IP Library Granted Patent US 12,554,585
Granted Patent B2
US 12,554,585 · App. 18/673,134 · Granted Feb 17, 2026

Proactive input-output failover based on predicting optical transceiver module hardware failures in host bus adapter

Inventors: Vinayak Karne (Hyderabad, IN); Yadagiri Rajaboina (Hyderabad, IN); Phani Kumar V. U. Ayyagari (Hyderabad, IN); Venkateshwar R. Yerravalli (Hyderabad, IN); Samyuktha Pantham (Rajanna Sircilla, IN)
Assignee: International Business Machines Corporation
G06F11/1423G06F11/0772G06F11/2028
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,554,585
App. No.
18/673,134
Granted
Feb 17, 2026
Kind
B2
Abstract

Method and apparatus for performing proactive input-output (IO) failover based on predicting optical transceiver module hardware failures in a host bus adapter (HBA) are described. An example method includes registering with an HBA to receive a notification that a first path via the HBA is predicted to fail. The notification is based on a set of parameters of an optical transceiver module coupled to the first path. A set of parameters of the optical transceiver module are monitored in accordance with the registration. At least one of the set of parameter is determined to satisfy a predetermined criteria. Responsive to the determination, the notification that the first path via the HBA is predicted to fail is sent, a second path via the HBA for issuing an IO operation is determined, and the IO operation is issued via the second path.

Claims (55)

1 . A computer-implemented method comprising:

registering with a host bus adapter (HBA) to receive a notification that a first path via the HBA is predicted to fail, wherein:

registering with the HBA comprises generating and transmitting an information element to subscribe for the notification that the first path via the HBA is predicted to fail;

the notification is based on a set of parameters of an optical transceiver module coupled to the first path; and

the information element comprises (i) the set of parameters and (ii) for each parameter of the set of parameters, a respective threshold value for the parameter that is associated with a likelihood that the optical transceiver module will fail;

monitoring the set of parameters of the optical transceiver module in accordance with the registration; and

upon determining, based on the monitoring, that at least one parameter of the set of parameters satisfies a predetermined criteria:

sending the notification that the first path via the HBA is predicted to fail;

determining a second path via the HBA for issuing an input-output (IO) operation, the second path being different from the first path; and

issuing the IO operation via the second path.

2 . The computer-implemented method of claim 1 , wherein determining that the at least one parameter of the set of parameters satisfies the predetermined criteria comprises determining that the at least one parameter has exceeded the respective threshold value for the at least one parameter.

3 . The computer-implemented method of claim 1 , wherein the notification is sent to a host device communicatively coupled to the HBA.

4 . The computer-implemented method of claim 1 , wherein the notification is sent to a plurality of virtual machines (VMs), each of the plurality of VMs being mapped to the first path via the HBA.

5 . The computer-implemented method of claim 1 , wherein the IO operation is issued via the second path prior to a failure of the first path.

6 . The computer-implemented method of claim 1 , wherein:

the first path comprises a first physical port of the HBA; and

the second path comprises a second physical port of the HBA.

7 . The computer-implemented method of claim 1 , wherein the set of parameters comprises at least one of a temperature, a voltage, a transmit bias, a transmit power, or a receive power.

8 . A computing device comprising:

one or more memories collectively storing executable instructions; and

one or more processors communicatively coupled to the one or more memories, the one or more processors being collectively configured to execute the executable instructions and cause the computing device to perform an operation comprising:

registering with a host bus adapter (HBA) to receive a notification that a first path via the HBA is predicted to fail, wherein;

registering with the HBA comprises generating and transmitting an information element to subscribe for the notification that the first path via the HBA is predicted to fail;

the notification is based on a set of parameters of an optical transceiver module coupled to the first path; and

the information element comprises (i) the set of parameters and (ii) for each parameter of the set of parameters, a respective threshold value for the parameter that is associated with a likelihood that the optical transceiver module will fail;

monitoring the set of parameters of the optical transceiver module in accordance with the registration; and

upon determining, based on the monitoring, that at least one parameter of the set of parameters satisfies a predetermined criteria:

sending the notification that the first path via the HBA is predicted to fail;

determining a second path via the HBA for issuing an input-output (IO) operation, the second path being different from the first path; and

issuing the IO operation via the second path.

9 . The computing device of claim 8 , wherein determining that the at least one parameter of the set of parameters satisfies the predetermined criteria comprises determining that the at least one parameter has exceeded the respective threshold value for the at least one parameter.

10 . The computing device of claim 8 , wherein the notification is sent to a host device communicatively coupled to the HBA.

11 . The computing device of claim 8 , wherein the notification is sent to a plurality of virtual machines (VMs), each of the plurality of VMs being mapped to the first path via the HBA.

12 . The computing device of claim 8 , wherein the IO operation is issued via the second path prior to a failure of the first path.

13 . The computing device of claim 8 , wherein:

the first path comprises a first physical port of the HBA; and

the second path comprises a second physical port of the HBA.

14 . A computer-readable storage medium having computer-readable program code embodied therewith, the computer-readable program code collectively executable by one or more computer processors to perform an operation comprising:

registering with a host bus adapter (HBA) to receive a notification that a first path via the HBA is predicted to fail, wherein:

registering with the HBA comprises generating and transmitting an information element to subscribe for the notification that the first path via the HBA is predicted to fail;

the notification is based on a set of parameters of an optical transceiver module coupled to the first path; and

the information element comprises (i) the set of parameters and (ii) for each parameter of the set of parameters, a respective threshold value for the parameter that is associated with a likelihood that the optical transceiver module will fail;

monitoring the set of parameters of the optical transceiver module in accordance with the registration; and

upon determining, based on the monitoring, that at least one parameter of the set of parameters satisfies a predetermined criteria:

sending the notification that the first path via the HBA is predicted to fail;

determining a second path via the HBA for issuing an input-output (IO) operation, the second path being different from the first path; and

issuing the IO operation via the second path.

15 . The computer-readable storage medium of claim 14 , wherein determining that the at least one parameter of the set of parameters satisfies the predetermined criteria comprises determining that the at least one parameter has exceeded the respective threshold value for the at least one parameter.

16 . The computer-readable storage medium of claim 14 , wherein the notification is sent to a host device communicatively coupled to the HBA.

17 . The computer-readable storage medium of claim 14 , wherein the notification is sent to a plurality of virtual machines (VMs), each of the plurality of VMs being mapped to the first path via the HBA.

18 . The computer-readable storage medium of claim 14 , wherein the IO operation is issued via the second path prior to a failure of the first path.

19 . The computer-readable storage medium of claim 14 , wherein:

the first path comprises a first physical port of the HBA; and

the second path comprises a second physical port of the HBA.

20 . The computer-readable storage medium of claim 14 , wherein the set of parameters comprises at least one of a temperature, a voltage, a transmit bias, a transmit power, or a receive power.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded May 28, 2024
From: KARNE, VINAYAK; RAJABOINA, YADAGIRI; AYYAGARI, PHANI KUMAR V. U.; YERRAVALLI, VENKATESHWAR R; PANTHAM, SAMYUKTHA
To: INTERNATIONAL BUSINESS MACHINES CORPORATION
Reel/Frame 067535/0291 →
Continuity (1)
Related Publication 20250363007A1 · Nov 27, 2025
References Cited (14)
US 7783779B1 · Scales · 2010 [cited by examiner]
US 7930583B1 · Vemuri · 2011 [cited by examiner]
US 9747180B1 · Fair · 2017 [cited by applicant]
US 10396897B1 · Malave et al. · 2019 [cited by applicant]
US 11500678B2 · Sumangala et al. · 2022 [cited by applicant]
US 11720369B2 · Kanjirathinkal et al. · 2023 [cited by applicant]
US 20120039331A1 · Astigarraga · 2012 [cited by examiner]
US 20180189152A1 · B S · 2018 [cited by examiner]
US 20230239073A1 · Bharadwaj · 2023 [cited by examiner]
Broadcom product brief, “Emulex SAN manager”, Mar. 2025 (Year: 2025). [cited by examiner]
Authors et al., Method and System to Detect Defective SFP Transceiver in a SAN Environment, ip.com, Jan. 31, 2012. [cited by applicant]
Authors et al., A Method for Reporting Port Status Health Checks Using Multiple Fibre Channel Ports on the Same Physical Machine, ip.com, May 24, 2021. [cited by applicant]
SNIA, Management Interface for SFP+, Mar. 31, 2021. [cited by applicant]
SFF Specifications, Retrieved from: https://www.snia.org/technology-communities/sff/specifications, Jun. 15, 2018, 6 pages. [cited by applicant]