Abstract: A ProbeSystem for one or more Replicas is disclosed. For each replica, the Probe System includes a ProbeHead that updates a Head field in a Target Record pertaining to the Replica and a ProbeEcho that updates an Echo field in the TargetRecord. The Head field includes a time stamp and an epoch value. The Echo field includes a latency value and an epoch value. The fields are such that, during reporting, the ProbeHead can analyze the TargetRecord for the replica to determine the latency between one or more paths between the Replica and other Replicas. The ProbeSystem has a peer-to-peer shared nothing architecture and applies to a large class of high-replica, distributed and high-scale cloud storage systems and services.

Abstract: Method, system, and programs for resolving data inconsistency in a distributed system having a plurality of replica instances. In response to a read request of at least one data element, a key-specific probability value for the at least one data element of the read request is retrieved. Based on the retrieved key-specific probability value and a system load of the distributed system, a self-healing probability of the at least one data element of the read request is computed. A self-healing transaction that includes the at least one data element of the read request is then generated based on the self-healing probability of the at least one data element. The generated self-healing transaction is further applied to at least some of the plurality of replica instances in the distributed system to resolve data inconsistency.

Abstract: A ProbeSystem for one or more Replicas is disclosed. For each replica, the Probe System includes a ProbeHead that updates a Head field in a Target Record pertaining to the Replica and a ProbeEcho that updates an Echo field in the TargetRecord. The Head field includes a time stamp and an epoch value. The Echo field includes a latency value and an epoch value. The fields are such that, during reporting, the ProbeHead can analyze the TargetRecord for the replica to determine the latency between one or more paths between the Replica and other Replicas. The ProbeSystem has a peer-to-peer shared nothing architecture and applies to a large class of high-replica, distributed and high-scale cloud storage systems and services.

Abstract: A ProbeSystem for one or more Replicas is disclosed. For each replica, the Probe System includes a ProbeHead that updates a Head field in a Target Record pertaining to the Replica and a ProbeEcho that updates an Echo field in the TargetRecord. The Head field includes a time stamp and an epoch value. The Echo field includes a latency value and an epoch value. The fields are such that, during reporting, the ProbeHead can analyze the TargetRecord for the replica to determine the latency between one or more paths between the Replica and other Replicas. The ProbeSystem has a peer-to-peer shared nothing architecture and applies to a large class of high-replica, distributed and high-scale cloud storage systems and services.

Abstract: Method, system, and programs for resolving data inconsistency in a distributed system having a plurality of replica instances. In response to a read request of at least one data element, a key-specific probability value for the at least one data element of the read request is retrieved. Based on the retrieved key-specific probability value and a system load of the distributed system, a self-healing probability of the at least one data element of the read request is computed. A self-healing transaction that includes the at least one data element of the read request is then generated based on the self-healing probability of the at least one data element. The generated self-healing transaction is further applied to at least some of the plurality of replica instances in the distributed system to resolve data inconsistency.

Abstract: A compile server and a class loader are disclosed. During execution of a program, the class loader transmits a compile service request and a run time service request, including a program instruction, over a network to a compile server. The class loader receives native machine code compiled from the instruction, where the native machine code is received over the network. Once received, the native machine code is executed. The compile server receives the compile service request and the run time service request, including the program instruction, over a network from the class loader, compiles the instruction into native machine code, and transmits the native machine code over the network to be executed. The compile server generates a compile thread for each compile service requests and processes multiple compile threads concurrently.

Abstract: A method for enabling an application program to communicate with a network server, includes the steps of downloading a document from a document server to the application program, downloading code from a code server associated with the document server to the application program, the code including a network protocol handler for the network server, and using the network protocol handler to communicate with the network server.

Abstract: A compile server and a class loader are disclosed. During execution of a program, the class loader transmits a compile service request and a run time service request, including a program instruction, over a network to a compile server. The class loader receives native machine code compiled from the instruction, where the native machine code is received over the network. Once received, the native machine code is executed. The compile server receives the compile service request and the run time service request, including the program instruction, over a network from the class loader, compiles the instruction into native machine code, and transmits the native machine code over the network to be executed. The compile server generates a compile thread for each compile service requests and processes multiple compile threads concurrently.

Abstract: A method for enabling an application program to communicate with a network server, includes the steps of downloading a document from a document server to the application program, downloading code from a code server associated with the document server to the application program, the code including a network protocol handler for the network server, and using the network protocol handler to communicate with the network server.

Abstract: A distributed garbage collection system and method is disclosed that is compatible with local ref-count or full garbage collection and that ensures that no local object's storage is deleted by the local garbage collector unless it is certain that there are no actual or potential remote references to that local object. The disclosed system and method are implemented in the context of a transparent distributed object system in which communications between objects in different processes are enabled by dedicated proxy objects that are linked to corresponding remote objects via a pair of transport objects. Additional proxy holder objects and proxy holder proxies ensure that objects for which third-party object references are passed (i.e., where one object in a first process passes a remote object in a second process a reference to a third object in a third process) are not collected until a direct link is established between the remote object in the second process and the third object in the third object space.

Abstract: A system and method provides for variable target outputs from a compiler with only a single execution. The compiler includes a front end, a generic back end, and plurality of individual back ends that are dynamically loaded by the compiler during execution. The front end produces an abstract syntax tree which is then processed by the generic back end and the individual back ends to produce a number of back end trees, each adapted for a specific back end, and representative of the desired structure of the various target outputs, including target code files, or events. The generic back end traverses the abstract syntax tree, and on each node of the tree, invokes each back end that has a node of its back end tree attached thereto. The back end can then modify its own back end tree, and attach further back end nodes to other nodes of the abstract syntax tree.