Systems and methods for performance monitoring

Patent metadata

Field	Value
Publication number	US-9983968-B2
Application number	US-201514749395-A
Country	US
Kind code	B2
Filing date	Jun 24, 2015
Priority date	Jun 24, 2015
Publication date	May 29, 2018
Grant date	May 29, 2018

Abstract

Techniques for obtaining computer system performance monitoring are described. One approach provides a performance monitoring system that is configured to obtain performance-related data from multiple monitored computing systems. The obtained data may be stored by the performance monitoring system in substantially real-time across multiple distinct data stores. The assignment of performance related data items (e.g., counter values) to storage locations may be determined in a various ways, including by random or pseudo-random assignment. Also, data may be stored in a tiered manner, where upper-level tiers aggregate data stored in lower-level tiers.

First claim

The embodiments of the invention in which an exclusive property or privilege is claimed are defined as follows:

1. A system for monitoring performance, the system comprising:
multiple data stores that are each configured to store a series of data items that reflect performance of a monitored computing system; and
a monitoring computing system that includes a processor and a memory that includes instructions that are configured, when executed by the processor, to monitor performance of the monitored computing system, by:
storing a first series of data items in a first randomly selected one of the multiple data stores, wherein the stored data items reflect performance of the monitored computing system during a first time interval;
storing a second series of data items in a second randomly selected one of the multiple data stores, wherein the stored data items reflect performance of the monitored computing system during a second time interval, wherein the first time interval precedes the second time interval; and
transmitting information about the stored data items; and

a monitoring agent that executes on the monitored computing system and that is configured to:
obtain counter values from a monitored application executing on the monitored computing system, wherein the counter values are provided by counter objects that are modified by instrumentation code in the monitored application and that represent one or more of: a number of operations performed by the monitored application, a response time of the monitored application, a processing time of the monitored application, or a throughput of the monitored application; and
transmit the counter values as the first series of data items to the monitoring computing system for storage on the first randomly selected data store.

2. The system of claim 1 , wherein the monitoring computing system is configured to:
aggregate the stored first and second series of data items; and
store the aggregated data in a third randomly selected one of the multiple data stores, wherein the stored aggregated data reflects performance of the monitored computing system during a third time interval that encompasses the first and second time intervals.

3. The system of claim 1 , wherein the monitoring computing system is configured to: monitor performance with respect to multiple distinct measurement tiers including a top tier and a bottom tier that each store data with respect to a corresponding sequence of time intervals of uniform size, wherein multiple time intervals of each tier that is not the top tier are encompassed by a single time interval of an immediate higher tier, and wherein each tier that is not the bottom tier aggregates data items stored in its immediate lower tier.

4. The system of claim 3 , wherein the monitoring computing system is further configured to:
monitor performance at a first measurement tier that stores data items that reflect performance of the monitored computing system during a first sample period of N seconds; and
monitor performance with respect to a second measurement tier that is above the first tier and that stores data items that reflect performance of the monitored computing system during a second sample period of M seconds, wherein each data item of the second measurement tier aggregates M/N data items obtained from the first measurement tier.

5. The system of claim 4 , wherein the first sample period is less than one minute and second sample period is more than one minute and less than one hour.

6. The system of claim 1 , wherein the monitoring computing system is configured to: for each of multiple successive time periods, randomly select one of the multiple data stores for storage of data items associated with that time period.

7. The system of claim 1 , wherein the counter objects include:
a first and a second data slot that are alternately used to store counter data during subsequent time intervals;
wherein, during a first time interval, the instrumentation code updates the first data slot while data from the second data slot is provided to the monitoring agent; and
wherein, during a second time interval immediately subsequent to the first time interval, the instrumentation code updates the second data slot while data from the first data slot is provided to the monitoring agent.

8. The system of claim 1 , wherein the monitored computing system is a digital transaction service comprising multiple distinct computing systems configured to facilitate digital transactions including associating electronic signatures with documents.

9. A method in a performance monitoring computing system, the method comprising:
providing multiple data stores that are each configured to store a series of data items that reflect performance of a monitored computing system;
storing a first series of data items in a first randomly selected one of the multiple data stores, wherein the stored data items reflect performance of the monitored computing system during a first time interval;
storing a second series of data items in a second randomly selected one of the multiple data stores, wherein the stored data items reflect performance of the monitored computing system during a second time interval, wherein the first time interval precedes the second time interval; and
transmitting information about the stored data items;
wherein the first randomly selected one of the multiple data stores and the second randomly selected one of the multiple data stores randomly distributes data items across the multiple data stores, such that on average 1/N of the data items are stored on each data store of the multiple data stores, wherein N equals the total number of data stores; and
in response to failure of one of the data stores, providing access to N−1/N of the stored data items.

10. The method of claim 9 , further comprising:
aggregating the stored first and second series of data items; and
storing the aggregated data in a third randomly selected one of the multiple data stores, wherein the stored aggregated data reflects performance of the monitored computing system during a third time interval that encompasses the first and second time intervals.

11. The method of claim 9 , further comprising:
monitoring performance with respect to multiple distinct measurement tiers that each store data with respect to a corresponding sequence of time intervals of uniform size, wherein multiple time intervals of each tier are encompassed by a single time interval of an immediate higher tier;
monitoring performance at a first measurement tier that stores data items that reflect performance of the monitored computing system during a sample period of N seconds; and
monitoring performance with respect to a second measurement tier that stores data items that reflect performance of the monitored computing system during a sample period of M seconds, wherein each data item of the second measurement tier aggregates M/N data items obtained from the first measurement tier.

12. The method of claim 9 , further comprising:
receiving counter values from a monitoring agent executing on the monitored computing system, the counter values including samples of counters that track the performance of an application executing on the monitored computing system;
storing a first portion of the received counter values as the first series of data items on the first randomly selected data store, the first portion of the received counter values reflecting performance of the monitored computing system during the first time interval; and
storing a second portion of the received counter values as the second series of data items on the second randomly selected data sto

Assignees

• Docusign Inc

Inventors

• Gwozdz Kazimierz Krzysztof

Classifications

• G06F11/3409

  for performance assessment · CPC title

• G06F11/3476

  Data logging (G06F11/14, G06F11/2205 take precedence) · CPC title

• G06F2201/865

  Monitoring of software · CPC title

• G06F11/3082

  the data filtering being achieved by aggregating or compressing the monitored data · CPC title

• G06F2201/835

  Timestamp · CPC title