Cloud databases changed the economics of storage, but the B+ tree is still the default answer for ordered data on disk.
Why B+ Trees Still Matter in the Age of Cloud Databases
Cloud databases changed the economics of storage, but the B+ tree is still the default answer for ordered data on disk.
Cloud storage made durability and capacity feel infinite, but it did not remove the cost of random reads. The B+ tree remains a strong answer because it keeps ordered data in large, sequential pages.
Fanout, page size and write amplification still decide how a storage engine behaves under load. Managed services hide the operations, not the physics.
The cloud made the disk someone else's problem. The access pattern is still yours.
Cache Eviction Policies in Storage Engines
Caching stores frequently accessed items in fast volatile storage. The interesting part is deciding what to evict, and when.
Read articleGenerating Zipfian Distributions Using Rejection Inversion
Rejection inversion is a sampling method for discrete monotone distributions. Here is how to use it to benchmark and test software.
Read articleUnderstanding Rejection–Inversion Sampling for Zipf Distributions
The theory behind sampling from Zipfian distributions, and why the inverse CDF is not always enough.
Read article