Abstract: A cloud network is a complex environment in which hundreds and thousands of users or entities can each host, create, modify, and develop multiple virtual machines. Each virtual machine can have complex behavior unknown to the provider or maintainer of the cloud. Technologies disclosed include methods, systems, and apparatuses to monitor the complex environment to detect network anomalies using machine learning techniques. In addition, techniques to modify and adapt to user feedback are provided allowing the developed models to be tuned for specific use cases, virtual machine types, and users.

Abstract: A cloud network is a complex environment in which hundreds and thousands of users or entities can each host, create, modify, and develop multiple virtual machines. Each virtual machine can have complex behavior unknown to the provider or maintainer of the cloud. Technologies disclosed include methods, systems, and apparatuses to monitor the complex environment to detect network anomalies using machine learning techniques. In addition, techniques to modify and adapt to user feedback are provided allowing the developed models to be tuned for specific use cases, virtual machine types, and users.

Abstract: A cloud network is a complex environment in which hundreds and thousands of users or entities can each host, create, modify, and develop multiple virtual machines. Each virtual machine can have complex behavior unknown to the provider or maintainer of the cloud. Technologies disclosed include methods, systems, and apparatuses to monitor the complex environment to detect network anomalies using machine learning techniques. In addition, techniques to modify and adapt to user feedback are provided allowing the developed models to be tuned for specific use cases, virtual machine types, and users.

Abstract: A method for synchronizing data operations in a multi-threaded computer system using a hybrid lock data structure that allows the computer system to dynamically implement a low contention cost lock or a low overhead cost lock based on the intensity of the memory operation.

Abstract: In one embodiment, a method of reclaiming data storage in a storage device slated as a reclamation target is disclosed. The method includes generating a first list of one or more portions of storage from the reclamation target that each possesses an application programming interface (API) state of unused per a system that uses the storage device. The method also includes identifying in the reclamation target, a reclamation state for each portion of storage from the first list. The method further includes comparing the API state and the reclamation state for each portion of storage in the first list. In addition, the method includes identifying a first subset of portions of storage from the first list as having a mismatched state. The method further includes converting the reclamation state of each of the portion of storage in the first subset from the used state to a marked for reclamation state.

Abstract: A method, a system and/or and an apparatus to minimize data loss by de-allocating storage space with the help of an Application Programming Interface (API). In one embodiment, the method includes reclaiming data storage in a storage device slated as a reclamation target by redirecting write commands intended for the reclamation target to a point-in-time volume of the reclamation target; generating a list of portions of storage from the reclamation target such that each portion of storage has an API state of unused as per a system that uses the storage device; and includes converting a reclamation state associated with each of the portions of storage to a state of ‘unused’. In addition, the method involves synchronization of the point-in-time volume with the reclamation target to capture any changes that might be associated with the point-in-time volume.