Implementations described and claimed herein provide methods and systems for routing data, message, and register access transactions through common ports in an on-chip interface interconnect module while supporting Quality of Service (QoS) and maintaining fair data throughput. In one implementation, a System on Chip system includes an AXI interconnect module to route data, message, and register access transactions via common ports in the AXI interconnect module. The system may include at least one master circuit module configured to initiate data transaction, wherein the at least one master circuit module behaves as a slave to receive message transactions and register access transactions. The system may include at least one slave circuit module configured to respond to data transactions, wherein the at least one slave circuit module behaves as a master to initiate message transactions. The system may use QoS signaling as a priority indicator to prioritize the transactions.

Embodiments generally relate to selecting a service instance in a service infrastructure. In some embodiments, a method includes sending, by a service registry, a status request to each service instance of a plurality of service instances, where the service registry maintains a data store of performance information associated each of the service instances. The method further includes receiving, by the service registry, a plurality of status responses, where each status response is received from a respective service instance of the plurality service instances, and where each status response includes one or more performance characteristics. The method further includes ranking, by the service registry, the service instances based at least in part on the one or more performance characteristics. The method further includes performing, by the service registry, service lookups based on the ranking.

A system includes a load balancer and storage including a first data structure and a second data structure. Each of the data structures includes a plurality of different weight levels. At least one of the weight levels of the first data structure includes an identifier of a target. At least one of the weight levels of the second data structure includes a map that associates a target identifier with a final weight value. Responsive to receipt of a request and responsive to a target identifier included at a given weight level in the first data structure corresponding to a pointer, the load balancer selects the target identified in the first data structure at the given weight level. Responsive to no target identifiers included at the given weight level in the first data structure, the load balancer selects a target identified in the second data structure at the given weight level. The load balancer forwards the request to the selected target.

Techniques and structures to prevent exhaustion of a database connection pool, including receiving a plurality of events from a plurality of application servers operating as a database connection pool, categorizing each of the plurality of events, including generating a categorization message for each of the plurality events, wherein each categorization message includes one or more fields having contextual information to identify an event type to which an event is associated, analyzing the categorization messages to generate an event record; and generating a report based on the event record including one or more performance measurements to indicate an effectiveness of management of the database connection pool.

Disclosed are systems and methods for forwarding service requests to an idle server from among a plurality of servers. For example, the method comprises, by a load balancer, tracking service requests on each backend server, estimating an additional load associated with currently pending requests based on resource usage statistics reported in a previous time interval, predicting which of the plurality of servers is most likely to be the most idle server for a next time interval using a statistical model based on the estimated additional load associated with currently pending requests, and forwarding service requests to the most idle server based on the prediction.

Examples provide a configurable connection reset system for customized distribution of incoming traffic to a plurality of servers. The connection reset component routes incoming service requests to active servers based on server performance, configurable server failure criteria, user-selected inactive threshold wait time-period for each server and per-client distribution configurations, including server priorities and user-selected distribution methodology or combination of distribution methodologies. A configurable wait time-period is assigned to each inactive server removed from a distribution rotation for the incoming traffic. A next received request to the next available server for the identified client. An inactive server is updated to active status and returned to the distribution rotation after expiration of the configurable wait time-period.

Disclosed are systems, apparatus, methods, and computer readable media for interfacing with a phone system in an on-demand service environment. In one embodiment, a client machine is in communication with a data provider and the phone system. The phone system is operated by a phone provider separate from the data provider. A message is received at the client machine from the phone system. The message includes information regarding a phone event in relation to a phone call. The phone event information is provided in a first part of a user interface in a web browser on the client machine. A second part of the user interface is capable of displaying information received at the client machine from the data provider. The first part of the user interface and the second part of the user interface are adapted to communicate with one another. For example, the first part of the user interface is adapted to send a message including the phone event information to the second part of the user interface.

A packet processing method includes: receiving, by a load balancing apparatus, a data packet sent by a UE; when the data packet includes historical association information, selecting, by the load balancing apparatus based on the historical association information, a service instance used to process the data packet; and sending, by the load balancing apparatus, the data packet to the selected service instance.

In an embedding caching system, embeddings generated from previous problems are re-used to improve performance on future problems. A data structure stores problems and their corresponding embeddings. When computing future embeddings, this data structure can be queried to determine whether an embedding has already been computed for a problem with the same structure. If it has, the embedding can be retrieved from the data structure, saving the time and computational expense of generating a new embedding. In one variation, the query is not based on exact matches. If a new problem is similar in structure to previous problems, those embeddings may be used to accelerate the generating of an embedding for the new problem, even if they cannot be used directly to embed the new problem.

A method for assigning clients to backup resources of a backup pool includes receiving backup resource requests from the clients. The method includes assigning a portion of the clients to a first backup resource of the backup pool while the first backup resource is not overloaded. The method includes after assigning the portion of the clients, obtaining status information for the first backup resource of the backup pool. The method includes making a first determination that the first backup resource is overloaded based on the status information and a load policy. The method includes in response to the first determination, determining that an attribute of the first backup resource is available for reconfiguration. The method includes reconfiguring the attribute of the first backup resource to obtain a reconfigured backup resource. The method includes assigning a client, which has not been assigned to the backup resources, to the reconfigured backup resource.