Granted Patent B2

US 10,075,549 · App. 15/005,846 · Granted Sep 11, 2018

Abstract

Claims (46)

1. A method, performed by an optimizer module that functions as a proxy between a server that provides access to a resource and a number of clients, for optimizing the performance of the server, the method comprising:

receiving, at the optimizer module, requests for accessing the resource from a number of clients;

buffering the received requests during a buffering interval;

creating a number of bulk requests by combining multiple buffered requests into each bulk request;

creating mappings which define where to send responses to the requests combined into each bulk request;

after the buffering interval, sending the bulk requests to the server, each bulk request comprising a single call to a web services application programming interface (API) that specifies each of the multiple requests that have been combined into the bulk request;

for each bulk request, receiving a bulk response that includes multiple responses corresponding to the multiple requests;

parsing each bulk response to extract the multiple responses;

employing the mappings to identify the corresponding client for each of the extracted responses; and

sending each of the responses to the corresponding client based on the mappings.

2. The method of claim 1 , wherein the buffering interval is a periodic interval.

3. The method of claim 1 , wherein the buffering interval is 100 milliseconds.

4. The method of claim 1 , wherein the web services API is a representational state transfer (REST) API.

5. The method of claim 1 , wherein the mappings identify a socket used by the optimizer module to receive the corresponding request.

6. The method of claim 1 , wherein creating a number of bulk requests by combining multiple requests into each bulk request comprises:

selectively combining multiple requests to create bulk requests that are optimized for the server based on one or more of a server specification or monitored performance characteristics of the server.

7. The method of claim 6 , wherein the multiple requests are selectively combined to yield an ordering of requests within the bulk request that is optimal for the server.

8. The method of claim 6 , wherein the multiple requests are selectively combined to yield a bulk request having an optimal size.

9. The method of claim 6 , wherein the multiple requests are selectively combined to yield an optimal number of bulk requests for parallel submission to the server.

10. The method of claim 6 , wherein the multiple requests are selectively combined to yield a bulk request having an optimal compression ratio.

11. The method of claim 6 , further comprising:

reordering parameters of one or more requests within at least one bulk request to be more optimal for the server.

12. The method of claim 1 , wherein the resource is a database and the requests comprise database operations.

13. A system for enabling more clients to concurrently submit hypertext transfer protocol (HTTP) requests to a server, the system comprising:

the server that is configured to provide web services API through which HTTP requests are submitted for accessing a resource; and

an optimizer module executing on the server or on another server, the optimizer module being configured to function as a proxy between the server and clients, the optimizer module being configured to:

receive and buffer the clients' HTTP requests during a buffering interval;

generate bulk requests by combining multiple buffered HTTP requests into each bulk request;

create mappings which define where to send responses to the HTTP requests combined into each bulk request;

submit the bulk requests to the server via the web services API such that the server receives and processes the bulk requests rather than the individual HTTP requests sent by the clients;

receive bulk responses to the bulk requests;

parse the bulk responses to extract each individual response;

employing the mappings to identify the corresponding client for each of the individual responses; and

for each individual response, send an HTTP response to the corresponding client.

14. The system of claim 13 , wherein the optimizer module is further configured to selectively combine multiple buffered HTTP requests into each bulk request to form bulk requests that are optimized for the server based on one or more of a server specification or monitored performance characteristics of the server.

15. One or more computer storage media storing computer executable instructions which when executed by one or more processors implements a method for increasing the performance of a server that provides access to a database via a REST API, the method comprising:

creating, by an optimizer module that functions as a proxy for the server, a buffer for buffering requests for accessing the database;

receiving, by the optimizer module and during a buffering interval, a number of requests;

storing the received requests in the buffer;

after the buffering interval, creating a number of bulk requests by combining multiple buffered requests into each bulk request;

creating mappings which define where to send responses to the requests combined into each bulk request;

sending the number of bulk requests to the server via the REST API, each bulk request comprising a single call to the REST API;

for each bulk request, receiving a bulk response that includes multiple responses corresponding to the multiple requests;

parsing each bulk response to extract the multiple responses;

employing the mappings to identify the corresponding client for each of the extracted responses; and

sending each of the responses to the corresponding client based on the mappings.

Assignments (32)