Software—Practice & Experience
On the self-similar nature of Ethernet traffic (extended version)
IEEE/ACM Transactions on Networking (TON)
Approximate medians and other quantiles in one pass and with limited memory
SIGMOD '98 Proceedings of the 1998 ACM SIGMOD international conference on Management of data
Self-similarity in file systems
SIGMETRICS '98/PERFORMANCE '98 Proceedings of the 1998 ACM SIGMETRICS joint international conference on Measurement and modeling of computer systems
A trace-driven analysis of the UNIX 4.2 BSD file system
Proceedings of the tenth ACM symposium on Operating systems principles
A relational model of data for large shared data banks
Communications of the ACM
Data Cube: A Relational Aggregation Operator Generalizing Group-By, Cross-Tab, and Sub-Totals
Data Mining and Knowledge Discovery
New NFS Tracing Tools and Techniques for System Analysis
LISA '03 Proceedings of the 17th USENIX conference on System administration
Passive NFS Tracing of Email and Research Workloads
FAST '03 Proceedings of the 2nd USENIX Conference on File and Storage Technologies
Buttress: A Toolkit for Flexible and High Fidelity I/O Benchmarking
FAST '04 Proceedings of the 3rd USENIX Conference on File and Storage Technologies
MEMS-based Storage Devices and Standard Disk Interfaces: A Square Peg in a Round Hole?
FAST '04 Proceedings of the 3rd USENIX Conference on File and Storage Technologies
The devil and packet trace anonymization
ACM SIGCOMM Computer Communication Review
TBBT: scalable and accurate trace replay for file server evaluation
FAST'05 Proceedings of the 4th conference on USENIX Conference on File and Storage Technologies - Volume 4
A flash-memory based file system
TCON'95 Proceedings of the USENIX 1995 Technical Conference Proceedings
Measurement and analysis of large-scale network file system workloads
ATC'08 USENIX 2008 Annual Technical Conference on Annual Technical Conference
DataSeries: an efficient, flexible data format for structured serial data
ACM SIGOPS Operating Systems Review
Discovery of application workloads from network file traces
FAST'10 Proceedings of the 8th USENIX conference on File and storage technologies
Improving the efficiency of information collection and analysis in widely-used IT applications
Proceedings of the 2nd ACM/SPEC International Conference on Performance engineering
Understanding and Improving Computational Science Storage Access through Continuous Characterization
ACM Transactions on Storage (TOS)
Analysis of Workload Behavior in Scientific and Historical Long-Term Data Repositories
ACM Transactions on Storage (TOS)
Extracting flexible, replayable models from large block traces
FAST'12 Proceedings of the 10th USENIX conference on File and Storage Technologies
LoadIQ: learning to identify workload phases from a live storage trace
HotStorage'12 Proceedings of the 4th USENIX conference on Hot Topics in Storage and File Systems
Workload diversity and dynamics in big data analytics: implications to system designers
Proceedings of the 2nd Workshop on Architectures and Systems for Big Data
Usage behavior of a large-scale scientific archive
SC '12 Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis
Building intelligence for software defined data centers: modeling usage patterns
Proceedings of the 6th International Systems and Storage Conference
Generating request streams on Big Data using clustered renewal processes
Performance Evaluation
Characterization of incremental data changes for efficient data protection
USENIX ATC'13 Proceedings of the 2013 USENIX conference on Annual Technical Conference
| Hi-index | 0.00 |
We describe methods to capture, convert, store and analyze NFS workloads that are 20-100× more intense, in terms of operations/day, than any previously published. We describe three techniques that improve capture performance by up to 10× over previous techniques. For conversion, we use a general-purpose format that is both highly space efficient and provides efficient access to the trace data. For analysis, we describe a number of techniques adopted from the database community and some new techniques that facilitate analysis of very large traces. We also describe a number of guidelines for trace collection that should prove useful to future practitioners. Finally, we analyze a commercial feature animation (movie) rendering workload using these techniques and discuss the characteristics of the workload. Our implementation of these techniques is available as open source and the exact anonymized datasets we analyze are available for free download.