Computing system and operation method thereof

Abstract

This invention proposes a computing system and its operating method. The computing system includes a master computer and slave computer devices. The slave computer devices are connected to a dedicated Internet Protocol (IP) address of the master computer. The master computer and slave computer devices each establish user accounts for passwordless connection authentication. The slave computer device establishes key information and provides its public key to the master computer, enabling the master computer to verify whether the slave computer has logged in without a password based on the slave computer's public key. After the slave computer logs in without a password, the master computer and slave computer devices perform distributed computing to train a large-scale model.

Claims

1. A computing system, comprising: One main computer device; The system includes a slave computer connected to a master computer via a dedicated Internet Protocol (IP) address. Both the master and slave computers establish user accounts for passwordless connection authentication. The slave computer creates key information and provides a public key of this key information to the master computer, allowing the master computer to verify passwordless login using the slave's public key. Each computer generates a temporary session key pair. After passwordless login, the master and slave computers perform distributed computation to train a large-scale model. They exchange keys and determine a session identifier. The slave computer signs the session identifier and sends an authentication request to the master computer. The master computer verifies the session identifier's validity based on the authentication request. When the session identifier is correct, the master computer device notifies the slave computer device that the identity verification is successful, and the slave computer device provides its public key to the master computer device so that the master computer device can send back the slave computer device's public key, a host public key of the master computer device, and a hash signature to the slave computer device.

2. The computing system as claimed in claim 1, wherein the slave computer device authenticates the master computer device based on the host public key of the master computer device, wherein the slave computer device obtains a shared secret based on the host public key and obtains a hash based on the hash signature, wherein the slave computer device generates a symmetric key and a session identifier for a data channel between the master computer device and the slave computer device based on the shared secret and the hash.

3. The computing system as described in claim 2, wherein the slave computer device signs the session identifier according to a private key.

4. The computing system as described in claim 2, wherein the slave computer device pre-copies the public key of the key data to an authorized public key list of the master computer device, so that the master computer device can determine whether to allow identity verification based on the authorized public key list.

5. The computing system as described in claim 4, wherein the master computer device verifies the session identifier based on the public key of the slave computer device to establish the data channel for training the large model.

6. The computing system as claimed in claim 1, wherein the slave computer device is connected to the host computer device via a Thunderbolt cable.

7. A method of operating a computing system, wherein the computing system includes a master computer device and a slave computer device, wherein the method of operating the system includes: Through the dedicated Internet Protocol address that connects the computer device to the host computer device; On both the master computer and the slave computer, a user account for passwordless connection authentication is established. A key database is created on the slave computer, and a public key of this key database is provided to the master computer. The master computer verifies whether the slave computer has logged in without a password based on the public key of the slave computer, including: exchanging keys between the master and slave computers and determining a session identifier; providing the slave computer's public key to the master computer, so that the master computer returns the slave computer's public key, a host public key of the master computer, and a hash signature to the slave computer; signing the session identifier and sending an authentication request to the master computer; verifying the session identifier's correctness based on the authentication request; and notifying the slave computer of successful authentication when the session identifier is correct. The master computer device and the slave computer device each generate temporary session key pairs; and after the slave computer device logs into the master computer device without a password, distributed computing is performed through the master computer device and the slave computer device to train a large model.

8. The method of operation as described in claim 7, wherein the step of exchanging keys further includes: The slave computer device authenticates the identity of the master computer device using the host public key; the slave computer device obtains a shared secret based on the host public key and a hash based on the hash signature; and the slave computer device generates a symmetric key and a session identifier for a data channel between the master computer device and the slave computer device based on the shared secret and the hash.

9. The method of operation as described in claim 8, wherein the step of signing the session identifier includes: The session identifier is signed using a private key via the computer device.

10. The operating method as described in claim 8, further comprising: The public key of the key data is copied from the computer device to an authorized public key list on the main computer device in advance; And through the main computer device, it determines whether to allow identity verification based on the list of authorized public keys.

11. The method of operation as described in request item 10, wherein the step of verifying whether the session identifier is correct includes: The host computer device verifies the session identifier based on the public key of the slave computer device to establish the data channel used in the training of the large model.

Images