thing is the

the problem with this is that there are people who dont know what the purpose of thier startups is

• fdsf

• dsafdasf
  dsafdsafds
  fsdafdsa
  fasdfasdf
  asfdsafdsa

Describing performance

What happens when the load increases:

• How is the performance affected?

• How much do you need to increase your resources?

In a batch processing system such as Hadoop, we usually care about throughput, or the number of records we can process per second.

Latency and response time

The response time is what the client sees. Latency is the duration that a request is waiting to be handled.

It's common to see the average response time of a service reported. However, the mean is not very good metric if you want to know your "typical" response time, it does not tell you how many users actually experienced that delay.

Better to use percentiles.

• Median (50th percentile or p50). Half of user requests are served in less than the median response time, and the other half take longer than the median

• Percentiles 95th, 99th and 99.9th (p95, p99 and p999) are good to figure out how bad your outliners are.

Amazon describes response time requirements for internal services in terms of the 99.9th percentile because the customers with the slowest requests are often those who have the most data. The most valuable customers.

On the other hand, optimising for the 99.99th percentile would be too expensive.

Service level objectives (SLOs) and service level agreements (SLAs) are contracts that define the expected performance and availability of a servic

yeah so this is the thing

fdsafsdafasdf

this is great but hte problem with this shit is that the people are not aware of the underlying issues