Abstract: A system and method for efficiently transferring virtual machine images across nodes in a cloud computing environment, includes analyzing each image on each node to create hash code clusters and a similarity matrix. An instruction to transfer an image from a source node to a target node is received. The clusters and the similarity matrix are used to determine to what extent the data from the image is already on the source node, or on any other node, and further determines the cost and speed of transferring such data to the target node. An optimal transfer plan is generated, and data that is not already on the target node is transferred to the target node from the most efficient node on which it is available, according to the optimal transfer plan.

Abstract: A method for automated provisioning of computing resources in a distributed computing network includes obtaining a data structure representing a resource dependency characterization of an application to be implemented over the distributed computing network, wherein the data structure indicates a set of resources to be deployed, obtaining a template that represents a predefined network topology for supporting deployment of the set of resources, mapping the set of resources represented in the data structure to the template so as to satisfy a set of component requirements to generate a deployment map, and provisioning the set of resources on a plurality of computers based on the deployment map to implement the application over the distributed computing network.

Abstract: A method for scaling a cloud infrastructure, comprises receiving at least one of resource-level metrics and application-level metrics, estimating parameters of at least one application based on the received metrics, automatically and dynamically determining directives for scaling application deployment based on the estimated parameters, and providing the directives to a cloud service provider to execute the scaling.

Abstract: Co-locating a virtual machine with nested virtualization, in one aspect, may comprise receiving a list of one or more virtual machine instances to co-locate with associated configuration information. A parent virtual machine instance may be identified to host the one or more virtual machine instances. Virtual machine images associated with the one or more virtual machine instances may be consolidated. A composite parent virtual machine image may be created based on the consolidated virtual machine images and the parent virtual machine instance. The parent virtual machine instance may be configured to accommodate for capacity requirement of the one or more child virtual machine instances. The parent virtual machine instance may be started on a processor via the composite parent virtual machine image. The one or more virtual machine instances may be started within the parent virtual machine instance as one or more child virtual machine instances.

Abstract: One embodiment of a system for provisioning a new virtual machine instance on a target host based on an image of the new virtual machine instance includes a local storage for storing images of virtual machine instances currently running on the target host and a virtual machine creator coupled to the local storage for constructing the image of the new virtual machine instance at least in part using the images of virtual machine instances currently running on the target host.

Abstract: An apparatus for enabling resiliency for cloud computing systems is provided. An apparatus includes a processor and a memory storing computer program code. The memory and the computer program code are configured to, with the processor, cause the apparatus to perform actions. The actions include modifying a topology graph of a network architecture by mapping processes flows onto the topology graph and creating a resiliency graph based on the modified topology graph. The actions also include modifying the resiliency graph by translating at least one SLA into the resiliency graph and identifying overlaps and dependencies in the modified resiliency graph. Apparatus and computer readable instructions are also described.

Abstract: One embodiment of a system for provisioning a new virtual machine instance on a target host based on an image of the new virtual machine instance includes a local storage for storing images of virtual machine instances currently running on the target host and a virtual machine creator coupled to the local storage for constructing the image of the new virtual machine instance at least in part using the images of virtual machine instances currently running on the target host.

Abstract: One embodiment of a method for provisioning a new virtual machine instance based on the content of an image of the new virtual machine instance includes identifying, from among a plurality of host machines, the host machine having the highest percentage of the content available in local storage, and provisioning the new virtual machine instance on the host machine having the highest percentage of the content available in local storage. Another embodiment of a method for provisioning a new virtual machine instance based on an image of the new virtual machine instance includes constructing at least a portion of the image using data stored locally on a target machine hosting the new virtual machine instance, and completing the image using data obtained over a network from remote storage.

Abstract: Co-locating a virtual machine with nested virtualization, in one aspect, may comprise receiving a list of one or more virtual machine instances to co-locate with associated configuration information. A parent virtual machine instance may be identified to host the one or more virtual machine instances. Virtual machine images associated with the one or more virtual machine instances may be consolidated. A composite parent virtual machine image may be created based on the consolidated virtual machine images and the parent virtual machine instance. The parent virtual machine instance may be configured to accommodate for capacity requirement of the one or more child virtual machine instances. The parent virtual machine instance may be started on a processor via the composite parent virtual machine image. The one or more virtual machine instances may be started within the parent virtual machine instance as one or more child virtual machine instances.

Abstract: Techniques for improving virtual machine-to-image affinity on a physical server. A method includes identifying physical machines in a network as candidate source physical machines, wherein each candidate source physical machine stores a first virtual machine image and a set of additional virtual machine images, identifying physical machines in the network as candidate target physical machines, wherein each candidate target physical machine stores one of the additional virtual machine images, and selecting a virtual machine image from the set of additional virtual machine images and selecting a physical machine from the candidate target physical machines such that migrating the selected virtual machine image from a candidate source physical machine to the selected target physical machine results in a maximized image affinity per virtual machine in comparison to each image migration scenarios for the set of additional virtual machine images.

Abstract: Systems and articles of manufacture for improving virtual machine-to-image affinity on a physical server include identifying physical machines in a network as candidate source physical machines, wherein each candidate source physical machine stores a first virtual machine image and a set of additional virtual machine images, identifying physical machines in the network as candidate target physical machines, wherein each candidate target physical machine stores one of the additional virtual machine images, and selecting a virtual machine image from the set of additional virtual machine images and selecting a physical machine from the candidate target physical machines such that migrating the selected virtual machine image from a candidate source physical machine to the selected target physical machine results in a maximized image affinity per virtual machine in comparison to each image migration scenarios for the set of additional virtual machine images.

Abstract: A method for enabling resiliency for cloud computing systems is described. The method includes modifying a topology graph of a network architecture by mapping processes flows onto the topology graph. A resiliency graph is created based on the modified topology graph. The method includes modifying the resiliency graph by translating at least one SLA into the resiliency graph. Overlaps and dependencies in the modified resiliency graph are identified. Apparatus and computer readable instructions are also described.

Abstract: A method for scaling a cloud infrastructure, comprises receiving at least one of resource-level metrics and application-level metrics, estimating parameters of at least one application based on the received metrics, automatically and dynamically determining directives for scaling application deployment based on the estimated parameters, and providing the directives to a cloud service provider to execute the scaling.

Abstract: A system and an article of manufacture for de-duplicating virtual machine image accesses include identifying one or more identical blocks in two or more images in a virtual machine image repository, generating a block map for mapping different blocks with identical content into a same block, deploying a virtual machine image by reconstituting an image from the block map and fetching any unique blocks remotely on-demand, and de-duplicating virtual machine image accesses by storing the deployed virtual machine image in a local disk cache.

Abstract: An apparatus for enabling resiliency for cloud computing systems is provided. An apparatus includes a processor and a memory storing computer program code. The memory and the computer program code are configured to, with the processor, cause the apparatus to perform actions. The actions include modifying a topology graph of a network architecture by mapping processes flows onto the topology graph and creating a resiliency graph based on the modified topology graph. The actions also include modifying the resiliency graph by translating at least one SLA into the resiliency graph and identifying overlaps and dependencies in the modified resiliency graph. Apparatus and computer readable instructions are also described.

Abstract: Disclosed is a method, computer-program product and a system to compose an application solution on a cloud-computing platform for a user. The method includes, in response to input from the user, selecting at least one predetermined template that includes a hardware and software instance of a data processing system that is implementable as an solution on a cloud platform; comparing the at least one selected template against a set of constraints; establishing an instance of a cloud based solution for the user based on the selected template; and validating the established cloud based solution against user-uploaded information such as software and workloads to ensure that the solution satisfies requirements of the user, such as both technical and financial requirements of the user.

Abstract: Systems and articles of manufacture for improving virtual machine-to-image affinity on a physical server include identifying physical machines in a network as candidate source physical machines, wherein each candidate source physical machine stores a first virtual machine image and a set of additional virtual machine images, identifying physical machines in the network as candidate target physical machines, wherein each candidate target physical machine stores one of the additional virtual machine images, and selecting a virtual machine image from the set of additional virtual machine images and selecting a physical machine from the candidate target physical machines such that migrating the selected virtual machine image from a candidate source physical machine to the selected target physical machine results in a maximized image affinity per virtual machine in comparison to each image migration scenarios for the set of additional virtual machine images.

Abstract: Techniques, systems, and articles of manufacture for improving virtual machine-to-image affinity on a physical server. A method includes identifying physical machines in a network as candidate source physical machines, wherein each candidate source physical machine stores a first virtual machine image and a set of additional virtual machine images, identifying physical machines in the network as candidate target physical machines, wherein each candidate target physical machine stores one of the additional virtual machine images, and selecting a virtual machine image from the set of additional virtual machine images and selecting a physical machine from the candidate target physical machines such that migrating the selected virtual machine image from a candidate source physical machine to the selected target physical machine results in a maximized image affinity per virtual machine in comparison to each image migration scenarios for the set of additional virtual machine images.

Abstract: Disclosed is a method, computer-program product and a system to compose an application solution on a cloud-computing platform for a user. The method includes, in response to input from the user, selecting at least one predetermined template that includes a hardware and software instance of a data processing system that is implementable as an solution on a cloud platform; comparing the at least one selected template against a set of constraints; establishing an instance of a cloud based solution for the user based on the selected template; and validating the established cloud based solution against user-uploaded information such as software and workloads to ensure that the solution satisfies requirements of the user, such as both technical and financial requirements of the user.

Abstract: A system and method for efficiently transferring virtual machine images across nodes in a cloud computing environment, includes analyzing each image on each node to create hash code clusters and a similarity matrix. An instruction to transfer an image from a source node to a target node is received. The clusters and the similarity matrix are used to determine to what extent the data from the image is already on the source node, or on any other node, and further determines the cost and speed of transferring such data to the target node. An optimal transfer plan is generated, and data that is not already on the target node is transferred to the target node from the most efficient node on which it is available, according to the optimal transfer plan.