Granted Patent B2

US 12,259,885 · App. 18/492,456 · Granted Mar 25, 2025

Inventors: Yang Yang (Hangzhou, CN); Qunshan Huang (Hangzhou, CN); Jun Qi (Hangzhou, CN); Shunde Cao (Hangzhou, CN); Pu Duan (Hangzhou, CN); Jian Du (Hangzhou, CN); Qingkai Mao (Hangzhou, CN); Yang Zhao (Hangzhou, CN); Kefeng Yu (Zhejiang, CN); Lei Wang (Hangzhou, CN); Benyu Zhang (Hangzhou, CN)

Assignee: Alipay (Hangzhou) Information Technology Co., Ltd.

G06F16/24545H04L9/06H04L2209/46

Abstract

Claims (67)

1. A computer-implemented method, comprising:

receiving, by a central device of a multi-party database system, a current query associated with a plurality of target databases of the multi-party database system;

generating, by the central device, a plurality of execution plans for the current query;

determining, by the central device for each execution plan, a respective cost computation formula of a plurality of cost computation formulas for computing an execution cost of jointly executing the execution plan by the plurality of target databases;

receiving, by the central device, a secure computation result from each of a plurality of query engines corresponding to the plurality of target databases, wherein the secure computation result is obtained by performing secure multi-party computation (MPC) based on a target secure computation method corresponding to the respective cost computation formula; and

determining, by the central device, an optimal execution plan having a lowest cost value in the plurality of cost computation formulas based on a cryptographic result of a cost value of each of the plurality of cost computation formulas.

2. The computer-implemented method according to claim 1 , wherein the execution cost comprises computing resource costs of the plurality of target databases and costs of communications between the plurality of target databases.

3. The computer-implemented method according to claim 1 , further comprising:

determining, by the central device, the target secure computation method based on the plurality of cost computation formulas; and

sending, by the central device, the target secure computation method to the plurality of query engines.

4. The computer-implemented method according to claim 1 , wherein the secure computation result is an index number of a cost computation formula with the lowest cost value in the plurality of cost computation formulas; and wherein

determining the optimal execution plan having the lowest cost value comprises:

determining an execution plan corresponding to the index number as the optimal execution plan.

5. The computer-implemented method according to claim 1 , wherein the secure computation result is

the cryptographic result of each of the plurality of cost computation formulas.

6. The computer-implemented method according to claim 5 , wherein the cryptographic result is a cost value shard computed based on secret sharing; and wherein

determining the optimal execution plan comprises:

aggregating a plurality of cost value shards sent by the plurality of query engines for a same execution plan of the plurality of execution plans to obtain a cost value of the same execution plan; and

determining the optimal execution plan by comparing cost values of the plurality of execution plans.

7. The computer-implemented method according to claim 5 , wherein the cryptographic result is a ciphertext cost value encrypted based on a public key of the central device; and wherein

determining the optimal execution plan comprises:

decrypting the ciphertext cost value based on a private key corresponding to the public key, to obtain a plaintext cost value; and

determining the optimal execution plan by comparing plaintext cost values of the plurality of execution plans.

8. A non-transitory, computer-readable medium storing one or more instructions executable by a computer system to perform operations comprising:

receiving, by a central device of a multi-party database system, a current query associated with a plurality of target databases of the multi-party database system;

generating, by the central device, a plurality of execution plans for the current query;

determining, by the central device for each execution plan, a respective cost computation formula of a plurality of cost computation formulas for computing an execution cost of jointly executing the execution plan by the plurality of target databases;

receiving, by the central device, a secure computation result from each of a plurality of query engines corresponding to the plurality of target databases, wherein the secure computation result is obtained by performing secure multi-party computation (MPC) based on a target secure computation method corresponding to the respective cost computation formula; and

determining, by the central device, an optimal execution plan having a lowest cost value in the plurality of cost computation formulas based on a cryptographic result of a cost value of each of the plurality of cost computation formulas.

9. The non-transitory, computer-readable medium according to claim 8 , wherein the execution cost comprises computing resource costs of the plurality of target databases and costs of communications between the plurality of target databases.

10. The non-transitory, computer-readable medium according to claim 8 , the operations further comprising:

determining, by the central device, the target secure computation method based on the plurality of cost computation formulas; and

sending, by the central device, the target secure computation method to the plurality of query engines.

11. The non-transitory, computer-readable medium according to claim 8 , wherein the secure computation result is an index number of a cost computation formula with the lowest cost value in the plurality of cost computation formulas; and wherein

determining the optimal execution plan having the lowest cost value comprises:

determining an execution plan corresponding to the index number as the optimal execution plan.

12. The non-transitory, computer-readable medium according to claim 8 , wherein the secure computation result is

the cryptographic result of each of the plurality of cost computation formulas.

13. The non-transitory, computer-readable medium according to claim 12 , wherein the cryptographic result is a cost value shard computed based on secret sharing; and wherein

determining the optimal execution plan comprises:

aggregating a plurality of cost value shards sent by the plurality of query engines for a same execution plan of the plurality of execution plans to obtain a cost value of the same execution plan; and

determining the optimal execution plan by comparing cost values of the plurality of execution plans.

14. The non-transitory, computer-readable medium according to claim 12 , wherein the cryptographic result is a ciphertext cost value encrypted based on a public key of the central device; and wherein

determining the optimal execution plan comprises:

decrypting the ciphertext cost value based on a private key corresponding to the public key, to obtain a plaintext cost value; and

determining the optimal execution plan by comparing plaintext cost values of the plurality of execution plans.

15. A central device of a multi-party database system, comprising:

one or more processors; and

one or more computer memory devices interoperably coupled with the one or more processors and storing programming instructions for execution by the one or more processors to perform operations comprising:

receiving a current query associated with a plurality of target databases of the multi-party database system;

generating a plurality of execution plans for the current query;

determining, for each execution plan, a respective cost computation formula of a plurality of cost computation formulas for computing an execution cost of jointly executing the execution plan by the plurality of target databases;

receiving a secure computation result from each of a plurality of query engines corresponding to the plurality of target databases, wherein the secure computation result is obtained by performing secure multi-party computation (MPC) based on a target secure computation method corresponding to the respective cost computation formula; and

determining an optimal execution plan having a lowest cost value in the plurality of cost computation formulas based on a cryptographic result of a cost value of each of the plurality of cost computation formulas.

16. The central device according to claim 15 , wherein the execution cost comprises computing resource costs of the plurality of target databases and costs of communications between the plurality of target databases.

17. The central device according to claim 15 , the operations further comprising:

determining the target secure computation method based on the plurality of cost computation formulas; and

sending the target secure computation method to the plurality of query engines.

18. The central device according to claim 15 , wherein the secure computation result is an index number of a cost computation formula with the lowest cost value in the plurality of cost computation formulas; and wherein

determining the optimal execution plan having the lowest cost value comprises:

determining an execution plan corresponding to the index number as the optimal execution plan.

19. The central device according to claim 15 , wherein the secure computation result is

the cryptographic result of each of the plurality of cost computation formulas.

20. The central device according to claim 19 , wherein the cryptographic result is a cost value shard computed based on secret sharing; and wherein

determining the optimal execution plan comprises:

aggregating a plurality of cost value shards sent by the plurality of query engines for a same execution plan of the plurality of execution plans to obtain a cost value of the same execution plan; and

determining the optimal execution plan by comparing cost values of the plurality of execution plans.

Assignments (6)

Priority Claims (1)

Continuity (2)

References Cited (14)