The Curious Engineer

All blogs / Systems

Dangling Pointers

danglingpointers.substack.com

Summaries of computer science research papers, with a focus on unsolved problems. Click to read Dangling Pointers, by Blake Pelton, a Substack publication with hundreds of subscribers.

SystemsDatabasesCPU / microarchitectureNetworking

’25’26
Posts
105 from 4 Aug 2025 to 27 Oct 2026
Posting
weekly (67 in the last year)
Kind
personal blog
Feed
https://danglingpointers.substack.com/feed

2026 53

  1. TwinString: Preserving String Semantics with Off-Heap Data on the JVM
  2. Warnock: Harnessing GPU Geometry Amplification for Vector Graphics
  3. Honey, I Shrunk the Headers With Flow.ZIP
  4. Presto: A Match-Action TCP Stack for the Terabit Era
  5. Understanding Host Network Stack Latency
  6. Analysis of Area and Power Overhead for a Synthesizable 4-Phase Bundled-Data Asynchronous RISC-V Processor
  7. Vector Value Prediction with Element-wise Stride Compression
  8. Modular GPU Programming with Typed Perspectives
  9. RNG: Flat Datacenter Networks at Scale
  10. Bonsai: Compiling Queries to Pruned Tree Traversals
  11. MatchBox: A Semantic Foundation for Data Plane Portability
  12. Breadcrumb Filters: Fast Fully Featured Filters
  13. RABIT: Efficient Range Queries with Bitmap Indexing
  14. Accelerating Stream Processing Engines via Hardware Offloading
  15. F3: The Open-Source Data File Format for the Future
  16. Enabling Packet Spraying over Commodity RNICs with In-Network Support
  17. LightDSA: Enabling Efficient DSA Through Hardware-Aware Transparent Optimization
  18. Accelerating Transactional Execution via Processing-In-Memory
  19. Yield Not Thy Core
  20. Pipeline Parallel Decompression
  21. SG-IOV: Socket-Granular I/O Virtualization for SmartNIC-Based Container Networks
  22. Efficient Remote Memory Ordering for Non-Coherent Systems
  23. Performance Predictability in Heterogeneous Memory
  24. Enabling Fast Networking in the Public Cloud
  25. Cicada: Dependably Fast Multi-Core In-Memory Transactions
  26. Beyond Page Migration: Enhancing Tiered Memory Performance via Integrated Last-Level Cache Management and Page Migration
  27. MagiCache: A Virtual In-Cache Computing Engine
  28. FlexGuard: Fast Mutual Exclusion Independent of Subscription
  29. RTSpMSpM: Harnessing Ray Tracing for Efficient Sparse Matrix Computations
  30. Dissecting and Modeling the Architecture of Modern GPU Cores
  31. Nexus Machine: An Energy-Efficient Active Message Inspired Reconfigurable Architecture
  32. TiNA: Tiered Network Buffer Architecture for Fast Networking in Chiplet-based CPUs
  33. Binary Compatible Critical Section Delegation
  34. Radshield: Software Radiation Protection for Commodity Hardware in Space
  35. Cacheman: A Comprehensive Last-Level Cache Management System for Multi-tenant Clouds
  36. Hapax Locks: Scalable Value-Based Mutual Exclusion
  37. Scalar Interpolation: A Better Balance between Vector and Scalar Execution for SuperScalar Architectures
  38. Flexible I/O for Database Management Systems with xNVMe
  39. A 1.27 fJ/B/transition Digital Compute-in-Memory Architecture for Non-Deterministic Finite Automata Evaluation
  40. Better Memory Tiering, Right from the First Placement
  41. Contiguitas: The Pursuit of Physical Memory Contiguity in Datacenters
  42. Efficient Lossless Compression of Scientific Floating-Point Data on CPUs and GPUs
  43. EDM: An Ultra-Low Latency Ethernet Fabric for Memory Disaggregation
  44. An Analysis of User-space Idle State Instructions on x86 Processors
  45. Re-architecting End-host Networking with CXL: Coherence, Memory, and Offloading
  46. Forest: Access-aware GPU UVM Management
  47. UPP: Universal Predicate Pushdown to Smart Storage
  48. RayN: Ray Tracing Acceleration with Near-memory Computing
  49. SHADOW: Simultaneous Multi-Threading Architecture with Asymmetric Threads
  50. Light-weight Cache Replacement for Instruction Heavy Workloads
  51. LoopFrog: In-Core Hint-Based Loop Parallelization
  52. Dynamic Load Balancer in Intel Xeon Scalable Processor
  53. The XOR Cache: A Catalyst for Compression

2025 52

  1. CHERIoT RTOS: An OS for Fine-Grained Memory-Safe Compartments on Low-Cost Embedded Devices
  2. CHERIoT: Complete Memory Safety for Embedded Devices
  3. Analog in-memory computing attention mechanism for fast and energy-efficient large language models
  4. A linear-time heuristic for improving network partitions
  5. Rearchitecting the Thread Model of In-Memory Key-Value Stores with μTPS
  6. Morsel-driven parallelism: a NUMA-aware query evaluation framework for the many-core age
  7. The Decomposition of Switching Functions
  8. Oasis: Pooling PCIe Devices Over CXL to Boost Utilization
  9. Tai Chi: A General High-Efficiency Scheduling Framework for SmartNICs in Hyperscale Clouds
  10. Backdoors To Typical Case Complexity
  11. How to Copy Memory? Coordinated Asynchronous Copy as a First-Class OS Service
  12. DPU-KV: On the Benefits of DPU Offloading for In-Memory Key-Value Stores at the Edge
  13. A Computing Procedure for Quantification Theory
  14. High-Performance Query Processing with NVMe Arrays: Spilling without Killing Performance
  15. Low-Latency Transaction Scheduling via Userspace Interrupts: Why Wait or Yield When You Can Preempt?
  16. SuSe: Summary Selection for Regular Expression Subsequence Aggregation over Streams
  17. Fast and Scalable Data Transfer Across Data Systems
  18. Falcon: A Reliable, Low Latency Hardware Transport
  19. Bounding Speculative Execution of Atomic Regions to a Single Retry
  20. CEIO: A Cache-Efficient Network I/O Architecture for NIC-CPU Data Paths
  21. Automata All the Way Down
  22. Gigaflow: Pipeline-Aware Sub-Traversal Caching for Modern SmartNICs
  23. Optimizing Datalog for the GPU
  24. Extended User Interrupts (xUI): Fast and Flexible Notification without Polling
  25. No Cap, This Memory Slaps: Breaking Through the Memory Wall of Transactional Database Systems with Processing-in-Memory
  26. Parendi: Thousand-Way Parallel RTL Simulation
  27. Skia: Exposing Shadow Branches
  28. Principles and Methodologies for Serial Performance Optimization
  29. Accelerate Distributed Joins with Predicate Transfer
  30. To PRI or Not To PRI, That's the question
  31. Pushing the Limits of In-Network Caching for Key-Value Stores
  32. State-Compute Replication: Parallelizing High-Speed Stateful Packet Processing
  33. 3D Gaussian Splatting for Real-Time Radiance Field Rendering
  34. Scaling IP Lookup to Large Databases using the CRAM Lens
  35. Enabling Silent Telemetry Data Transmission with InvisiFlow
  36. Enabling Portable and High-Performance SmartNIC Programs with Alkali
  37. Ripple: Asynchronous Programming for Spatial Dataflow Architectures
  38. Iso: Request-Private Garbage Collection
  39. Making Concurrent Hardware Verification Sequential
  40. Instant neural graphics primitives with a multiresolution hash encoding
  41. Disentangling the Dual Role of NIC Receive Rings
  42. Efficiently Processing Joins and Grouped Aggregations on GPUs
  43. Partial Evaluation, Whole-Program Compilation
  44. Filtering After Shading With Stochastic Texture Filtering
  45. Boosting OLTP Performance with Per-Page Logging on NVDIMM
  46. Zombie Hashing: Reanimating Tombstones in Graveyard
  47. SPID-Join
  48. Don't Repeat Yourself! Coarse-Grained Circuit Deduplication to Accelerate RTL Simulation
  49. Necro-reaper: Pruning away Dead Memory Traffic in Warehouse-Scale Computers
  50. Predicate Transfer: Efficient Pre-Filtering on Multi-Join Queries
  51. Low-latency, high-throughput garbage collection
  52. Welcome To Dangling Pointers