Time: 15:30 - 16:20
Room: Houston Ballroom B/C
Postgres has for a long time been lauded to handle concurrency well. I.e. to process more requests if addtional resources (CPUs) are added. Unfortunately after that had been determined many years ago, not much work had been put into the topic. But since then the number of cores and the general architecture of parallel systems have changed. Leading to the stiuation that we don't scale as well as we'd like.
I'll talk about recent scalability improvements (9.2, 9.5) that improved things considerably and what the biggest remaining scalability problems are. Some of the remaining issues can mainly addressed by changing postgres, others can be worked around on the application side.