The Low Level Persistence Library offers access to persistent memory using MemoryBlocks allocated from a persistent Heap.
This Java library uses the libpmemobj library from the Persistent Memory Development Kit (PMDK). For more information on PMDK, please visit http://pmem.io and https://github.com/pmem/pmdk.
The following are the prerequisites for building this Java library:
- Linux operating system (tested on CentOS 7.2)
- Persistent Memory Development Kit (PMDK)
- Java 8 or above
- Build tools - g++ compiler and make
This library assumes the availability of hardware persistent memory or emulated persistent memory. Instructions for creating emulated persistent memory are shown below.
The preferred way is to create an in-memory DAX file system. This requires Linux kernel 4.2 or greater. Please follow the steps at:
http://pmem.io/2016/02/22/pm-emulation.html
Alternatively, for use with older kernels, create a tmpfs partition as follows (as root):
$ mount -t tmpfs -o size=4G tmpfs /mnt/mem # creates a 4GB tmpfs partition
$ chmod -R a+rw /mnt/mem # enables read/write permissions to all users
Once all the prerequisites have been satisfied:
$ git clone https://github.com/pmem/llpl.git
$ cd llpl
$ make && make tests
Available Makefile targets include:
sources
- builds only sourcesexamples
- builds the sources and examplestests
- builds and runs testsdocs
- builds javadocsjar
- builds jar file intotarget
directory
To import this library into an existing Java application, include the project's target/classes
directory in your Java classpath and the project's target/cppbuild
directory in your
java.library.path
. For example:
$ javac -cp .:<path>/llpl/target/classes <source>
$ java -cp .:<path>/llpl/target/classes -Djava.library.path=<path>/llpl/target/cppbuild <class>
Thanks for your interest! Right now, architectural changes are still happening in the project. This makes it difficult to contribute code and difficult to effectively process pull requests. We expect these changes to settle out around March of this year and we look forward to code contributions once this happens. We will update this README then.
In the meantime, we would love to hear your comments and suggestions via https://github.com/pmem/llpl/issues.
For more information on this library, contact Olasoji Denloye ([email protected]) or Steve Dohrmann ([email protected]).