Systems, methods, and machine-readable and executable instructions are provided for network system management. Network system management can include receiving a network system size and a number of system parameters. Network system management can also include receiving a desired monitoring performanc
Systems, methods, and machine-readable and executable instructions are provided for network system management. Network system management can include receiving a network system size and a number of system parameters. Network system management can also include receiving a desired monitoring performance and a desired monitoring quality. Furthermore, network system management can include generating a monitoring system topology for a monitoring and analysis system based on the network system size, the number of system parameters, the desired monitoring performance, and the desired monitoring quality.
대표청구항▼
1. A computer-implemented method for network system management comprising: receiving, at a monitoring system topology constructor, a network system size and a number of system parameters;receiving, at the monitoring system topology constructor, a desired monitoring system performance and a desired m
1. A computer-implemented method for network system management comprising: receiving, at a monitoring system topology constructor, a network system size and a number of system parameters;receiving, at the monitoring system topology constructor, a desired monitoring system performance and a desired monitoring system quality; andgenerating, at the monitoring system topology constructor, a monitoring system topology for a monitoring and analysis system based on the network system size, the number of system parameters, the desired monitoring system performance, and the desired monitoring system quality, wherein the monitoring system topology has a cost that meets or exceeds a threshold based on an actual monitoring system performance, an actual monitoring system quality, and an actual cost of the monitoring and analysis system; andwherein the monitoring system topology is least sensitive to changes in the network system size to maintain the threshold. 2. The method of claim 1 wherein the number of system parameters includes at least one of a number of sensors per node, a network system latency, an available end-to-end bandwidth of the network system, a per-node data size, a per-node failure probability, a per-metric comparison time, a bandwidth budget of the network system for monitoring, and a number of nodes in a particular location of a data center. 3. The method of claim 1 wherein the network system size includes a number of nodes in an infrastructure of the network system. 4. The method of claim 1 wherein the desired monitoring system performance includes at least one of a desired time to insight and a desired number of continuous analytics operations to be delivered per second to the monitoring and analysis system. 5. The method of claim 1 wherein the desired monitoring system quality includes at least one of a desired ratio of available data used for an analysis operation in the presence of a failure, a desired ratio of available data used for an analysis operation in a mismatch of sampling rates during aggregation, and a desired false alarm rate. 6. The method of claim 1, wherein the threshold includes (P*Q)/C wherein P is the actual monitoring system performance, Q is the actual monitoring system quality, and C is the actual cost of the monitoring and analysis system. 7. The method of claim 1 further comprising: receiving, at the monitoring system topology constructor, a desired cost of the monitoring and analysis system; andgenerating, at the monitoring system topology constructor, the system topology for the monitoring and analysis system based on the desired cost. 8. A non-transitory machine-readable medium storing a set of instructions for network system management executable by a computer to cause the computer to: receive a network system size and a number of system parameters;receive a number of analytical models for each of a number of monitoring system topologies;evaluate the number of analytical models using the network system size and the number of system parameters;determine an actual monitoring performance, an actual monitoring quality, and an actual cost for each of the number of analytical models based on the evaluation;receive a desired monitoring performance and a desired monitoring quality;compare the actual monitoring performance and the actual monitoring quality to the desired monitoring performance and the desired monitoring quality;in response to meeting a comparison threshold, select one of the number of analytical models for a monitoring and analysis system based on the comparison and generate a corresponding one of the number of monitoring system topologies, wherein the one of the number of monitoring system topologies has a cost that meets or exceeds a threshold based on the actual monitoring performance, the actual monitoring quality, and the actual cost; andwherein the one of the number of monitoring system topologies is least sensitive to changes in the network system size to maintain the threshold; andin response to not meeting the comparison threshold, dynamically create a hybrid analytical model for the monitoring and analysis system and generate a corresponding monitoring system topologies. 9. The medium of claim 8 wherein the one of the number of monitoring system topologies is generated at runtime. 10. The medium of claim 8 wherein the number of monitoring system topologies include at least one of centralized, hierarchical, decentralized and hybrid monitoring system topologies. 11. The medium of claim 10 wherein the hybrid monitoring system topologies include a combination of at least one of the number of monitoring system topologies, wherein the combination results in the desired monitoring system performance and the desired monitoring system quality within the threshold. 12. A computing system for network system management comprising: a memory;a processor resource coupled to the memory, to:receive a network system size and a number of system parameters;receive a number of analytical models for each of a number of monitoring system topologies;evaluate the number of analytical models using the network system size and the number of system parameters;receive a desired monitoring system performance and a desired monitoring system quality; andgenerate an analytical model of a hybrid monitoring system topology that results in the desired monitoring system performance and the desired monitoring system quality within a cost threshold in response to the evaluation, wherein the cost threshold includes a cost that meets or exceeds a threshold based on an actual monitoring system performance, an actual monitoring quality, and an actual cost of the monitoring system; andwherein the hybrid monitoring system topology is least sensitive to changes in the network system size to maintain the threshold. 13. The system of claim 12 wherein the hybrid monitoring system topology includes a monitoring system topology combining at least one of the number of monitoring system topologies within a particular location of a data center. 14. The system of claim 12 wherein the hybrid monitoring system topology includes a monitoring system topology combining at least one the number of monitoring system topologies across multiple locations of a data center.
연구과제 타임라인
LOADING...
LOADING...
LOADING...
LOADING...
LOADING...
이 특허에 인용된 특허 (18)
Jee-Hoon Song KR; Young-Hyun Kang KR; Cheong-Rim Choi KR, Automatic discovery and positioning method of the network elements in the network management system in case that the network topology is configured.
Binding,Carl; Dolivo,Francois B.; Hermann,Reto J.; Husemann,Dirk; Schade,Andreas, Low-rate wireless personal area network system for tracking containers.
Drysdale R. Scott ; Ennis ; Jr. James D. ; Hasselkus John E. ; Nisbet Thomas R. ; Troutman Robert C. ; Rehbehn Kenneth J. ; Wheeler ; Jr. David E., Method and apparatus for performing service level analysis of communications network performance metrics.
D'Alo,Salvatore; Donatelli,Alex; Marinelli,Claudio; Bernardini,Gianluca; Santoli,Giulio, Method and system for monitoring distributed applications on-demand.
Takubo Shunji,JPX ; Sagawa Nobutoshi,JPX ; Ohta Tadashi,JPX ; Yamaga Susumu,JPX, Method of monitoring a computer system, featuring performance data distribution to plural monitoring processes.
Alan H. Epstein ; Stephen D. Senturia ; Ian A. Waitz ; Jeffrey H. Lang ; Stuart A. Jacobson ; Fredric F. Ehrich ; Martin A. Schmidt ; G. K. Ananthasuresh ; Mark S. Spearing ; Kenneth S. Breu, Microturbomachinery.
Vallone,Eric; Hassell,Suzanne; McFarland,Michael; Johnson,David A., System and method for the collection and display of network performance data in a communication network.
Filip Michael J. (Saratoga CA) Karunungan Kathy L. (Rome MN ITX) Kramer Jeffrey C. (Apple Valley MN) Lee Lucille C. (Rome ITX) Moore Danielle L. (Rome CA ITX) Shih Charles C. (Mountain View CA) Sydir, System for using object-oriented hierarchical representation to implement a configuration database for a layered compute.
Natarajan, Shankar; Harvey, Andrew G.; Lee, Hsuan-Chung; Rawat, Vipin; Pereira, Leo, Technique for providing dynamic modification of application specific policies in a feedback-based, adaptive data network.
Aggarwal, Gautam; Shah, Pritam; Thrivikramannair, Ajith; Blair, Dana, Technique for using OER with an ECT solution for multi-homed spoke-to-spoke sites.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.