Closed PADMAARUN closed 3 years ago
If you are referring to the ability to make X different client connections, and drive a smart contract from all of these connections, then yes. That is the "worker number" part of the benchmark. Be warned though, you are talking about a very large number of clients, for this I would recommend using the distributed worker approach so that you can scale the workers on the appropriate hardware. Launching 300 separate worker processes on a single machine is ill advised.
thanks for ur reply.....
blockchain{ type : fabric config :./fabric.json
command: ...
test: "clients"{ "type":local "number":5 } "rounds":
In the above, the parameter NUMBER represents what? number of client-users that we want to test or something else? is this i need to change, if i want to test with varying number of client-users?
thank u
number represents the number of worker processes that are spawned.
Each worker has the ability to submit transactions as any of the identities specified in the network configuration file
Each worker is indexed (it knows its own zero based ID)
You can specify which identity to use within the workload generation file by passing invokerIdentity
as part of the workload arguments. If you do not pick one, one will be selected for you.
It is possible (and common) to use multiple workers acting under the same identity during benchmarks
closing as the question has been answered. For anyone coming across this issue from an internet search, please feel free to ask questions in Rocket Chat https://chat.hyperledger.org/channel/caliper
i want to test the hyperledger fabric use case using hyperledger caliper benchmark.... i wnat to test the application varying the number of users. ie what is the performance of the application whaen there is 100 users or when there is 300 users... is it possible to do? if so, kindly help me how to do this....
Context
Expected Behavior
Actual Behavior
Possible Fix
Steps to Reproduce
1. 2. 3. 4.
Existing issues
Context
Your Environment