Pydis: Redis clone in Python 3 to make points about performance

Re:Pydis: Redis clone in Python 3 to make points about perfo... by orbifold on Saturday March 02, 2019 @11:04AM (#19289725)

The conclusion should probably be that redis is super slow, not that the python 3 implementation is particularly fast. Also it is unclear how python dictionaries would scale for large number of keys compared to redis.

[ Parent | on HN ]

Re:Pydis: Redis clone in Python 3 to make points about perfo... by jnwatson on Saturday March 02, 2019 @01:06PM (#19290284)

The Python dictionary implementation is among the most optimized on the planet. Given that Python uses dictionaries for almost everything, very close attention is paid to dictionary performance.

[ Parent | on HN ]

Re:Pydis: Redis clone in Python 3 to make points about perfo... by benekastah on Saturday March 02, 2019 @02:02PM (#19290566)

That doesn't automatically mean that dictionaries are just as fast with hundreds of thousands or millions of keys. I don't know one way or the other, but it's conceivable that performance could degrade somewhat with large numbers of keys.

[ Parent | on HN ]

Re:Pydis: Redis clone in Python 3 to make points about perfo... by coldtea on Saturday March 02, 2019 @12:27PM (#19290093)

>The conclusion should probably be that redis is super slow, not that the python 3 implementation is particularly fast.

That sounds like the definition of confirmation bias.

How is that substantiated?

[ Parent | on HN ]