This Article 
 Bibliographic References 
 Add to: 
A Checkpointing Page Store for Write-Once Optical Disk
January 1990 (vol. 39 no. 1)
pp. 2-9

A model paging system is described for write-once optical disk. The paging system checkpoints pages on a nonerasable, failure-resistant medium as a side effect of virtual memory operation. Most pages are already checkpointed at any given time. It is practicable to checkpoint the global state periodically by forcing the write-out of pages that have been modified since their last checkpoint, together with the page tables. The proposed checkpointing algorithm was simulated to evaluate the performance of paging and checkpoint to write-once optical disk. There are plausible values for the operating parameters that make the overhead of checkpointing negligible compared to paging. The lifetime of 1-GB write-once optical disk media used for paging and checkpointing can range up to three months under plausible operating conditions.

[1] O. Babaoglu and W. Joy, "Converting a swap based system to do paging in an architecture lacking page reference bits," inProc. 8th Symp. Operating System Principles, Asilomar, CA, Dec. 1981, pp. 78-86.
[2] A. Bell, "Optical data storage--A status report," inProc. 6th IEEE Symp. Mass Storage Syst., 1984, pp. 93-98.
[3] L. Cabrera, "The influence of workload on load balancing strategies,"Usenix Proc., Atlanta, GA, June 1986, pp. 446-458.
[4] S. Christodoulakis and C. Faloutsos, "Design and Performance Considerations for an Optical Disk Based Multimedia Object Server,"Computer, Dec. 1986, pp. 45-56.
[5] G. Copeland, "What if mass storage were free?,"IEEE Comput. Mag., pp. 27-35, July 1982.
[6] M. C. Easton, "Key Sequence Data Sets on Indelible Storage,"IBM J. R&D, May 1986, pp. 230-241.
[7] B. Ellis, "A stable storage package,"Usenix Proc., Portland, OR, June 1985, pp. 209-212.
[8] J. Eppinger and A. Spector, "Virtual memory management for recoverable objects in the TABS prototype," CS-85-163, Dep. Comput. Sci, Carnegie Mellon Univ., Dec. 1985.
[9] R. Finlayson and D. Cheriton, "Log Files: An Extended File Service Exploiting Write-Once Storage,"Proc. 11th ACM Symp. on Op. Sys. Princ., Nov. 1987, pp. 139-148.
[10] R. Floyd, "Short term file reference patterns in a Unix environment," TR 177, Comput. Sci. Dep., Univ. of Rochester, Mar. 1986.
[11] L. Fujitani, "Laser Optical Disk: The Coming Revolution in On-Line Storage,"Comm. ACM, June 1984, pp. 546-554.
[12] J. Gait, "The optical file cabinet: A random access file system for write once optical disk,"IEEE Comput. Mag., to be published.
[13] S. Garfinkel, "A file system for write once media, MIT Media Lab., Oct. 1986.
[14] R. B. Hagmann, "A crash recovery scheme for a memory-resident database system,"IEEE Trans. Comput., vol. C-35, no. 9, pp. 839- 843, Sept. 1986.
[15] J. Israel,et al., "Separating data from function in a distributed file system," inProc. 2nd Int. Symp. Oper. Syst., IRIA, Rocquencourt, France, Oct. 1978.
[16] M. Jegado, "Recoverability aspects of a distributed file system,"SPAE, vol. 13, pp. 33-44, 1983.
[17] E. Lazowaka, J. Zahorjan, D. Cheriton, and W. Zwaenepoel, "File access performance of diskless workstations,"ACM Trans. Comput. Syst., vol. 4, no. 3, Aug. 1986.
[18] J. Ledleu, "Commercial introduction of digital optical disk technology,"J. Inform. Sci. Princ. Prac., vol. 9, pp. 209-215, 1984.
[19] T. Lehman, "Design and performance evaluation of a main memory relational database system," Tech. Rep. 656, Comput. Sci. Dep., Univ. Wisconsin, Aug. 1986.
[20] M. McKendryet al., "Architecture for a global operating system," inProc. IEEE INFOCOM, San Diego, CA, Apr. 1983.
[21] S. Mullender and A. Tannenbaum, "A distributed file service based on optimistic concurrency control," inProc. 10th Symp. Operating System Principle, Orcas Island, WA, Dec. 1985, pp. 51-62.
[22] J. Ousterhoutet al., "Trace-driven analysis of the Unix 4.2 BSD file system," inProc. 10th ACM Symp. on Operating System Principles, Dec. 1985.
[23] P. Rathman, "Dynamic Data Structures on Optical Disks,"Proc. IEEE Data Eng. Conf., April 1984, pp. 35-42.
[24] A. J. Smith, "Disk cache-miss ratio analysis and design considerations,"ACM Trans. Comput. Syst., vol. 3, no. 3, pp. 161-203, Aug. 1985.
[25] A. Spector and P. Schwartz, "Transactions: A construct for reliable distributed computing,"Oper. Syst. Rev., vol. 17, no. 2, pp. 18-35, 1983.
[26] L. Svoboda, "A reliable object oriented data repository for a distributed computer system," inProc. 8th Symp. Oper. Syst. Principles, Asilomar, CA, Dec. 1981, pp. 47-58.
[27] D. J. Taylor and C. J. Seger, "Robust storage structures for crash recovery,"IEEE Trans. Comput., vol. C-35, pp. 288-295, Apr. 1986.
[28] J. Vitter, "An efficient I/O interface for optical disks," CS-84-15, Dep. Comput. Sci., Brown Univ., June 1984.
[29] J. Wallace, "DMERT crash resistant file systems,"SPAE, vol. 13, pp. 385-387, 1983.
[30] S. Zhouet al., "A file system tracing package for Berkeley Unix,"Usenix Proc., Portland, OR, June 1985, pp. 407-419.

Index Terms:
checkpointing page store; write-once optical disk; model paging system; virtual memory; optical disc storage; virtual storage.
J. Gait, "A Checkpointing Page Store for Write-Once Optical Disk," IEEE Transactions on Computers, vol. 39, no. 1, pp. 2-9, Jan. 1990, doi:10.1109/12.46276
Usage of this product signifies your acceptance of the Terms of Use.