2. • Interested in Open Source technologies
• Interested in Performance Engineering, MongoDB,
DevOps & DevOpSec Practices
• Tech Speaker/Blogger
• CTO, Mydbops IT Solution
Manosh Malai
About Me
7. Goals Of Performance Monitoring
▪ Utilization: How much of a resource is engaged in percentage overtime
▪ Cost: The price/Performance ratio
▪ Throughput: Typicallly data movement in bits or bytes per second
▪ Rate: A transaction/request per second
▪ Latency: Runtime of a query or operation, typically measured in milliseconds
9. Anti Method
Street Light Anti-Method
Drunk Man Anti-Method
Blame Someone Else Anti-Method
Pick some familiar observability tools and look for obvious issues
Tune things at random until the problem goes away
Point to other team then sit and relax
11. Workload Characterization
Who is causing the load(PID, Process Name, UID, IP address)
Why is th load Called?(Trace in code)
How the load changing over time
What is the load(IOPS, Throughput)
12. Drill-Down Analysis
Is a process of breaking problems
Into its basic elements
To find the root cause(s)