Re: [h2] database benchmark performace mysql vs cache vs system vs directory file

[Noel Grandin]

I'm sorry to say, but you need to do more reading on how caches work, that
is not at all how it functions.

A brief overview of our architecture is here:
http://h2database.com/html/architecture.html

When people say that SQLite is faster than the filesystem, they mean it is
faster for the specific use-case of storing and retrieving structured
records.
It can be faster because it knows more about the structure and
access-pattern of the data than the filesystem.

-- 
You received this message because you are subscribed to the Google Groups "H2 
Database" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to h2-database+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/h2-database/CAFYHVnXyeHom0xh%3DD4fUxCPQO7fCXmzH%2BOL-5T%2B7VgEt3qE58A%40mail.gmail.com.

Re: [h2] database benchmark performace mysql vs cache vs system vs directory file

[Danilo Santos]

as I understand it, this is copying and pasting a frequent part of the 
storage file into another smaller storage file.

but I was wondering if it would be faster to insert data directly into the 
directory system instead of storage files.

on the internet there are several theories, but none shows a benchmark, and 
I think that 1 creator of 1 database may have tested adversarial databases 
to analyze performance.

they say sqlite is 70% faster than the file system, but I can’t understand 
why, I still don’t have the means to test whether it’s true and I don’t 
even find test videos like they do on video cards even for ram, processor, 
ghz, core due to fps

Em domingo, 12 de abril de 2020 03:50:52 UTC-3, Noel Grandin escreveu:
>
> Not really, look up how an LRU cache works e.g.
>
> https://en.wikipedia.org/wiki/Cache_replacement_policies
>

-- 
You received this message because you are subscribed to the Google Groups "H2 
Database" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to h2-database+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/h2-database/160a8a36-1879-4823-8e06-3be0060846cf%40googlegroups.com.

Re: [h2] database benchmark performace mysql vs cache vs system vs directory file

[Noel Grandin]

Not really, look up how an LRU cache works e.g.

https://en.wikipedia.org/wiki/Cache_replacement_policies

-- 
You received this message because you are subscribed to the Google Groups "H2 
Database" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to h2-database+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/h2-database/CAFYHVnV3NRqE9HOWmzVFPPJAriyJg5MB%2BxgMauW%3DZnFmbyhvwQ%40mail.gmail.com.

Re: [h2] database benchmark performace mysql vs cache vs system vs directory file

[Danilo Santos]

has cache to analyze the frequency of the line and position it for a faster 
query in the future based on the probability that the lines appear?

-- 
You received this message because you are subscribed to the Google Groups "H2 
Database" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to h2-database+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/h2-database/75bbca0b-78ca-4a46-8985-147dc54efae2%40googlegroups.com.

Re: [h2] database benchmark performace mysql vs cache vs system vs directory file

[Noel Grandin]

On Sat, 11 Apr 2020 at 21:04, Danilo Santos 
wrote:

> exists one https://database.serverbenchmark.com/ like
> https://gpu.userbenchmark.com/ or https://cpu.userbenchmark.com/
> 
>
>
Not that I am aware of. Lots of databases prohibit the publication of
unauthorised benchmarks, so it is a little tricky.



> why are there still people using mysql instead of 1 portable database like
> h2database?
>
>
There are a lot of features in different databases, and different
tradeoffs. Your job as a developer is to find the database that has the
right tradeoffs for your application.

> how does the database cache system work? is it faster than the operating
> system's directory system? or that sqlite?
>
>
We have two cache systems, a fairly normal LRU cache and a scan-resistant
LRU cache. Our cache is more "efficient" than the operating systems cache
for the specific purpose of serving rows, but that is because our cache
operates at a different level.

-- 
You received this message because you are subscribed to the Google Groups "H2 
Database" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to h2-database+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/h2-database/CAFYHVnUYV0QKMEH23OggzDW6%3DE9TtMDXA05rUJ1cQKWPEHqsiw%40mail.gmail.com.

[h2] database benchmark performace mysql vs cache vs system vs directory file

[Danilo Santos]

exists one https://database.serverbenchmark.com/ like 
https://gpu.userbenchmark.com/ or https://cpu.userbenchmark.com/ 


why are there still people using mysql instead of 1 portable database like 
h2database?

how does the database cache system work? is it faster than the operating 
system's directory system? or that sqlite?


-- 
You received this message because you are subscribed to the Google Groups "H2 
Database" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to h2-database+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/h2-database/c7946e06-7812-4b02-a9fd-75d4020cea14%40googlegroups.com.