OracleConnections

Oracle Community Network

Lightweight database-like library for storing key-value pairs

What is the best way of storing key-value pairs of Strings in a file in Java, that is scalable (can work with a large number of pairs, i.e. doesn't read or write entire file on access), but is as lightweight as possible?

I am asking this because even the lightest database libraries, like SQLite and H2 seem like an overkill for this purpose, and are even impossible to use for ME programs (although I would need this mainly for SE programs for now).

2 members like this

▶ Reply to This

Replies to This Discussion

Permalink Reply by rihanna Adams on August 19, 2013 at 12:51pm

Oracle BerkeleyDB java edition allows you to store key-value objects, it is simple to use and administer, and up-scalable to heaven (or so). At 820k is not that big.

But if you are thinking about down scaling to j2me, you may try TinySQL

Pros:

It is small (93k!)
It is embeddable
It uses DBF or text files files to store data, so they are easy to read.

Cons:

It is an old unmaintained project
It is not designed to work in j2me, but since it can work in JDK 1.1.8 it won't be hard to make it work in j2me. Of course you will have to change some code from using RandomAccessFile to FileConnection and stuff like that, but at least you wont need to mess with generics related code.
It is not very fast, because it does not use indexes, so you need to try and see if it is fits your needs
It is not feature complete, just gives you a small subset of SQL

▶ Reply

Permalink Reply by Martha Cathy on August 19, 2013 at 12:53pm

For a static set of key-value pairs, Dan Bernstein's cdb comes to mind. To quote from the cdb description:

cdb is a fast, reliable, simple package for creating and reading constant databases. Its database structure provides several features:

Fast lookups: A successful lookup in a large database normally takes just two disk accesses. An unsuccessful lookup takes only one.

Low overhead: A database uses 2048 bytes, plus 24 bytes per record, plus the space for keys and data.

No random limits: cdb can handle any database up to 4 gigabytes. There are no other restrictions; records don't even have to fit into memory. Databases are stored in a machine-independent format.

Fast atomic database replacement: cdbmake can rewrite an entire database two orders of magnitude faster than other hashing packages.

Fast database dumps: cdbdump prints the contents of a database in cdbmake-compatible format.

cdb is designed to be used in mission-critical applications like e-mail. Database replacement is safe against system crashes. Readers don't have to pause during a rewrite.