System and method for providing intelligent diagnostic support for cloud-based infrastructure

A method for providing intelligent support for a communications network, the method comprising:

• configuring a cloud ecosystem state (CES) model with respect to at least a network portion of the communications network, wherein the CES model is operative to define a cloud ecosystem state comprising a plurality of variables having a scalable dimensionality based on one or more operational characteristics of the network portion and wherein the cloud ecosystem state corresponds to one or more time-stamped records of diagnostic data, the cloud ecosystem state including instructions regarding retrieving a diagnostic entity associated with the cloud ecosystem state and performing an evaluation of the diagnostic data of the diagnostic entity;
• obtaining diagnostics data corresponding to the cloud ecosystem state from one or more infrastructure elements of the network portion;
• analyzing the diagnostics data corresponding to the cloud ecosystem state and providing one or more feedback messages operative to adaptively update the cloud ecosystem state with respect to at least one of the scalable dimensionality of the plurality of variables or the operational characteristics of the network portion, such analyzing further comprising:
• obtaining a time-stamped record of the cloud ecosystem state at a particular time;
• retrieving a sequence of stored time-stamped records of the cloud ecosystem state over a configurable timing window prior to the particular time, each stored time-stamped record comprising diagnostic data corresponding to a respective time within the configurable timing window prior to the particular time;
• generating a cumulative time series of the time-stamped records including the time-stamped record at the particular time and the retrieved sequence of the stored time-stamped records over the configurable timing window;
• determining if a matching pattern of the cumulative time series of the time-stamped records associated with the cloud ecosystem state exists in a reference database;
• responsive to determining that there is no matching pattern of the cumulative time series of the time-stamped records in the reference database, performing:
• storing the time-stamped record of the cloud ecosystem state obtained at the particular time in the reference database;
• applying machine learning to identify a pattern including the time-stamped record for the cloud ecosystem state;
• and
• updating the reference database to include one or more corrective actions applicable with respect to the identified pattern including the time-stamped record.

A non-transitory machine-readable storage medium having program instructions thereon, which are configured to effectuate an intelligent diagnostics support system for a communications network when executed by one or more processors of a network diagnostics center, the non-transitory machine-readable storage medium comprising:

• a code portion to configure a cloud ecosystem state (CES) model with respect to at least a network portion of the communications network, wherein the CES model is operative to define a cloud ecosystem state comprising a plurality of variables having a scalable dimensionality based on one or more operational characteristics of the network portion and wherein the cloud ecosystem state corresponds to one or more time-stamped records of diagnostic data, the cloud ecosystem state including instructions regarding retrieving a diagnostic entity associated with the cloud ecosystem state and performing an evaluation of the diagnostic data of the diagnostic entity;
• a code portion for obtaining diagnostics data corresponding to the cloud ecosystem state from one or more infrastructure elements of the network portion;
• a code portion for analyzing the diagnostics data corresponding to the cloud ecosystem state and providing one or more feedback messages operative to adaptively update the cloud ecosystem state with respect to at least one of the scalable dimensionality of the plurality of variables or the operational characteristics of the network portion;
• a code portion for obtaining a time-stamped record of the cloud ecosystem state at a particular time;
• a code portion for performing a status check of the cloud ecosystem state with respect to the time-stamped record;
• a code portion, responsive to determining that the status check of the cloud ecosystem state is not successful, for determining whether there is a similar cloud ecosystem state stored in a reference database;
• and
• a code portion, responsive to determining that there is no similar cloud ecosystem state found in the reference database, for storing the time-stamped record of the cloud ecosystem state in the reference database and updating the reference database to include one or more corrective actions applicable with respect to the diagnostic data comprising the time-stamped record.

A non-transitory machine-readable storage medium having program instructions thereon, which are configured to effectuate an intelligent diagnostics support system for a communications network when executed by one or more processors of a network diagnostics center, the non-transitory machine-readable storage medium comprising:

• a code portion to configure a cloud ecosystem state (CES) model with respect to at least a network portion of the communications network, wherein the CES model is operative to define a cloud ecosystem state comprising a plurality of variables having a scalable dimensionality based on one or more operational characteristics of the network portion and wherein the cloud ecosystem state corresponds to one or more time-stamped records of diagnostic data, the cloud ecosystem state including instructions regarding retrieving a diagnostic entity associated with the cloud ecosystem state and performing an evaluation of the diagnostic data of the diagnostic entity;
• a code portion for obtaining diagnostics data corresponding to the cloud ecosystem state from one or more infrastructure elements of the network portion;
• a code portion for analyzing the diagnostics data corresponding to the cloud ecosystem state and providing one or more feedback messages operative to adaptively update the cloud ecosystem state with respect to at least one of the scalable dimensionality of the plurality of variables or the operational characteristics of the network portion;
• a code portion for obtaining a time-stamped record of the cloud ecosystem state at a particular time;
• a code portion for retrieving a sequence of stored time-stamped records of the cloud ecosystem state over a configurable timing window prior to the particular time, each stored time-stamped record comprising diagnostic data corresponding to a respective time within the configurable timing window prior to the particular time;
• a code portion for generating a cumulative time series of the time-stamped records including the time-stamped record at the particular time and the retrieved sequence of the stored time-stamped records over the configurable timing window;
• a code portion for determining if a matching pattern of the cumulative time series of the time-stamped records associated with the cloud ecosystem state exists in a reference database;
• a code portion, responsive to determining that there is no matching pattern of the cumulative time series of the time-stamped records in the reference database, for performing:
• storing the time-stamped record of the cloud ecosystem state obtained at the particular time in the reference database;
• applying machine learning to identify a pattern including the time-stamped record for the cloud ecosystem state;
• and
• updating the reference database to include one or more corrective actions applicable with respect to the identified pattern including the time-stamped record.