Fault-Tolerant File-I/O for Portable Checkpointing Systems

Authors:
Igor Lyubashevskiy;Volker Strumpen
Affiliations:
Oracle Corp. 781-768-5600 Facilities, 6th Floor 200 Fifth Avenue Waltham, MA 02451-8779 ilyubash@us.oracle.com;Yale University, Departments of Electrical Engineering and Computer Science strumpen@ee.yale.edu
Venue:
The Journal of Supercomputing - Special issue on embedded fault-tolerance systems
Year:
2000

Citing 9
Cited 2

Machine-Independent Virtual Memory Management for Paged Uniprocessor and Multiprocessor Architectures

IEEE Transactions on Computers - Special issue on architectural support for programming languages and operating systems
The standard C library

The standard C library
Modern operating systems

Modern operating systems
A new approach to I/O performance evaluation: self-scaling I/O benchmarks, predicted I/O performance

ACM Transactions on Computer Systems (TOCS) - Special issue on computer architecture
Transaction Processing: Concepts and Techniques

Transaction Processing: Concepts and Techniques
Portable Checkpointing for Heterogeneous Archtitectures

FTCS '97 Proceedings of the 27th International Symposium on Fault-Tolerant Computing (FTCS '97)
Integrating Checkpointing with Transaction Processing

FTCS '97 Proceedings of the 27th International Symposium on Fault-Tolerant Computing (FTCS '97)
The possibilities and limitations of heterogeneous process migration

The possibilities and limitations of heterogeneous process migration
Libckpt: transparent checkpointing under Unix

TCON'95 Proceedings of the USENIX 1995 Technical Conference Proceedings

CprFS: a user-level file system to support consistent file states for checkpoint and restart

Proceedings of the 22nd annual international conference on Supercomputing
HSG-LM: hybrid-copy speculative guest OS live migration without hypervisor

Proceedings of the 6th International Systems and Storage Conference

Quantified Score

Hi-index	0.00

Visualization

Abstract

The ftIO-system provides portable and fault-tolerant file-I/O by enhancing the functionality of the ANSI C file system without changing its application programmer interface and without depending on system-specific implementations of the standard file operations. The ftIO-system is an extension of the porch compiler and its runtime system. The porch compiler automatically generates code to save bookkeeping information about ftIO's transactional file operations in portable checkpoints. These portable checkpoints can be recovered on a binary incompatible architecture. We developed a new algorithm for supporting transactional file operations in ftIO. Rather than using the well-known two-phase commit protocol, this algorithm uses only a single bit of information and an atomic rename file operation to guarantee fault tolerance. In this paper, we describe our new ftIO algorithm, discuss design choices for ftIO, and provide experimental data of our ftIO prototype.