jradius performance
wlanmac
wlan at mac.com
Tue Dec 8 17:19:10 CET 2009
Hello,
The question of performance of jradius has been raised a couple times. I
thought I'd share some recent findings and improvements. Attached are
some radperf results of various setups and a brief discussion is below.
The tests are after changes we have made, which we will be integrating
into jradius/rlm_jradius soon. What we have done amounts to:
- buffers read/write to jradius socket in rlm_jradius and jradius-server
- object reuse of heavily used Java objects (radius packets, attributes,
etc - using Apache commons "pool" for this)
- using ByteBuffer instead of wrapped Streams throughout jradius
(resulting in less objects and method calls)
- and some more things...
It should also be noted that jradius per-default will maintain a RADIUS
"session" keyed on the combination of some attributes - analogous to the
"Acct-Unique-Session-Id" - and a thread-safe RadiusSession cached
object. For the testing, we disabled the session manager (and therefor
the session locking) since all the packets from radperf appear to be of
the same session, and requests are serialized around the session lock.
The tests are with the setup:
Thread pool: 128
thread pool {
start_servers = 128
max_servers = 128
min_spare_servers = 3
max_spare_servers = 128
max_requests_per_server = 0
}
JRadius config:
jradius {
name = "test"
primary = "localhost"
timeout = 2
keepalive = yes
connections = 128
onfail = NOOP
allow_codechange = yes
}
RadPerf performed using: radperf -x -s -f test.csv -T nas.rad -a pap -p
200 -c 20000 localhost auth <sec> -q
Summary of test:
FreeRADIUS 'users' file: ~ 17500 p/s
FreeRADIUS rlm_perl (example.pl): ~ 7600 p/s
FreeRADIUS rlm_jradius (LocalUsers): ~ 2500 p/s
Native Java jradius (CoovaRADIUS): ~ 12000 p/s
Still some room for improvement... but, we are getting there.
Valgrind reveals that FreeRADIUS is spending a lot of time in
rlm_jradius as it parses and creates/add attributes. This part could be
streamlined a lot as the attributes (in request and response) don't
always change yet rlm_jradius is freeing/creating new attributes
constantly. I suspect fixing that would greatly improve the rlm_jradius
performance.
Cheers,
David
-------------- next part --------------
======================================================================================
FreeRADIUS / users file (native)
(Only the 'files' modules in authorize, no 'perl', no 'jradius')
Total succeeded : 20000
Total failed : 0
Total no reply : 0
Total time (s) : 1.151
Packets/s : 17372
Response times:
< 10 usec : 0
< 100 usec : 0
< msec : 0
< 10 msec : 6260
< 0.1s : 13740
< s : 0
< 10s : 0
< 100s : 0
Total succeeded : 20000
Total failed : 0
Total no reply : 0
Total time (s) : 1.111
Packets/s : 17986
Response times:
< 10 usec : 0
< 100 usec : 0
< msec : 0
< 10 msec : 7339
< 0.1s : 12661
< s : 0
< 10s : 0
< 100s : 0
Total succeeded : 20000
Total failed : 0
Total no reply : 0
Total time (s) : 1.140
Packets/s : 17536
Response times:
< 10 usec : 0
< 100 usec : 0
< msec : 0
< 10 msec : 5631
< 0.1s : 14369
< s : 0
< 10s : 0
< 100s : 0
======================================================================================
FreeRADIUS / Simple rlm_perl handler
(Only 'files' and 'perl' in authorize, no 'jradius')
Total succeeded : 20000
Total failed : 0
Total no reply : 0
Total time (s) : 2.578
Packets/s : 7758
Response times:
< 10 usec : 0
< 100 usec : 0
< msec : 0
< 10 msec : 903
< 0.1s : 19075
< s : 22
< 10s : 0
< 100s : 0
Total succeeded : 20000
Total failed : 0
Total no reply : 0
Total time (s) : 2.622
Packets/s : 7625
Response times:
< 10 usec : 0
< 100 usec : 0
< msec : 0
< 10 msec : 693
< 0.1s : 19302
< s : 5
< 10s : 0
< 100s : 0
Total succeeded : 20000
Total failed : 0
Total no reply : 0
Total time (s) : 2.672
Packets/s : 7484
Response times:
< 10 usec : 0
< 100 usec : 0
< msec : 0
< 10 msec : 791
< 0.1s : 19157
< s : 52
< 10s : 0
< 100s : 0
======================================================================================
FreeRADIUS / rlm_jradius / LocalUsers
(Only 'files' and 'jradius' in authorize, no 'perl')
Total succeeded : 20000
Total failed : 0
Total no reply : 0
Total time (s) : 8.135
Packets/s : 2458
Response times:
< 10 usec : 0
< 100 usec : 0
< msec : 0
< 10 msec : 37
< 0.1s : 16283
< s : 3680
< 10s : 0
< 100s : 0
Total succeeded : 20000
Total failed : 0
Total no reply : 0
Total time (s) : 8.025
Packets/s : 2492
Response times:
< 10 usec : 0
< 100 usec : 0
< msec : 0
< 10 msec : 45
< 0.1s : 16022
< s : 3933
< 10s : 0
< 100s : 0
Total succeeded : 20000
Total failed : 0
Total no reply : 0
Total time (s) : 7.874
Packets/s : 2540
Response times:
< 10 usec : 0
< 100 usec : 0
< msec : 0
< 10 msec : 35
< 0.1s : 16995
< s : 2970
< 10s : 0
< 100s : 0
======================================================================================
Native Java / LocalUsers
Total succeeded : 20000
Total failed : 0
Total no reply : 0
Total time (s) : 1.636
Packets/s : 12221
Response times:
< 10 usec : 0
< 100 usec : 0
< msec : 0
< 10 msec : 2329
< 0.1s : 17671
< s : 0
< 10s : 0
< 100s : 0
Total succeeded : 20000
Total failed : 0
Total no reply : 0
Total time (s) : 1.635
Packets/s : 12229
Response times:
< 10 usec : 0
< 100 usec : 0
< msec : 0
< 10 msec : 2730
< 0.1s : 17270
< s : 0
< 10s : 0
< 100s : 0
Total succeeded : 20000
Total failed : 0
Total no reply : 0
Total time (s) : 1.644
Packets/s : 12160
Response times:
< 10 usec : 0
< 100 usec : 0
< msec : 0
< 10 msec : 2418
< 0.1s : 17582
< s : 0
< 10s : 0
< 100s : 0
More information about the Freeradius-Users
mailing list