Metrics in distributed computing
According to an example aspect of the present invention, there is provided apparatus comprising at least one processing core, at least one memory including computer program code, the at least one memory and the computer program code being configured to, with the at least one processing core, cause the apparatus at least to store group definition information, receive plural reports from a set of distributed-computing entities each report comprising performance information and a transaction identifier, form a group of at least a proper subset of the plural reports based on their respective transaction identifiers and the group definition information, and obtain an aggregate performance report using both the performance information in each one of the reports in the group, and a performance metric of the group.
1 . An apparatus comprising at least one processing core, at least one memory including computer program code, the at least one memory and the computer program code being configured to, with the at least one processing core, cause the apparatus at least to:
store group definition information;
receive plural reports from a set of distributed-computing entities each report comprising performance information and a transaction identifier;
form a group of at least a proper subset of the plural reports based on their respective transaction identifiers and the group definition information, wherein the transaction identifiers in the group are distinct from each other, and
obtain an aggregate performance report using both the performance information in each one of the reports in the group, and a performance metric of the group.
2 . The apparatus according to claim 1 , wherein the at least one memory and the computer program code are configured to, with the at least one processing core, further cause the apparatus to determine at least one performance report of a specific one of the distributed-computing entities based on a specific report received in the apparatus from the set of distributed-computing entities.
3 . The apparatus according to claim 1 , wherein the group of reports and the aggregate performance report comprise information on a distributed execution of a single service or function invocation.
4 . The apparatus according to claim 1 , wherein the reports in the group each comprise the same transaction identifier.
5 . The apparatus according to claim 1 , wherein the reports in the group each comprise, as the transaction identifier, a pointer to a memory location.
6 . The apparatus according to claim 1 , wherein the apparatus is configured to form the group by determining the distinct transaction identifiers are comprised in a same identifier group.
7 . The apparatus according to claim 6 , wherein the apparatus is configured to perform the determining that the distinct transaction identifiers are comprised in a same identifier group by using a filter comprised in the group definition information.
8 . The apparatus according to claim 1 , wherein the performance information in each report in the group comprises an indication as to at least one of the following: a time delay, at least one timestamp, a processor cycle count, a processor load status of the distributed-computing entity transmitting the respective report, an indication of fault status, a networking resource use, a disk resource use, memory use, and an indication of an absence of a fault status.
9 . The apparatus according to claim 1 , wherein the performance metric indicates that the performance information in each report in the group are to be joined together in an additive manner to obtain the aggregate performance report.
10 . The apparatus according to claim 1 , wherein at least one of the reports in the group comprises performance information concerning more than one of the distributed-computing entities.
11 . The apparatus according to claim 1 , wherein at least one of the reports in the group comprises performance information concerning one and only one of the distributed-computing entities.
12 . The apparatus according to claim 1 , wherein the distributed-computing entities of the set of distributed-computing entities are physically distinct hardware computation nodes.
13 . A method, comprising:
storing group definition information;
receiving plural reports from a set of distributed-computing entities each report comprising performance information and a transaction identifier;
forming a group of at least a proper subset of the plural reports based on their respective transaction identifiers and the group definition information, wherein the transaction identifiers in the group are distinct from each other, and
obtaining an aggregate performance report using both the performance information in each one of the reports in the group, and a performance metric of the group.
14 . A non-transitory computer readable medium having stored thereon a set of computer readable instructions that, when executed by at least one processor, cause an apparatus to at least:
store group definition information;
receive plural reports from a set of distributed-computing entities each report comprising performance information and a transaction identifier;
form a group of at least a proper subset of the plural reports based on their respective transaction identifiers and the group definition information, wherein the transaction identifiers in the group are distinct from each other, and
obtain an aggregate performance report using both the performance information in each one of the reports in the group, and a performance metric of the group.