IP Library Granted Patent US 11,922,226
Granted Patent B2
US 11,922,226 · App. 17/362,774 · Granted Mar 5, 2024

Adaptive control of deadline-constrained workload migrations

Inventor: Manish Gupta (New Delhi, IN)
Assignee: Kyndryl, Inc.
G06F9/5088G06F9/5038G06F9/505G06F11/0757
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 11,922,226
App. No.
17/362,774
Granted
Mar 5, 2024
Kind
B2
Abstract

Adaptive control of deadline-constrained workload migrations can include monitoring migrations of workloads forming a wave migrating from a source computing node to a target computing node. The monitoring can be performed in real time. The migrations can be performed by transferring image replications of each workload over a data communication network. Based on an expected bandwidth availability, a likelihood that a cutover deadline associated with the wave is exceeded prior to completing a migration of each of the wave's workloads can be predicted. Migration of one or more selected workloads can be suspended in response to determining that exceeding the cutover deadline prior to completing migration of each of the wave's workloads is likely.

Claims (31)

1. A method, comprising:

monitoring, with a computer, migrations of workloads forming a wave migrating from a source computing node to a target computing node, wherein the monitoring is performed in real time and wherein the migrations are performed by transferring image replications of each workload over a data communication network;

predicting, with the computer, based on an expected bandwidth availability, a likelihood that a cutover deadline associated with the wave is exceeded prior to completing a migration of each of the wave's workloads; and

suspending, with the computer, migration of one or more selected workloads of another wave in response to determining that exceeding the cutover deadline prior to completing migration of each of the wave's workloads is likely.

2. The method of claim 1 , further comprising reinitiating migration of the one or more workloads selected and suspending migration of one or more other workloads in response to determining that suspending migrations of the one or more selected workloads failed to improve the expected bandwidth availability to make completion of the migrations prior to the cutover deadline likely.

3. The method of claim 1 , further comprising selecting one or more pending workloads and initiating migrations of the one or more pending workloads selected, wherein the one or more pending workloads is selected in response to determining that completing the migration prior to the cutover deadline is likely.

4. The method of claim 1 , further comprising generating an alarm in response to determining that migration of each workload of the wave prior to the cutover deadline is not likely.

5. The method of claim 4 , further comprising performing a predetermined number of predictions over a given interval and generating the alarm in response to a predetermined portion of the predictions indicating a likelihood that the cutover deadline associated with the wave is exceeded prior to completing a migration of each of the wave's workloads.

6. The method of claim 1 , wherein the expected bandwidth availability is based on a comparison of a mean transfer rate with a transfer rate required to complete migration of the workloads of the wave prior to the cutover deadline.

7. The method of claim 6 , wherein the mean transfer rate is determined based on a predetermined number of weighted samples of past transfer rates wherein the samples are weighted to give greater effect to samples corresponding to more recent transfers of image replications over the data communication network.

8. A system, comprising:

a processor configured to initiate operations including:

monitoring migrations of workloads forming a wave migrating from a source computing node to a target computing node, wherein the monitoring is performed in real time and wherein the migrations are performed by transferring image replications of each workload over a data communication network;

predicting, based on an expected bandwidth availability, a likelihood that a cutover deadline associated with the wave is exceeded prior to completing a migration of each of the wave's workloads; and

suspending migration of one or more selected workloads of another wave in response to determining that exceeding the cutover deadline prior to completing migration of each of the wave's workloads is likely.

9. The system of claim 8 , wherein the processor is configured to initiate operations further including reinitiating migration of the one or more workloads selected and suspending migration of one or more other workloads in response to determining that suspending migrations of the one or more selected workloads failed to improve the expected bandwidth availability to make completion of the migrations prior to the cutover deadline likely.

10. The system of claim 8 , wherein the processor is configured to initiate operations further including selecting one or more pending workloads and initiating migrations of the one or more pending workload selected, wherein in the one or more pending workloads is selected in response to determining that completing the migration prior to the cutover deadline is likely.

11. The system of claim 8 , wherein the processor is configured to initiate operations further including generating an alarm in response to determining that migration of each workload of the wave prior to the cutover deadline is not likely.

12. The system of claim 11 , wherein the processor is configured to initiate operations further including performing a predetermined number of predictions over a given interval and generating the alarm in response to a predetermined portion of the predictions indicates a likelihood that the cutover deadline associated with the wave is exceeded prior to completing a migration of each of the wave's workloads.

13. The system of claim 8 , wherein the expected bandwidth availability is based on a comparison of a mean transfer rate with a transfer rate required to complete migration of the workloads of the wave prior to the cutover deadline.

14. A computer program product, the computer program product comprising:

one or more computer-readable storage media and program instructions collectively stored on the one or more computer-readable storage media, the program instructions executable by a processor to cause the processor to initiate operations including:

monitoring migrations of workloads forming a wave migrating from a source computing node to a target computing node, wherein the monitoring is performed in real time and wherein the migrations are performed by transferring image replications of each workload over a data communication network;

predicting, based on an expected bandwidth availability, a likelihood that a cutover deadline associated with the wave is exceeded prior to completing a migration of each of the wave's workloads; and

suspending migration of one or more selected workloads of another wave in response to determining that exceeding the cutover deadline prior to completing migration of each of the wave's workloads is likely.

15. The computer program product of claim 14 , wherein the program instructions are executable by the processor to cause the processor to initiate operations further including reinitiating migration of the one or more workloads selected and suspending migration of one or more other workloads in response to determining that suspending migrations of the one or more selected workloads failed to improve the expected bandwidth availability to make completion of the migrations prior to the cutover deadline likely.

16. The computer program product of claim 14 , wherein the program instructions are executable by the processor to cause the processor to initiate operations further including selecting one or more pending workloads and initiating migrations of the one or more pending workload selected, wherein in the one or more pending workloads is selected in response to determining that completing the migration prior to the cutover deadline is likely.

17. The computer program product of claim 14 , wherein the program instructions are executable by the processor to cause the processor to initiate operations further including generating an alarm in response to determining that migration of each workload of the wave prior to the cutover deadline is not likely.

18. The computer program product of claim 17 , wherein the program instructions are executable by the processor to cause the processor to initiate operations further including performing a predetermined number of predictions over a given interval and generating the alarm in response to a predetermined portion of the predictions indicates a likelihood that the cutover deadline associated with the wave is exceeded prior to completing a migration of each of the wave's workloads.

19. The computer program product of claim 14 , wherein the expected bandwidth availability is based on a comparison of a mean transfer rate with a transfer rate required to complete migration of the workloads of the wave prior to the cutover deadline.

20. The computer program product of claim 14 , wherein the mean transfer rate is determined based on a predetermined number of weighted samples of past transfer rates wherein the samples are weighted to give greater effect to samples corresponding to more recent transfers of image replications over the data communication network.

Assignments (2)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Nov 18, 2021
From: INTERNATIONAL BUSINESS MACHINES CORPORATION
To: KYNDRYL, INC.
Reel/Frame 058213/0912 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jun 29, 2021
From: GUPTA, MANISH
To: INTERNATIONAL BUSINESS MACHINES CORPORATION
Reel/Frame 056710/0547 →
Continuity (1)
Related Publication 20220413942A1 · Dec 29, 2022