This is an implementation of a memory manager based on buddy memory algorithm. https://en.wikipedia.org/wiki/Buddy_memory_allocation
Modified buddy algorithm is used in Linux Kernel and few commercial products. It is not an advanced collector and requires manual deallocation.
https://www.kernel.org/doc/gorman/html/understand/understand009.html
Package provides two implementations; an on-heap algorithm backed by byte[] and off JVM heap.
Deallocation is explicit. Memory defragmentation is invoked manually. There is no daemon thread to combine freed memory blocks.
User who forgets to deallocate will cause a memory leak.
//
logger.info("Starting test");
// use either of two algorithms
MemoryClientInterface impl_offheap=MemoryMgrFactory.getImplementation(AlgoImplEnum.OFF_HEAP);
//MemoryClientInterface impl_onheap=MemoryMgrFactory.getImplementation(AlgoImplEnum.ON_HEAP);
// set up 64MB with 1K page
impl_offheap.setup(64*1024*1024,1024);
// another notation: impl_offheap.setup("64 kb", "1 kb");
// allocate 10K bytes
long address=impl_offheap.allocate(10000);
printSeparator();
// write integer values into address space [0...10000-1]
int myIntWriting1=7;
int myIntWriting2=-500;
impl_offheap.writeIntToByteArray( myIntWriting1, address, 0 );
impl_offheap.writeIntToByteArray( myIntWriting2, address, 4 ); // for ints step by 4.
// read back
int myIntRead1=impl_offheap.readIntFromByteArray( address, 0 );
int myIntRead2=impl_offheap.readIntFromByteArray( address, 4 );
printSeparator();
// when done release memory back to the pool
impl_offheap.deallocate(address);
printSeparator();
// show memory usage
impl_offheap.print();
Project is born to support
- efficient resizing of arrays of primitives
- exceed heap limitation of 2^31-1 array elements.
- improve performance by moving allocation off heap and avoiding jvm array checks.
-
Concurrent access to memory manager is not supported. Results will be unpredictable. Solution: synchronize access to memory mutator methods.
-
Block barriers. If user allocated 1000 bytes and decided to write 1010 bytes, next memory block(s) will be corrupted. Solution: Proxy memory I/O access and track block boundaries.
-
De-fragmentation is not automatic like GC in Java. Solution: Create a daemon thread, run concurrently with the allocator creating contiguous free space.
-
Each allocation request stores a header. Due to this overhead, available space is always less 100% of setup memory. Solution: none.
Execute mvn command then go into /bin
mvn clean compile assembly:single
See folder /docs
See package /examples
Single contributor.
As a single author, I would appreciate user feedback and suggestions. Please reply to github email on file.
The MIT License (MIT)
Copyright (c) 2015-2016 ocean927
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.