Dynamic metering adjustment for service management of computing platform

Patent metadata

Field	Value
Publication number	US-10467036-B2
Application number	US-201514926384-A
Country	US
Kind code	B2
Filing date	Oct 29, 2015
Priority date	Sep 30, 2014
Publication date	Nov 5, 2019
Grant date	Nov 5, 2019

Abstract

Systems and methods are provided for dynamic metering adjustment for service management of a computing platform. For example, a plurality of virtual machines are provisioned across a plurality of computing nodes of a computing platform. Data samples are collected for a metric that is monitored with regard to resource utilization in the computing platform by the virtual machines. The data samples are initially collected at a predefined sampling frequency. The data samples collected over time for the metric are analyzed to determine an amount of deviation in values of the collected data samples. A new sampling frequency is determined for collecting data samples for the metric based on the determined amount of deviation. The new sampling frequency is applied to collect data samples for the metric, wherein the new sampling frequency is less than the predefined sampling frequency.

First claim

What is claimed is:

1. A method for managing a computing platform, comprising:
provisioning a plurality of virtual machines that execute on a plurality of computing nodes of a computing platform, wherein the provisioned virtual machines utilize computing resources of the computing nodes; and
executing a centralized service management system on at least one computing node of the computing platform to perform service management functions of the computing platform, wherein the service management functions performed by the centralized service management system comprise a process for dynamically adjusting metering operations for monitoring utilization of a computing resource of the plurality of computing nodes, which is utilized by the plurality of virtual machines executing on the plurality of computing nodes of the computing system, wherein the process for dynamically adjusting metering operations comprises:
monitoring the utilization of the computing resource by the provisioned virtual machines executing on the plurality of computing nodes of the computing platform, wherein monitoring comprises collecting data samples from the plurality of computing nodes, wherein the data samples comprise information regarding a metric of the utilization of the monitored resource by the provisioned virtual machines executing on the plurality of computing nodes, wherein the data samples are initially collected at a given sampling frequency, wherein each data sample comprises (i) a timestamp to mark a time that the data sample was collected, and (ii) a sample value of the metric of the utilization of the monitored resource;
storing the collected data samples in a persistent storage system;
analyzing a set of the data samples that are initially collected at the given sampling frequency and stored for the metric of the utilization of the monitored resource to determine an amount of deviation in the sample values of the data samples within the set of data samples that are initially collected at the given sampling frequency for the metric of the utilization of the monitored resource, wherein analyzing the set of data samples comprises (i) generating change point time series data by detecting changes in the sample values of the collected data samples associated with the metric of the utilization of the monitored resource and (ii) and converting the change point time series data into a sequence of symbols which encodes a change behavior of the metric of the utilization of the monitored resource;
determining a new sampling frequency for collecting new data samples for the metric of the utilization of the monitored resource based on the amount of deviation in the sample values of the data samples within the set of data samples for the metric of the utilization of the monitored resource as determined from the sequence of symbols which encodes the change behavior of the metric of the utilization of the monitored resource; and
applying the new sampling frequency for collecting new data samples for the metric of the utilization of the monitored resource by the provisioned virtual machines executing on the plurality of computing nodes of the computing platform;
wherein the new sampling frequency for collecting new data samples is less than the given sampling frequency when an encoded symbol for the metric indicates a period of invariable behavior of the metric of utilization of the monitored resource, to thereby reduce an amount of new data samples for the metric of the utilization of the monitored resource which are collected and stored in the persistent storage system;
wherein the method is implemented at least in part by a processor executing program code.

2. The method of claim 1 , further comprising assigning a metric policy to the metric of the utilization of the monitored resource based on values of one or more metric profile configuration items associated with the metric.

3. The method of claim 2 , wherein determining a new sampling frequency for collecting new data samples for the metric of the utilization of the monitored resource is further based on a metric policy assigned to the metric.

4. The method of claim 2 , wherein the metric policy for the metric of the utilization of the monitored resource comprises at least one of conservative sampling, conservative storage, aggregated storage, per tier sampling or a combination thereof, wherein the conservative sampling applies the predefined given sampling frequency for the metric, wherein conservative storage implies that all collected data samples for the metric are stored, wherein aggregated storage implies that an aggregate of the sample values of the collected data samples are stored, or that only changes in the sample values of the collected data samples are stored, and wherein pier tier sampling implies that different sampling frequencies are applied for different behaviors of the metric of the utilization of the monitored resource.

5. The method of claim 2 , wherein the one or more metric profile configuration items comprises one of an importance item, a usage item, a dependency item, or a combination thereof.

6. The method of claim 1 , wherein the monitored resource comprises one of CPU (central processing unit) usage, memory usage, TCP/IP connection rate, and page access per time.

7. The method of claim 1 , wherein determining the new sampling frequency for collecting new data samples for the metric of the utilization of the monitored resource comprises:
clustering the sequence of symbols into tiers of metrics with similar sequences;
determining a new sampling frequency for each of the tiers based on the sequences of symbols that are included within the tier.

8. The method of claim 1 , further comprising re-applying the given sampling frequency to collect new data samples for the metric of the utilization of the monitored resource when there is one of (i) a detected change in behavior of the metric and (ii) a metric policy update for the metric.

9. The method of claim 1 , further comprising:
aggregating the sample values of the collected data samples for the metric of the utilization of the monitored resource; and
storing the aggregated sample values of the collected data samples.

10. The method of claim 1 , further comprising:
determining if the metric of the utilization of the monitored resource is correlated to another metric;
if the metric of the utilization of the monitored resource is determined to be correlated to another metric, then comparing each newly collected data sample for the metric of the utilization of the monitored resource to a last collected and stored data sample for the metric of the utilization of the monitored resource; and
storing the newly collected data sample only if the sample value of the newly collected sample is different from the sample value of the last collected and stored data sample.

11. An article of manufacture comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a computer to cause the computer to perform a method for managing a computing platform, the method comprising:
provisioning a plurality of virtual machines that execute on a plurality of computing nodes of a computing platform, wherein the provisioned virtual machines utilize computing resources of the computing nodes; and
executing a centralized service management system on at least one computing node of the computing platform to perform service management functions of the computing platform, wherein the service management functions performed by the centralized service management system comprise a process for dynamically adjusting metering operations for monitoring utilization o

Assignees

• IBM

Inventors

• Anwar Ali
• Kochut Andrzej
• Sailer Anca
• Schulz Charles O
• Segal Alla

Classifications

• G06F2201/815

  Virtual · CPC title

• G06F11/3006

  where the computing system is distributed, e.g. networked systems, clusters, multiprocessor systems (multiprogramming arrangements G06F9/46; allocation of resources G06F9/50) · CPC title

• H04L43/024

  by adaptive sampling · CPC title

• G06F2009/45595

  Network integration; Enabling network access in virtual machine instances · CPC title

• H04L41/142

  using statistical or mathematical methods · CPC title