapping GPU compute to data buckets and using batch binary search, delivering up to 6.5x lower query latency and 4x better throughput per memory than leading GPU B-trees and LSM-trees.
Generalizes rely-guarantee to parametric memory models and presents Piccolo, the first such logic for causally consistent shared memory using potential-based operational semantics.
citing papers explorer
-
Ocean: Fast Estimation-Based Sparse General Matrix-Matrix Multiplication on GPU
Ocean uses HyperLogLog estimators to skip the costly symbolic phase of GPU SpGEMM, pairs it with dynamic workflow choice and a shared-plus-global hash accumulator, and reports 1.4-2.8x speedups over prior GPU implementations.
-
FliX: Flipped-Indexing for Scalable GPU Queries and Updates
FliX flips the indexing paradigm by mapping GPU compute to data buckets and using batch binary search, delivering up to 6.5x lower query latency and 4x better throughput per memory than leading GPU B-trees and LSM-trees.
-
Rely-Guarantee Reasoning for Causally Consistent Shared Memory (Extended Version)
Generalizes rely-guarantee to parametric memory models and presents Piccolo, the first such logic for causally consistent shared memory using potential-based operational semantics.