HTAP is Dead

This blog is inspired by Jordan Tigani’s blog titled “Big Data is Dead”. Jordan and I actually spent some time building a HTAP database at SingleStore. The good old days ('70s) Back in the ’70s, one relational database did everything. Transactions (OLTP) during the day and reports after hours (OLAP). Databases like Oracle V2 and IBM DB2 ran OLTP and OLAP on the same system; largely because data sets still fit on a few disks and compute was costly. Nobody called it Hybrid Transactional/Analytical Processing (HTAP); it was simply the database. The great divide (‘80s) As businesses had more data and asked tougher questions, the database began to show its limits. See, transactional and analytical workloads pull in opposite directions. OLTP requires microsecond inserts and single-row lookups, while OLAP demands full-table scans and large-scale aggregates. This created constant contention; analytics consuming I/O and cache needed for latency-sensitive transactions, and vice versa. The ... Read full article.