alibaba / clusterdata

cluster data collected from production clusters in Alibaba for cluster management research
1.54k stars 402 forks source link

For the dm and um in MSCallGraph, can not find their corresponding msname in MSResource in cluster-trace-microservices-v2021 #202

Closed THUzxj closed 6 months ago

THUzxj commented 6 months ago

Thank you for your efforts in releasing this data!

I encountered this problem when joining the tables. For the dm and um in MSCallGraph, can not find their corresponding msname in MSResource in cluster-trace-microservices-v2021. However, the README says that in MS_Metrics_Table, the msname is "The name of MS, to be joined with MSName in MS_MCR_RT_Table, and DM and UM in MS_CallGraph_Table. MSName only contains stateless services, as stateful services run in other dedicated clusters."

Thus I wonder whether the um and dm fields in MSCallGraph are the names of microservices.

How to reproduce it:

$ head -n2 MSCallGraph/MSCallGraph_0.csv 
,traceid,timestamp,rpcid,um,rpctype,dm,interface,rt
0,0b133c1915919238193454000e5d37,219678,0.1.3.1.1.1.12,5cca70246befb1f4c9546d2912b9419dee54439218efa55a7a2e0e26e86ad749,mc,b1dbd3a649a3cc790fa12573c9c1aa00988e07a8818a2214208b9697238c1b11,,0
$ cat MSResource/*.csv | grep 5cca70246befb1f4c9546d2912b9419dee54439218efa55a7a2e0e26e86ad749
# no output, but it is expected to show the lines with msname = '5cca...'
niewuya commented 6 months ago

Thank you for your question and interest. MSResource only involves the microservices in two clusters, as mentioned in the README, while MS_CallGraph_Table contains the microservices in more than two clusters. Therefore, it happens that some microservices only appear in MS_CallGraph_Table.

THUzxj commented 6 months ago

Thank you for your quick answer! I have confirmed that there are traces in MSCallGraph whose um or dm appears in MSResource.