Storage Class Memories

Description

Magnetic RAM (MRAM) is one of several memory technologies that promises speeds and densities comparable to DRAM with the non-volatility of other magnetic media. Our research investigates the use of such technology to build faster, more reliable file systems. We are also exploring new file system access paradigms, such as the integration of file systems with databases, that can be implemented efficiently using the non-volatile byte-addressable storage provided by MRAM.

Status

After a developing a prototype for the linking file system, using a database to simulate the required data structures, we are now trying to write a version of LiFS that may reside in MRAM. This version will use a large amount of DRAM to simulate MRAM. We are developing the file system implementation in user space using FUSE to make it available through the VFS. We are also modifying a linux kernel to provide new system calls for the advanced features of our file system using links and attributes.

Useful Links

Here are some links relevant to storage class memories.

• Samsung announcement about phase-change RAM, a replacement for flash

Publications

• Sasha Ames, LiFSBrowse: A Visual, User Environment for the Linking File System, Technical Report UCSC-SSRC-07-08, August 2007.
• Kevin Greenan, Ethan L. Miller, PRIMS : Making NVRAM Suitable for Extremely Reliable Storage, short paper in Proceedings of the 3rd Workshop on Hot Topics in System Dependability (HotDep '07), June 2007.
• Timothy Bisson, Scott A. Brandt, Flushing Policies for NVCache Enabled Hard Disks, Technical Report UCSC-SSRC-07-04, May 2007.
• Timothy Bisson, Scott A. Brandt, Reducing Hybrid Disk Write Latency with Flash-Backed I/O Requests, Technical Report UCSC-SSRC-07-03, May 2007.
• Timothy Bisson, Scott A. Brandt, Darrell D. E. Long, A Hybrid Disk-Aware Spin-Down Algorithm with I/O Subsystem Support, Proceedings of the International Performance Conference on Computers and Communication (IPCCC '07), April 2007. Won Best Paper Award.
• Carlos Maltzahn, Nikhil Bobb, Mark W. Storer, Damian Eads, Scott A. Brandt, Ethan L. Miller, Graffiti: A Framework for Testing Collaborative Distributed Metadata, Proceedings in Informatics 21, March 2007, pages 97–111.
• Mark W. Storer, Graffiti Server - Design and Implementation, Technical Report UCSC-SSRC-07-02, January 2007.
• Kevin Greenan, Ethan L. Miller, Reliability Mechanisms for File Systems Using Non-Volatile Memory as a Metadata Store, Proceedings of the 6th ACM & IEEE Conference on Embedded Software (EMSOFT '06), October 2006, pages 178-187.
• Sasha Ames, Nikhil Bobb, Kevin Greenan, Owen Hofmann, Mark W. Storer, Carlos Maltzahn, Ethan L. Miller, Scott A. Brandt, LiFS: An attribute-rich file system for storage class memories, Proceedings of the 23rd IEEE / 14th NASA Goddard Conference on Mass Storage Systems and Technologies, May 2006.
• Nikhil Bobb, Damian Eads, Mark W. Storer, Scott A. Brandt, Carlos Maltzahn, Ethan L. Miller, Graffiti: A framework for testing collaborative distributed metadata, Proceedings of the 7th International Workshop on Distributed Data and Structures (WDAS 2006), January 2006.
• Sasha Ames, Nikhil Bobb, Scott A. Brandt, Adam Hiatt, Carlos Maltzahn, Ethan L. Miller, Alisa Neeman, Deepa Tuteja, Richer file system metadata using links and attributes, Proceedings of the 22nd IEEE / 13th NASA Goddard Conference on Mass Storage Systems and Technologies, April 2005.
• Timothy Bisson, Scott A. Brandt, Reducing Energy Consumption with a Non-Volatile Storage Cache, International Workshop on Software Support for Portable Storage (IWSSPS) held in conjunction with the IEEE Real-Time and Embedded Systems and Applications Symposium (RTAS 2005), March 2005.
• Nate Edel, Deepa Tuteja, Ethan L. Miller, Scott A. Brandt, MRAMFS: a compressing file system for non-volatile RAM, Proceedings of the 12th International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems (MASCOTS '04), October 2004, pages 596-603.
• Nate Edel, Karl Brandt, Ethan L. Miller, Scott A. Brandt, Measuring the compressibility of metadata and small files for disk/nvram hybrid storage systems., Proceedings of the 2004 International Symposium on Performance Evaluation of Computer and Telecommunication Systems (SPECTS '04), July 2004.
• Timothy Bisson, Scott A. Brandt, Adaptive Disk Spin-down Algorithms in Practice, Work-in-Progress Proceedings of the Usenix Conference of File and Storage Technology (FAST 2004), March 2004.
• Nate Edel, Karl Brandt, Ethan L. Miller, Scott A. Brandt, Measuring the compressibility of metadata and small files for disk/NVRAM hybrid storage systems, Technical Report UCSC-CRL-03-04, Storage Systems Research Center, University of California, July 2003.
• Alicja Szczurowska, MRAM–preliminary analysis for file system design, Master's thesis, University of California, Santa Cruz, March 2002, March 2002.
• Ethan L. Miller, Scott A. Brandt, Darrell D. E. Long, HeRMES: High-Performance Reliable MRAM-Enabled Storage, Proceedings of the 8th IEEE Workshop on Hot Topics in Operating Systems (HotOS-VIII), May 2001, pages 83-87.