Program

The SPAA proceedings can be accessed here.

Friday, June 16th (Workshops and Tutorials)

09:00 – 11:00 Tutorial: SpeedCode: Software performance engineering education via Coding of didactic exercises

11:15 – 12:30 Tutorial: Tutorial on Approximate Nearest Neighbor Search (ANNS) – Techniques and Open Problems

12:30 – 14:00 Lunch Break

14:00 – 18:00 Workshop on Highlights of Parallel Computing (HOPC)

in parallel with

14:00 – 18:00 Workshop on Filters: From Bloom to Quotient and Everything in Between

18:00 – 20:30 Welcome Reception / Poster Sessions for HOPC

Saturday, June 17

8:30-8:45am Opening remarks

Session 1: Data Structures / Scheduling

8:45-9:06am PIM-Trie: A Skew-Resistant Trie for Processing-in-Memory
9:06-9:27am Quancurrent: A Concurrent Quantiles Sketch
9:27-9:48am An Efficient Scheduler for Interactive Task-Parallel Applications
9:48-10:09am Efficient Synchronization-Light Work Stealing
10:09-10:30am Balanced Allocation in Batches: The Tower of Two Choices

10:30-11am Break

11am-12:15pm SPAA Parallel Computing Award Keynote – Guy Blelloch

12:30-2pm Lunch

Session 2: Distributed Algorithms

2:00-2:21pm On Parallel k-Center Clustering
2:21-2:42pm Massively Parallel Tree Embeddings for High-Dimensional Spaces
2:42-3:03pm Deterministic Massively Parallel Symmetry Breaking for Sparse Graphs
3:03-3:24pm Location-Sensitive String Problems in MPC
3:24-3:30pm Brief Announcement: Regular and Dyck Languages in MPC
3:30-4pm Break

Session 3: Caching / Networks

4:00-4:21pm An Associativity Threshold Phenomenon in Set-Associative Caches
4:21-4:42pm Increment–and–Freeze: Every Cache, Everywhere, All of the Time
4:42-5:03pm Multidimensional Approximate Agreement with Asynchronous Fallback
5:03-5:24pm A Tight Characterization of Fast Failover Routing: Resiliency to Two Link Failures is Possible
5:24-5:45pm In-network Allreduce with Multiple Spanning Trees on PolarFly

6-8pm Business Meeting

Sunday, June 18

Session 4: Concurrency

8:45-9:06am Releasing Memory with Optimistic Access: A Hybrid Approach to Memory Reclamation and Allocation in Lock-Free Programs
9:06-9:27am Transactional Composition of Nonblocking Data Structures
9:27-9:48am Protecting Locks Against Unbalanced Unlock()
9:48-10:09am Applying Hazard Pointers to More Concurrent Data Structures
10:09-10:30am A Time and Space Efficient Lock with Dynamic Joining under System-wide Failures

10:30-11am Break

11am-12:15pm SPAA Test-of-Time Award Keynote – Bradley Kuszmaul and Charles Leiserson

12:30-2pm Lunch

Session 5: Best Paper Candidates

2:00-2:21pm Almost Optimal Massively Parallel Algorithms for k-Center Clustering and Diversity Maximization
2:21-2:42pm Nearly optimal parallel algorithms for longest increasing subsequence
2:42-3:03pm Provably-Efficient and Internally-Deterministic Parallel Union-Find
3:03-3:24pm Nearly Work-Efficient Parallel DFS in Undirected Graphs

3:30-4pm Break

Session 6: Brief Announcements

4:00-4:06pm Brief Announcement: On Solving Recoverable Mutual Exclusion Under System-Wide Failures
4:06-4:12pm Brief Announcement: A Parallel Architecture for Dynamic Approximate Membership
4:12-4:18pm Brief Announcement: DeepZoning: Accelerate CNN Inference with Zoning Graph at Dynamic Granularity
4:18-4:24pm Brief Announcement: Optimized GPU-accelerated Feature Extraction for ORB-SLAM Systems
4:24-4:30pm Brief Announcement: is the problem-based benchmark suite fearless with Rust?
4:30-4:36pm Brief Announcement: Dynamic Vector Bin Packing for Online Resource Allocation in the Cloud
4:36-4:42pm Brief Announcement: Streaming Balanced Clustering

Monday, June 19

Session 7: Parallel Algorithms

8:45-9:06am A Simple and Efficient Parallel Laplacian Solver
9:06-9:27am Parallel Longest Increasing Subsequence and van Emde Boas Trees
9:27-9:48am High-Performance and Flexible Parallel Algorithms for Semisort and Related Problems
9:48-10:09am Optimal Parallel Sorting with Comparison Errors
10:09-10:30am Quadratic Speedups in Parallel Sampling from Determinantal Distributions

10:30-11am Break

11am-12:30pm FCRC Plenary Talk

12:30-2pm Lunch

Session 8: Linear Algebra / Graph Partitioning
2:00-2:21pm Multiplying 2 x 2 Sub-Blocks Using 4 Multiplications
2:21-2:42pm Parallel Memory-Independent Communication Bounds for SYRK
2:42-3:03pm Polylog-Competitive Algorithms for Dynamic Balanced Graph Partitioning for Ring Demands
3:03-3:24pm Partitioning Hypergraphs is Hard: Models, Inapproximability, and Applications
3:24-3:30pm Brief Announcement: Communication Optimal Sparse LU factorization for Planar Matrices

3:30-4pm Break

Session 9: Distributed Algorithms

4:00-4:21pm Adaptive Massively Parallel Connectivity in Optimal Space
4:21-4:42pm Fast dynamic programming in trees in the MPC model
4:42-5:03pm Coloring Fast with Broadcasts
5:03-5:24pm Optimal Round and Sample-Size Complexity for Partitioning in Parallel Sorting
5:24-5:45pm Distributed Multi-writer Multi-reader Atomic Register with Optimistically Fast Read and Write
5:45-5:51pm Brief Announcement: List Defective Colorings: Distributed Algorithms and Applications