What's interesting about this post is that the regular heap access is fast. But the sub-class implementation is hit rather badly, which is rather more about th

What's interesting about this post is that the regular heap access is fast. But the sub-class implementation is hit rather badly, which is rather more about the state of the JIT than anything else. HeapMemorySegment (standalone) : 1,441 msecs OffHeapMemorySegment (standalone) : 1,628 msecs HeapMemorySegment (subclass) : 3,841 msecs OffHeapMemorySegment (subclass) : 3,847 msecs