Transparent software-defined network management
Systems and methods for network management, including adaptively installing one or more monitoring rules in one or more network devices on a network using an intelligent network middleware, detecting application traffic on the network transparently using an application demand monitor, and predicting future network demands of the network by analyzing historical and current demands. The one or more monitoring rules are updated once counters are collected; and network paths are determined and optimized to meet network demands and maximize utilization and application performance with minimal congestion on the network.
1. A method for network management, comprising:
adaptively installing one or more monitoring rules in one or more network devices on a network using an intelligent network middleware,
wherein the one or more monitoring rules are generated by breaking wildcard rules into specialized rules, each corresponding to a single application;
wherein each of the one or more monitoring rules overlaps, and has a higher priority than one distinct wildcard rule so that it matches a particular traffic flow for a particular duration; and
wherein a wildcard rule that is an exact match for one or more Origin-Destination pairs on initialization is reinstalled as a same rule with a higher priority, same action, after the wildcard rule becomes inactive;
detecting application traffic on the network transparently using an application demand monitor;
predicting future network demands for the network by analyzing historical and current demands;
updating the one or more monitoring rules once counters are collected,
the updating the one or more monitoring rules comprising updating a frequency F to F′, wherein F′=α*F, if the one or more monitoring rules are unstable, wherein α is an update speed and F′ is an updated frequency; and
determining and optimizing network paths to meet network demands and maximize utilization and application performance with minimal congestion on the network.
2. The method according to claim 1 , wherein the predicting future network demands includes bandwidth-based modeling.
3. The method according to claim 1 , wherein the predicting future network demands includes frequency-based modeling.
4. The method according to claim 1 , wherein the one or more monitoring rules are at least one of a temporary rule which counts bytes of a single flow and an exact-match original rule which polls byte counters.
5. The method according to claim 1 , wherein the one or more monitoring rules are governed by granularity, frequency, duration, and switch.
6. The method according to claim 1 , wherein the updating the one or more monitoring rules further comprises updating a frequency F to F′, wherein F′=1/α, if the one or more monitoring rules are stable, wherein α is an update speed and F′ is an updated frequency.
7. A system for network management, comprising:
an intelligent middleware configured to adaptively install one or more monitoring rules in one or more network devices on a network,
wherein the one or more monitoring rules are generated by breaking wildcard rules into specialized rules, each corresponding to a single application;
wherein each of the one or more monitoring rules overlaps, and has a higher priority than one distinct wildcard rule so that it matches a particular traffic flow for a particular duration; and
wherein a wildcard rule that is an exact match for one or more Origin-Destination pairs on initialization is reinstalled as a same rule with a higher priority, same action, after the wildcard rule becomes inactive;
an application demand monitor configured to detect application traffic on the network transparently;
a predictor module configured to predict future network demands of the network by analyzing historical and current demands;
an updating module configured to update the one or more monitoring rules once counters are collected, the update of the one or more monitoring rules comprising updating a frequency F to F′, wherein F′=α*F, if the one or more monitoring rules are unstable, wherein α is an update speed and F′ is an updated frequency; and
an optimization module configured to determine and optimize network paths to meet network demands and maximize utilization and application performance with minimal congestion on the network.
8. The system according to claim 7 , wherein the predicting future network demands includes bandwidth-based modeling.
9. The system according to claim 7 , wherein the predicting future network demands includes frequency-based modeling.
10. The system according to claim 7 , wherein the one or more monitoring rules are at least one of a temporary rule which counts bytes of a single flow and an exact-match original rule which polls byte counters.
11. The system according to claim 7 , wherein the one or more monitoring rules are governed by granularity, frequency, duration, and switch.
12. The system according to claim 7 , wherein the updating the one or more monitoring rules further comprises updating a frequency F to F′, wherein F′=1/α, if the one or more monitoring rules are stable, wherein α is an update speed and F′ is an updated frequency.