Abstract: A request to store a new value to a data entity to be fault tolerant across a plurality of nodes is received. The data entity being in an intermediate state is indicating to at least one less than a majority of the plurality of nodes. A request to read a stored value of the data entity is received. A latest value instance of the data entity among a majority nodes of the plurality of nodes is determined. A value of the latest value instance in a majority nodes of the plurality of nodes is stored as a next value instance of the data entity.

Abstract: A request to store a new value to a data entity to be fault tolerant across a plurality of nodes is received. The data entity being in an intermediate state is indicating to at least one less than a majority of the plurality of nodes. A request to read a stored value of the data entity is received. A latest value instance of the data entity among a majority nodes of the plurality of nodes is determined. A value of the latest value instance in a majority nodes of the plurality of nodes is stored as a next value instance of the data entity.

Abstract: A decision tree analysis system and method for navigating a multi-dimensional decision tree uses acceptable alternative child nodes of a target child node to select an end child node for a parent node, where the parent node and the final child node define a single step of a navigation path for the multi-dimensional decision tree. The acceptable alternative child nodes are based on an acceptance delta parameter for a particular attribute, which defines a value range about an attribute value of the target child node within which a child node is determined to be an acceptable alternative child node of the target child node.

Abstract: A resource management system and method for performing resource management operations in a distributed computer system uses a dispersion rule to try to uniformly disperse clients in a cluster of host computers in the distributed computer system. The dispersion rule is used to compute a dispersion score for at least one candidate distribution of the clients in the cluster of host computers, which is used to select a particular candidate distribution of the clients in the cluster of host computers to disperse the clients.

Abstract: A resource management system and method for performing resource management operations in a distributed computer system uses a dispersion rule to try to uniformly disperse clients in a cluster of host computers in the distributed computer system. The dispersion rule is used to compute a dispersion score for at least one candidate distribution of the clients in the cluster of host computers, which is used to select a particular candidate distribution of the clients in the cluster of host computers to disperse the clients.

Abstract: A decision tree analysis system and method for navigating a multi-dimensional decision tree uses acceptable alternative child nodes of a target child node to select an end child node for a parent node, where the parent node and the final child node define a single step of a navigation path for the multi-dimensional decision tree. The acceptable alternative child nodes are based on an acceptance delta parameter for a particular attribute, which defines a value range about an attribute value of the target child node within which a child node is determined to be an acceptable alternative child node of the target child node.

Abstract: A method for adjusting the configuration of host computers in a cluster on which virtual machines are running in response to a failed change in state is disclosed. The method involves receiving at least one reason a change in state failed the present check or the future check, associating the at least one reason with at least one remediation action, wherein the remediation action would allow the change in state to pass both a present check and a future check, assigning the at least one remediation action a cost, and determining a set of remediation actions to perform based on the cost assigned to each remediation action. In an embodiment, the steps of this method may be implemented in a non-transitory computer-readable storage medium having instructions that, when executed in a computing device, causes the computing device to carry out the steps.

Abstract: A method for supporting a change in state within a cluster of host computers that run virtual machines is disclosed. The method involves identifying a change in state within a cluster of host computers that run virtual machines, determining if predefined criteria for available resources within the cluster of host computers can be met by resources available in the cluster of host computers, and determining if predefined criteria for available resources within the cluster of host computers can be maintained after at least one different predefined change in state. In an embodiment, the steps of this method may be implemented in a non-transitory computer-readable storage medium having instructions that, when executed in a computing device, causes the computing device to carry out the steps.

Abstract: A method for adjusting the configuration of host computers in a cluster on which virtual machines are running in response to a failed change in state is disclosed. The method involves receiving at least one reason a change in state failed the present check or the future check, associating the at least one reason with at least one remediation action, wherein the remediation action would allow the change in state to pass both a present check and a future check, assigning the at least one remediation action a cost, and determining a set of remediation actions to perform based on the cost assigned to each remediation action. In an embodiment, the steps of this method may be implemented in a non-transitory computer-readable storage medium having instructions that, when executed in a computing device, causes the computing device to carry out the steps.

Abstract: A method for supporting a change in state within a cluster of host computers that run virtual machines is disclosed. The method involves identifying a change in state within a cluster of host computers that run virtual machines, determining if predefined criteria for available resources within the cluster of host computers can be met by resources available in the cluster of host computers, and determining if predefined criteria for available resources within the cluster of host computers can be maintained after at least one different predefined change in state. In an embodiment, the steps of this method may be implemented in a non-transitory computer-readable storage medium having instructions that, when executed in a computing device, causes the computing device to carry out the steps.

Abstract: A storage system that accomplishes both robustness and scalability. The storage system includes replicated region servers configured to handle computation involving blocks of data in a region. The storage system further includes storage nodes configured to store the blocks of data in the region, where each of the replicated region servers is associated with a particular storage node of the storage nodes. Each storage node is configured to validate that all of the replicated region servers are unanimous in updating the blocks of data in the region prior to updating the blocks of data in the region. In this manner, the storage system provides end-to-end correctness guarantees for read operations, strict ordering guarantees for write operations, and strong durability and availability guarantees despite a wide range of server failures (including memory corruptions, disk corruptions, etc.) and scales these guarantees to thousands of machines and tens of thousands of disks.