Salesforce Com, Inc.

Ronnie Fong

Ashraya Raj Mathur

Jianxun Jason Ding

Shweta Joshi

Jiten Oswal

Steven Bougon

Mihirsinh Raol

Bruno Manuel Torres Gomez

Samir Pilipovic

Jean Steriti

Anshul Kumar

Ranjeeth Karthik Selvan Kathiresan

Systems and methods for a cost-to-serve (CTS) service to measure total cost-to-serve and cost attributions by a plurality of CTS agents spawned across dynamic resources to capture information from a set of instances associated with the plurality of resources; a CTS agent transaction module to publish a set of metrics established by the CTS agent for each instance and usage type; a CTS service collector module to aggregate from each CTS agent, one or more instances from the set of instances to generate transaction metrics; a CTS measurement service module for measuring a total cost for each selected transaction stored at the CTS store based on at least a cost per unit; a CTS metrics processor module for aggregating metrics related to the transactions to determine total cost and set of cost attributions for a selected cloud; and a CTS metrics analytic module to provide cost attribution analytics in the selected cloud in an analytics display.

Systems and methods for real-time determination of cost-to-serve metrics and cost attribution for cloud applications in the public cloud

Systems and methods for an elastic cost-to-serve system including a first module to orchestrate an elastic server set; a second module to orchestrate a load test and to apply one or more use-case scenarios for each orchestrated server set; a third module to generate a cost metrics model of the orchestrated server set for predictive cost modeling; a fourth module coupled to the third module to collect a plurality of performance metrics across the server resources and associated client devices; a fifth module to post-process the collected performance metrics across a load testing duration and to provide analytics of the server set performance; and a sixth module coupled to analyze the performance metrics adapting available resources and to apply a heuristic of the cost metrics model to predict a model of cost optimization of the server set.

Systems and methods for determining optimal cost-to-serve for cloud applications in the public cloud

Embodiments regard composite batching to manage throughput for online commerce applications. An embodiment of one or more storage mediums include instructions for processing online transaction requests for an order management service to generate transaction records, the online transaction requests including at least a first type of transaction and a second type of transaction; accumulating the transaction records in a storage; automatically determining a batch size for application program interface (API) requests for the transaction records and a frequency rate for the API requests based at least in part on a throughput rate for the processed transaction requests; and process the API requests to the order management service.

Composite batching to manage throughput for online commerce applications

Method embodiments are disclosed for characterizing network latency for a component of a webpage provided by an application server device, using boxcarring of action requests. The method comprises measuring the network latency for a component provided by an application server device. A latency category is established based on the network latency. An action request of a user occurring within a queue wait time is associated with the latency category. The action request of the user associated with the latency category is enqueued into an enqueued action request, which is batched in a boxcar to create a batched action request. The batched action request is dispatched in the boxcar to the application server device. The queue wait time is adapted based on an updated network latency and a transmission status of the action request of the user and the batched action request. System and computer program product embodiments are also disclosed.

Characterization of network latency using boxcarring of action requests from component-driven cloud applications

Performance of web pages and mobile device applications with multiple components rendered on a user device is dynamically speeded up, including dynamically measuring network communication latency, adjusting inter-boxcar intervals used to dispatch batches of action requests from the user device to a production server, and dispatching boxcarred requests to the server. Adjustments to the boxcar intervals are based on the dynamically measured network communication latency and a number of connections supported between the user device and the server. The measured network communication latency is calculated as dispatch-to-completed response time minus server processing time and the server processing time is received from the server for a boxcar of completed responses. The system adjusts according to feedback received, as a browser or mobile device changes network connections or the network conditions change, and adapting over time for a particular user. Inter-boxcar intervals are tunable and programmatically changeable, with values learned from experience.

Dynamic adjustment of boxcarring of action requests from component-driven cloud applications

Improved perceived load time for browser and mobile application pages is achieved by adjusting boxcarring of action requests from coupled data consuming applications on the user device, using the priority level of regions and components in component-driven cloud applications. Priority labels differentiate among display regions rendered by the data consuming application and the priority labels further differentiate among components within respective display regions. The middleware application batches the action requests into batches based at least in part on the priority labels, into boxcars segregated by priority label according to a predetermined segregation schedule, and dispatches the boxcars of batched action requests to the server. Performance is also dynamically speeded up, by adjusting inter-boxcar intervals used to dispatch batches of action requests from the user device to a production server, based on the dynamically measured network communication latency between the user device and the server.

Growing community of inventors

San Jose, CA, United States of America

Ronnie Fong