New Record in DDoS, Managing 450M+ Images and Calling Uber
🛡 5.6 Tbps: a new DDoS record blocked by Cloudflare
What happened – In January 2025 Cloudflare mitigated the now largest distributed‑denial‑of‑service, peaking at 5.6 Tbps.
Why it matters – The post dissects shifting attacker tactics and offers mitigation tips.
🖼️ How Agoda wrangles 450 million property images
What happened – Agoda reveals the architecture behind its Content Enrichment Platform that ingests, deduplicates and serves almost half a billion images.
Why it matters – Clever caching layers and isolated “micro‑transformers” kept S3 costs flat while traffic doubled – patterns you can steal for any large media repo.
🏗️ Canva’s Snowflake pipeline in production
What happened – Canva explains how it built a high‑throughput ingestion path with Snowpipe Streaming, handling billions of analytics events daily.
Why it matters – Their cost comparison vs. Kafka/BigQuery is gold if you’re debating warehouse‑native streaming.
⚙️ Airflow + Composer at lastminute.com
What happened – lastminute.com migrated 1000+ ETL DAGs to Apache Airflow on Google Cloud Composer, sharing the business case and the bill.
Why it matters – Rare to see hard numbers: they shaved 40% off ops overhead by auto‑suspending idle workers.
💾 Pinterest drops memory footprint on its core API
What happened – Switching from RocksDB to LMDB and tweaking pool allocation cut RSS by 52% in Pinterest’s GraphQL edge layer.
Why it matters – A reminder that “just buy more RAM” is not a strategy when you’re at a Pinterest scale
🚗 Uber bets on Ray to optimize ride matching
What happened – Uber replaced Spark jobs with a Ray cluster that powers surge‑pricing simulations and market‑balancing heuristics.
Why it matters – Nice, candid comparison table: Ray cut iteration time from 30 min to 3 min for AB experiments.
📈 Lyft models supply & demand with marketplace marginal values
What happened – Lyft shares the math (G‑formula, causal inference) behind its interference‑aware pricing model.
Why it matters – Probably the clearest real‑world example of causal graph techniques in a live marketplace.
Got a link that belongs here, or any feedback? Reach out to me on LinkedIn, and I’ll check it out. Until next time – stay scalable! ✌️