nvm malloc: Memory Allocation for NVRAM

Non-volatile main memory (NVRAM) has the potential to fundamentally change the persistency of software. Applications can make their state persistent by directly placing data structures on NVRAM instead of volatile DRAM. However, the persistent nature of NVRAM requires significant changes for memory allocators that are now faced with the additional tasks of data recovery and failure-atomicity. In this paper, we present nvm malloc, a general-purpose memory allocator concept for the NVRAM era as a basic building block for persistent applications. We introduce concepts for managing named allocations for simplified recovery and using volatile and non-volatile memory in combination to provide both high performance and failure-atomic allocations.

[1]  Rajesh K. Gupta,et al.  NV-Heaps: making persistent objects fast and safe with next-generation, non-volatile memories , 2011, ASPLOS XVI.

[2]  D. Andersen,et al.  Persistent , Protected and Cached : Building Blocks for Main Memory Data Stores , 2011 .

[3]  Kathryn S. McKinley,et al.  Hoard: a scalable memory allocator for multithreaded applications , 2000, SIGP.

[4]  Ethan L. Miller,et al.  PRIMS: making NVRAM suitable for extremely reliable storage , 2007 .

[5]  Roy H. Campbell,et al.  Consistent and Durable Data Structures for Non-Volatile Byte-Addressable Memory , 2011, FAST.

[6]  A. L. Narasimha Reddy,et al.  SCMFS: A file system for Storage Class Memory , 2011, 2011 International Conference for High Performance Computing, Networking, Storage and Analysis (SC).

[7]  Ada Gavrilovska,et al.  Reducing the cost of persistence for nonvolatile heaps in end user devices , 2014, 2014 IEEE 20th International Symposium on High Performance Computer Architecture (HPCA).

[8]  Jason Evans April.  A Scalable Concurrent malloc(3) Implementation for FreeBSD , 2006 .

[9]  Christopher Frost,et al.  Better I/O through byte-addressable, persistent memory , 2009, SOSP '09.

[10]  Michael M. Swift,et al.  Mnemosyne: lightweight persistent memory , 2011, ASPLOS XVI.

[11]  Chao Wang,et al.  NVMalloc: Exposing an Aggregate SSD Store as a Memory Partition in Extreme-Scale Machines , 2012, 2012 IEEE 26th International Parallel and Distributed Processing Symposium.

[12]  Hovav Shacham,et al.  On the effectiveness of address-space randomization , 2004, CCS '04.

[13]  Sanjay Kumar,et al.  System software for persistent memory , 2014, EuroSys '14.

[14]  Orion Hodson,et al.  Whole-system persistence , 2012, ASPLOS XVII.

[15]  Guillaume Mercier,et al.  hwloc: A Generic Framework for Managing Hardware Affinities in HPC Applications , 2010, 2010 18th Euromicro Conference on Parallel, Distributed and Network-based Processing.

[16]  Chuck Lever,et al.  Malloc() Performance in a Multithreaded Linux Environment , 2000, USENIX Annual Technical Conference, FREENIX Track.

[17]  M.H. Kryder,et al.  After Hard Drives—What Comes Next? , 2009, IEEE Transactions on Magnetics.

[18]  Paul R. Wilson,et al.  Dynamic Storage Allocation: A Survey and Critical Review , 1995, IWMM.

[19]  Sooyong Kang,et al.  Performance Trade-Offs in Using NVRAM Write Buffer for Flash Memory-Based Storage Devices , 2009, IEEE Transactions on Computers.

[20]  Paul R. Wilson,et al.  The memory fragmentation problem: solved? , 1998, ISMM '98.