US 12,093,261 B2
Storage formats for in-memory caches
Aurosish Mishra, Belmont, CA (US); Shasank K. Chavan, Menlo Park, CA (US); Vinita Subramanian, Campbell, CA (US); Ekrem S. C. Soylemez, Lexington, MA (US); Adam Kociubes, Bedford, MA (US); Eugene Karichkin, Fremont, CA (US); and Garret F. Swart, Palo Alto, CA (US)
Assigned to Oracle International Corporation, Redwood Shores, CA (US)
Filed by Oracle International Corporation, Redwood Shores, CA (US)
Filed on Apr. 2, 2018, as Appl. No. 15/943,335.
Claims priority of provisional application 62/565,832, filed on Sep. 29, 2017.
Prior Publication US 2019/0102391 A1, Apr. 4, 2019
Int. Cl. G06F 16/24 (2019.01); G06F 16/172 (2019.01); G06F 16/22 (2019.01); G06F 16/2455 (2019.01)
CPC G06F 16/24552 (2019.01) [G06F 16/172 (2019.01); G06F 16/221 (2019.01); G06F 16/2455 (2019.01); G06F 16/24553 (2019.01)] 28 Claims
OG exemplary drawing
 
1. A method comprising:
for each numeric value in a set of numeric values, storing, in a cache, a respective application-level representation that represents said each numeric value and a respective hardware-level representation that represents said each numeric value, wherein said respective application-level representation belongs to a set of application-level representations and said respective hardware-level representation belongs to a set of hardware-level representations, wherein each hardware-level representation of said set of hardware-level representations includes a value configured to be loadable into a register;
responsive to receiving a query to be executed over said set of numeric values, making a determination as to whether or not it would be more efficient to execute said query over said set of hardware-level representations than to execute said query over said set of application-level representations;
executing said query over said set of hardware-level representations when said determination indicates that it would be more efficient to execute said query over said set of hardware-level representations than to execute said query over said set of application-level representations;
wherein the method is performed by one or more computing devices.