CODEWITHSUNDEEP
httperf
user2277149
user2277149

Reputation: 65

Httperf wsesslog Understanding

I have file by the name sessions.txt which has contents: 

url1

url2

url3

That means in the file I have 3 sessions with burst of length 1

The Httperf command I ran is

httperf --server=localhost --port=8080 --wsesslog=50,1,uri.txt --rate=300

In the below result I am not able to understand whether this response time is for all the sessions or for each individual request. Also I have mentioned rate=300 , but in the result, Connection rate still says=1.6 conns/s

Total: connections 50 requests 50 replies 50 test-duration 30.739 s

Connection rate: 1.6 conn/s (614.8 ms/conn, <=50 concurrent connections)
Connection time [ms]: min 16647.5 avg 22309.2 max 30598.5 median 21520.5 stddev 3859.7
Connection time [ms]: connect 711.9
Connection length [replies/conn]: 1.000

Request rate: 1.6 req/s (614.8 ms/req)
Request size [B]: 462.0

Reply rate [replies/s]: min 0.0 avg 1.5 max 6.6 stddev 2.6 (6 samples)
Reply time [ms]: response 21597.1 transfer 0.1
Reply size [B]: header 143.0 content 661.0 footer 2.0 (total 806.0)
Reply status: 1xx=0 2xx=50 3xx=0 4xx=0 5xx=0

CPU time [s]: user 14.31 system 11.90 (user 46.6% system 38.7% total 85.3%)
Net I/O: 2.0 KB/s (0.0*10^6 bps)

Errors: total 0 client-timo 0 socket-timo 0 connrefused 0 connreset 0
Errors: fd-unavail 0 addrunavail 0 ftab-full 0 other 0

Session rate [sess/s]: min 0.00 avg 1.63 max 6.60 stddev 2.58 (50/50)
Session: avg 1.00 connections/session
Session lifetime [s]: 22.3
Session failtime [s]: 0.0
Session length histogram: 0 50

Upvotes: 1

Views: 2589

Answers (1)

Rabi
Rabi

Reputation: 111

The 'reply time' is average reply time for all request. In your case the URL/API latency is 21597.1 ms.

The 'rate' value indicates number of session per second. In you case it is 300 session per second, each session has one connection.

Note as per httperf documentation

"The concurrent calls are issued either as pipelined calls on an existing persistent connection or as individual calls on separate connections. Whether a persistent connection is used depends on whether the server responds to the first call with a reply that includes a ‘‘Connection: close’’ header line. If such a line is present, separate connections are used."

Upvotes: 1

Related Questions