Haskell Spatial Effective Embedded Solution

Question

I'm looking for a save solution (maybe NoSQL db? Or something else ...) that has the following criteria:

1) Has a Haskell API

2) Effective disk space - db can easily get into many gigabytes of data, but I need it to work well on a regular desktop. I need something that stores data as efficiently as possible. So, for example, storing field names in a record would be bad.

3) High performance for reading sequential records. A typical use case starts somewhere and then reads ahead directly through the data - reading perhaps millions of records as quickly as possible.

4) In principle, the data never changes (they would be changed only if the data were found incorrectly), just registered

5) It must act directly on files (files) that can be easily moved / copied. It should not call a separate running server.

Answer 1

If you delete the “single file” requirement without any other running process, everything else can be performed by all standard DBMSs and depending on the data type, it is especially good in column stores in particular.

, , - sqlite. sqlite, . , , .

, , berkeleydb, , , .

Haskell sqlite berkeleydb.

: , - , . . . , . - - , , .

Answer 2

CouchDB :

1) http://hackage.haskell.org/package/CouchDB

2) , . , , , . XML JSON, , ( ).

3) , -.

4) CouchDB CM- , . , , .

5) . ( ) . ?