US2018260144A1PendingUtilityA1

I/o amplification for determining to increase workload

Assignee: IBMPriority: Mar 9, 2017Filed: Sep 25, 2017Published: Sep 13, 2018
Est. expiryMar 9, 2037(~10.6 yrs left)· nominal 20-yr term from priority
G06F 3/0613G06F 3/0604G06F 13/1642G06F 3/0619G06F 3/0676G06F 3/065G06F 3/0659G06F 3/0683G06F 16/957
51
PatentIndex Score
0
Cited by
0
References
0
Claims

Abstract

An amplification number may be input into a storage subsystem interface. A processor in a storage system may receive an original input/output (I/O) request from an application. The processor may determine, in response to the amplification number being input, to duplicate the original I/O request one or more times. The processor may generate one or more duplicate I/O requests of the original I/O request. The processor may store the original I/O request in a general address space in the storage subsystem. The processor may store the one or more duplicate I/O requests in a reserved address space in the storage subsystem. The processor may execute the original I/O request and the one or more duplicate requests.

Claims

exact text as granted — not AI-modified
1 . A computer-implemented method comprising:
 inputting, by a processor in a storage subsystem, an amplification number into a storage subsystem interface, wherein the processor inputs the amplification number using a random number generator;   receiving, by the processor in the storage subsystem, an original input/output (I/O) request from an application;   determining, in response to the amplification number being input, to duplicate the original I/O request one or more times;   generating one or more duplicate I/O requests of the original I/O request for simultaneous execution with the original I/O request, wherein the number of the one or more duplicate: I/O requests corresponds to the amplification number, and wherein generating the one or more duplicate I/O requests comprises identifying a block-size of the original I/O request and duplicating the block-size of the original I/O request identically to each of the one or more duplicate I/O requests;   determining that the original I/O request type is a write operation, wherein the write operation alters the behavior of the storage subsystem;   duplicating the write operation of the original I/O request to the one or more duplicate I/O requests;   identifying a storage address of the original I/O request, wherein the storage address directs the original I/O request to be stored in a general address space of the storage subsystem;   generating a tag, wherein the tag is a fixed offset value;   adding the tag to the storage address of the original I/O request, wherein adding the tag to the storage address of the original I/O request creates a new storage address;   storing the one or more duplicate I/O requests in the new storage address, wherein the new storage address directs the one or more duplicate I/O requests to be stored in a reserved address space of the storage subsystem;   storing the original I/O request in the general address space in the storage subsystem;   executing, simultaneously, the original I/O request in the general address space and all of the one or more duplicate I/O requests in the reserved address space;   altering the behavior of the reserved address space of the storage subsystem by executing the write operation, wherein altering the behavior of the reserved address space includes changing a display feature of the application;   hiding the altered behavior of the reserved address space of the storage subsystem from a user, wherein hiding the altered behavior of the reserved address space includes not displaying the changed display feature of the application to the user;   providing the user with an observational performance impact report, wherein the observational performance impact report includes the effect of the one or more duplicate I/O requests on the storage subsystem after execution of the one or more duplicate I/O requests;   determining, from the observational performance impact report, a maximum workload for the application, wherein determining the maximum workload for the application comprises the processor identifying that a load time of the executed one or more duplicate I/O requests is above a load time threshold; and   increasing, in response to determining the maximum workload for the application, the maximum workload for the application by automatically merging the general address space and the reserved address space together.

Join the waitlist — get patent alerts

Track US2018260144A1 — get alerts on status changes and closely related new filings.

We store only your email — no account needed. See our privacy policy.