Requesting storage performance models for a configuration pattern of storage resources to deploy at a client computing environment

Patent metadata

Field	Value
Publication number	US-9888078-B2
Application number	US-201514823720-A
Country	US
Kind code	B2
Filing date	Aug 11, 2015
Priority date	Jan 22, 2015
Publication date	Feb 6, 2018
Grant date	Feb 6, 2018

Abstract

Provided are a method, system and program for requesting storage performance models for a configuration pattern of storage resources to deploy at a client computing environment. A determination is made of a new configuration pattern of storage resources to deploy. A request is sent to a service provider with information on the new configuration pattern. A result set is received from the service provider having at least one provided configuration pattern having a degree of similarity to the new configuration pattern and a storage performance model for each of the provided configuration patterns. Each of the storage performance models indicate workload and performance characteristics for one of the provided configuration patterns. One of the provided configuration patterns is selected from the result set and the storage performance model for the selected configuration pattern is used to model performance at the client.

First claim

What is claimed is:

1. A method performed by a processor executing program code in a non-transitory computer readable storage medium for deploying storage resources at a client computing environment, comprising:
determining a new configuration pattern of storage resources to deploy in the client computing environment;
determining if a local catalog includes one or more published configuration patterns similar to the new configuration pattern;
selecting a published configuration pattern in the local catalog having a similar configuration pattern to the new configuration pattern and highest confidence level if there are multiple published configuration patterns in the local catalog having similar configuration patterns;
sending a request to a service provider with information of the new configuration pattern if there is no published configuration pattern in the local catalog having a similar configuration pattern;
receiving, from the service provider, a result set having at least one provided configuration pattern data structure having a degree of similarity to the new configuration pattern, wherein each of the at least one provided configuration pattern data structure includes a configuration pattern of storage resources including attributes values for storage attributes of the storage resources deployed at the client, a storage performance model for the configuration pattern of storage resources, and a confidence level indicating a degree of detail of an internal structure of the configuration pattern of storage resources considered in determining storage performance metrics, wherein the storage performance model indicates workload and performance characteristics for the configuration pattern of storage resources based on the storage performance metrics as a function of different workloads at the configuration pattern of storage resources that models how the configuration pattern of storage resources responds to different workloads, wherein the storage performance model is determined according to the storage performance metrics that test internal structures and workings of the configuration pattern of storage resources and operations and results of the configuration pattern of storage resources;
selecting a configuration pattern data structure from one of the at least one provided configuration pattern data structure from the result set in response to receiving the result set; and
using the storage performance model for the selected configuration pattern data structure from the result set or the selected published configuration pattern from the local catalog to model performance at the client computing environment of the new configuration pattern of storage resources.

2. The method of claim 1 , further comprising:
requesting a new software service from the service provider, wherein the determining the new configuration pattern comprises determining the new configuration pattern to support the requested new software service, wherein the using the storage performance model comprises applying the storage performance model to determine provisioning of the new configuration pattern of storage resources to be used for the requested new software service.

3. The method of claim 1 , further comprising:
subscribing, by the client computing environment, to a type of storage performance model, wherein the the at least one provided configuration pattern data structure of the received result set has the storage performance model of the subscribed type of storage performance model.

4. The method of claim 1 , wherein the storage performance model for each provided configuration pattern comprises at least one of an Input/Output (I/O) performance latency curve, estimated I/O peak, Input/Output, I/O response time as a function of additional I/O traffic, and storage utilization.

5. The method of claim 1 , wherein the information on the new configuration pattern sent to the service provider includes attributes of the storage resources including at least one of an enclosure type, disk type, encryption settings, compression settings, storage configuration, a Redundant Array of Independent Disk (RAID) configuration, and a Just a Bunch of Disk (JBOD) configuration.

6. The method of claim 1 , wherein the selected configuration pattern data structure from the result set has a highest confidence level for the at least one provided configuration pattern data structure from the result set.

Assignees

• IBM

Inventors

• Jain Rakesh
• Routray Ramani R
• Padbidri Sumant
• Song Yang

Classifications

• G06F3/0689

  Disk arrays, e.g. RAID, JBOD · CPC title

• G06F3/0611

  in relation to response time · CPC title

• H04L41/0806

  for initial configuration or provisioning, e.g. plug-and-play · CPC title

• H04L41/0826

  for reduction of network costs (H04L41/0833 takes precedence) · CPC title

• H04L67/1097 Primary

  for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS] · CPC title