Creative Insights

Exploring the intersection of code, design, and technology.

Designing an io_uring-Driven LSM-Tree for Zoned Namespaces (ZNS) SSDs tech
Oct 25, 2026 2 min read

Designing an io_uring-Driven LSM-Tree for Zoned Namespaces (ZNS) SSDs

Learn how combining Linux io_uring with Zoned Namespaces (ZNS) SSDs eliminates double garbage collection and kernel overhead in high-performance log-structured storage engines.

Read Article
Mitigating Cache Coherency Hazards in Zero-Copy DMA Ring Buffers on ARM Cortex-M7 tech
Oct 22, 2026 3 min read

Mitigating Cache Coherency Hazards in Zero-Copy DMA Ring Buffers on ARM Cortex-M7

Learn how to eliminate cache coherency bugs and false sharing in Cortex-M7 zero-copy DMA ring buffers while maximizing CMSIS-DSP throughput.

Read Article
Taming Raft Tail Latency: Building an eBPF-Driven Adaptive Heartbeat Controller for Asymmetric Partitions tech
Oct 21, 2026 2 min read

Taming Raft Tail Latency: Building an eBPF-Driven Adaptive Heartbeat Controller for Asymmetric Partitions

Learn how to eliminate tail latency spikes in Raft clusters during asymmetric network partitions by tracking kernel-level socket metrics with an adaptive eBPF heartbeat …

Read Article
Benchmarking io_uring vs. Linux AIO for LSM-Tree Commit Logs: Mitigating Tail Latency Under NVMe Write Saturation tech
Oct 19, 2026 2 min read

Benchmarking io_uring vs. Linux AIO for LSM-Tree Commit Logs: Mitigating Tail Latency Under NVMe Write Saturation

Discover how io_uring slashes p99 tail latencies compared to Linux AIO during high-throughput LSM-tree write-ahead log ingestion on saturated NVMe storage.

Read Article
tech
Oct 18, 2026 2 min read

Deterministic Durability: Slashing P99 Latency in LSM-Trees with ZNS and io_uring

Learn how replacing legacy block I/O with NVMe Zoned Namespaces and io_uring eliminates Write-Ahead Log stalls and crushes P99 tail latency in distributed storage …

Read Article
Real-Time Acoustic Anomaly Detection on Cortex-M4: Fixed-Point STFT and INT8 TFLM in Under 64KB SRAM tech
Oct 17, 2026 2 min read

Real-Time Acoustic Anomaly Detection on Cortex-M4: Fixed-Point STFT and INT8 TFLM in Under 64KB SRAM

Learn how to build an end-to-end bare-metal acoustic anomaly detector on an ARM Cortex-M4 using fixed-point CMSIS-DSP STFT and INT8 TFLite Micro within a …

Read Article
Building an In-Process Semantic Search Pipeline in Go with ONNX, pgvector, and RRF tech
Oct 11, 2026 2 min read

Building an In-Process Semantic Search Pipeline in Go with ONNX, pgvector, and RRF

Learn how to build a low-latency semantic search pipeline in Go using embedded ONNX inference, pgvector HNSW indexing, and Reciprocal Rank Fusion.

Read Article
Scalable Hybrid Search in Go: Merging pgvector HNSW and Full-Text Search with RRF Under High Ingestion tech
Oct 11, 2026 2 min read

Scalable Hybrid Search in Go: Merging pgvector HNSW and Full-Text Search with RRF Under High Ingestion

Learn how to build a resilient hybrid search pipeline in Go, combining PostgreSQL pgvector HNSW and lexical search via Reciprocal Rank Fusion under heavy …

Read Article
Architecting Resilient Hybrid Search in Go: pgvector, BM25, and In-Process ONNX Reranking tech
Oct 11, 2026 2 min read

Architecting Resilient Hybrid Search in Go: pgvector, BM25, and In-Process ONNX Reranking

Learn how to build a high-throughput hybrid search engine in Go, combining PostgreSQL semantic retrieval, BM25 lexical matching, and embedded ONNX cross-encoder reranking.

Read Article