Shoring up persistent applications
SIGMOD '94 Proceedings of the 1994 ACM SIGMOD international conference on Management of data
An analytic behavior model for disk drives with readahead caches and request reordering
SIGMETRICS '98/PERFORMANCE '98 Proceedings of the 1998 ACM SIGMETRICS joint international conference on Measurement and modeling of computer systems
Using System-Level Models to Evaluate I/O Subsystem Designs
IEEE Transactions on Computers
Minerva: An automated resource provisioning tool for large-scale storage systems
ACM Transactions on Computer Systems (TOCS)
Machine Learning
Capturing the spatio-temporal behavior of real traffic data
Performance Evaluation
Analytic Modeling of Clustered RAID with Mapping Based on Nearly Random Permutation
IEEE Transactions on Computers
Hippodrome: Running Circles Around Storage Administration
FAST '02 Proceedings of the Conference on File and Storage Technologies
Data Mining Meets Performance Evaluation: Fast Algorithms for Modeling Bursty Traffic
ICDE '02 Proceedings of the 18th International Conference on Data Engineering
A Modular, Analytical Throughput Model for Modern Disk Arrays
MASCOTS '01 Proceedings of the Ninth International Symposium in Modeling, Analysis and Simulation of Computer and Telecommunication Systems
Issues and Challenges in the Performance Analysis of Real Disk Arrays
IEEE Transactions on Parallel and Distributed Systems
Storage Device Performance Prediction with CART Models
MASCOTS '04 Proceedings of the The IEEE Computer Society's 12th Annual International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunications Systems
Using the Distiller to Direct the Development of Self-Configuration Software
ICAC '04 Proceedings of the First International Conference on Autonomic Computing
Using utility to provision storage systems
FAST'08 Proceedings of the 6th USENIX Conference on File and Storage Technologies
Communications of the ACM - A Direct Path to Dependable Software
SCAN-Lite: enterprise-wide analysis on the cheap
Proceedings of the 4th ACM European conference on Computer systems
DIADS: addressing the "my-problem-or-yours" syndrome with integrated SAN and database diagnosis
FAST '09 Proccedings of the 7th conference on File and storage technologies
Reference-driven performance anomaly identification
Proceedings of the eleventh international joint conference on Measurement and modeling of computer systems
Do you know your IQ?: a research agenda for information quality in systems
ACM SIGMETRICS Performance Evaluation Review
Towards versatile performance models for complex, popular applications
ACM SIGMETRICS Performance Evaluation Review
A load balancing framework for clustered storage systems
HiPC'08 Proceedings of the 15th international conference on High performance computing
Practical performance models for complex, popular applications
Proceedings of the ACM SIGMETRICS international conference on Measurement and modeling of computer systems
BASIL: automated IO load balancing across storage devices
FAST'10 Proceedings of the 8th USENIX conference on File and storage technologies
Storage device performance prediction with selective bagging classification and regression tree
NPC'10 Proceedings of the 2010 IFIP international conference on Network and parallel computing
NPC'10 Proceedings of the 2010 IFIP international conference on Network and parallel computing
Differentiated storage services
ACM SIGOPS Operating Systems Review
Maestro: quality-of-service in large disk arrays
Proceedings of the 8th ACM international conference on Autonomic computing
Italian for beginners: the next steps for SLO-based management
HotStorage'11 Proceedings of the 3rd USENIX conference on Hot topics in storage and file systems
YouChoose: Choosing your Storage Device as a Performance Interface to Consolidated I/O Service
ACM Transactions on Storage (TOS)
No one (cluster) size fits all: automatic cluster sizing for data-intensive analytics
Proceedings of the 2nd ACM Symposium on Cloud Computing
Pesto: online storage performance management in virtualized datacenters
Proceedings of the 2nd ACM Symposium on Cloud Computing
Differentiated storage services
SOSP '11 Proceedings of the Twenty-Third ACM Symposium on Operating Systems Principles
Verifiable resource accounting for cloud computing services
Proceedings of the 3rd ACM workshop on Cloud computing security workshop
QoS support for end users of I/O-intensive applications using shared storage systems
Proceedings of 2011 International Conference for High Performance Computing, Networking, Storage and Analysis
TRACON: interference-aware scheduling for data-intensive applications in virtualized environments
Proceedings of 2011 International Conference for High Performance Computing, Networking, Storage and Analysis
Model building for dynamic multi-tenant provider environments
ACM SIGOPS Operating Systems Review
Responding rapidly to service level violations using virtual appliances
ACM SIGOPS Operating Systems Review
Modeling I/O interference for data intensive distributed applications
Proceedings of the 28th Annual ACM Symposium on Applied Computing
ACIC: automatic cloud I/O configurator for HPC applications
SC '13 Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis
Fourier-assisted machine learning of hard disk drive access time models
PDSW '13 Proceedings of the 8th Parallel Data Storage Workshop
| Hi-index | 0.01 |
Relative fitness is a new black-box approach to modeling the performance of storage devices. In contrast with an absolute model that predicts the performance of a workload on a given storage device, a relative fitness model predicts performance differences between a pair of devices. There are two primary advantages to this approach. First, because are lative fitness model is constructed for a device pair, the application-device feedback of a closed workload can be captured (e.g., how the I/O arrival rate changes as the workload moves from device A to device B). Second, a relative fitness model allows performance and resource utilization to be used in place of workload characteristics. This is beneficial when workload characteristics are difficult to obtain or concisely express (e.g., rather than describe the spatio-temporal characteristics of a workload, one could use the observed cache behavior of device A to help predict the performance of B. This paper describes the steps necessary to build a relative fitness model, with an approach that is general enough to be used with any black-box modeling technique. We compare relative fitness models and absolute models across a variety of workloads and storage devices. On average, relative fitness models predict bandwidth and throughput within 10-20% and can reduce prediction error by as much as a factor of two when compared to absolute models.