Byte addressable non-volatile memory (NVRAM) is likely to supplement, and perhaps eventually replace, DRAM. Applications can then persist data structures directly in memory instead of serializing them and storing them onto a durable block device. However, failures during execution can leave data structures in NVRAM unreachable or corrupt. In this paper, we present Makalu, a system that addresses non-volatile memory management. Makalu offers an integrated allocator and recovery-time garbage collector that maintains internal consistency, avoids NVRAM memory leaks, and is efficient, all in the face of failures.
We show that a careful allocator design can support a less restrictive and a much more familiar programming model than existing persistent memory allocators. Our allocator significantly reduces the per allocation persistence overhead by lazily persisting non-essential metadata and by employing a post-failure recovery-time garbage collector. Experimental results show that the resulting online speed and scalability of our allocator are comparable to well-known transient allocators, and significantly better than state-of-the-art persistent allocators.
Thu 3 NovDisplayed time zone: Amsterdam, Berlin, Bern, Rome, Stockholm, Vienna change
13:30 - 15:10
|Efficient and Thread-Safe Objects for Dynamically-Typed Languages|
Benoit Daloze JKU Linz, Austria, Stefan Marr Johannes Kepler University Linz, Daniele Bonetta Oracle Labs, Hanspeter Mössenböck JKU Linz, AustriaDOI Pre-print Media Attached
|Hybrid STM/HTM for Nested Transactions on OpenJDK|
Keith Chapman Purdue University, Tony Hosking Australian National University, Data61, and Purdue University, Eliot Moss University of Massachusetts AmherstLink to publication DOI Pre-print Media Attached
|Makalu: Fast Recoverable Allocation of Non-volatile Memory|
OOPSLADOI Pre-print Media Attached
|Prioritized Garbage Collection: Explicit GC Support for Software Caches|
Diogenes Nunez Tufts University, Sam Guyer Tufts University, Emery D. Berger University of Massachusetts, AmherstDOI Pre-print Media Attached