# Any fast database working with Julia?
**URL:** https://discourse.julialang.org/t/any-fast-database-working-with-julia/17960
**Category:** Data
**Created:** [November 24, 2018, 11:34pm UTC](https://discourse.julialang.org/t/any-fast-database-working-with-julia/17960 "2018-11-24T23:34:58Z")
**Posts on this page:** 4
**Page:** 1
### Author:  [@Juan](https://discourse.julialang.org/u/Juan)
#### Post date: [November 24, 2018, 11:34pm UTC](https://discourse.julialang.org/t/any-fast-database-working-with-julia/17960/1 "2018-11-24T23:34:58Z")
What fast on-disk database (such as SciDB, MonetDB…) would you suggest to use to work with Julia.
I mean something Julia can easily connect to and able to deal with large datasets.
Being able to work distributed and on Windows systems is a plus.
Or do you just prefer to use feather or HDF5 to store your data?
---
### Author:  [@Yifan\_Liu](https://discourse.julialang.org/u/Yifan_Liu)
#### Post date: [November 24, 2018, 11:45pm UTC](https://discourse.julialang.org/t/any-fast-database-working-with-julia/17960/2 "2018-11-24T23:45:34Z")
You can use RCall to use the R MonetDBLite package so you don’t need to install MonetDB. It is fast not because it is on-disk but because it is column based. These lazy evaluation stuffs including feather are not faster than basic data frames. Personally I think HDF5 is a better choice.
---
### Author:  [@Juan](https://discourse.julialang.org/u/Juan)
#### Post date: [November 25, 2018, 12:01am UTC](https://discourse.julialang.org/t/any-fast-database-working-with-julia/17960/3 "2018-11-25T00:01:07Z")
Thanks, though I’d prefer something not relying on R.
I’ve also found some projects dealing with some of these databases but they haven’t been updated for 4 years.
---
### Author:  [@dfdx](https://discourse.julialang.org/u/dfdx)
#### Post date: [November 25, 2018, 12:43am UTC](https://discourse.julialang.org/t/any-fast-database-working-with-julia/17960/4 "2018-11-25T00:43:44Z")
If you are looking for a columnar database, consider [ClickHouse](https://clickhouse.yandex/) which supports simple HTTP API.