New Prosper 1.7 Proves 7% Faster

by Mike Chiasson on October 12, 2011 · 7 comments

After reading CTR Tard’s quick review of the new release of Prosper 1.7.2 (view here) I really wanted to get my hands on it. I previously did a case study showing the speed of my Prosper202 redirects vs Tracking202 and the Bevo Media platform. You can view those results here. With 1.7.2 I was really excited to see exactly how much faster my server redirects would be once I cleared the click data. For anyone who doesn’t know, Prosper202 retains a record of all kinds of information for you to perform past queries on old information. It is widely believed that this excess of information kills your server speeds. I’ve spoken to a few internet marketers who preach that they wipe their click database tables on a weekly basis otherwise their ‘server crawls’. Mind you some of these people are doing hundreds of thousands of clicks a day, but some were definitely in the sub 1,000 a day and still claim the speed bonuses.

Testing Time!

I used the same platform and same landing page as before to perform the tests. All prosper tests were performed on a low duty dedicated server. I was really only interested in viewing the Prosper202 stats but I figured it would be nice to compare to another platform’s speed as well. Even though the results show one way, I still opt to continue to use Prosper as my primary tracking platform.

Prosper 1.6 vs. Bevo Media

Prosper 1.6 vs Bevo Media

Prosper 1.7 (No Click Data Removed) vs. Bevo Media

Prosper 1.7 (No Cl

Prosper 1.7 (ALL Click Data Removed) vs. Bevo Media

Prosper 1.7 (All Click Data Removed) vs. Bevo Media

All in all you can see the Prosper 1.7 with click data removed is significantly faster. It topped my original benchmark by just under 7%. If you are focusing on redirect speeds I would recommend an immediate upgrade as the standard upgrade alone also showed a 6% increase in speed.

I’ve never cleared my Prosper202 click data before, some currently do it manually. It reported about 250MB of click data on this install. I was really thinking that it was time for me to wipe the SQL database and just refresh to really get my snappy redirects back, however in the end it seems they were just as snappy as ever and the click data deletion might not be as big of a deal as we all suspected. I still appreciate them adding it in though!

Anyone else get a chance to try it out yet?

About the author

Mike Chiasson Mike Chiasson is the Director of IT for a publicly traded company by day and an Internet Marketer by night. He absolutely hates the words 'serial entrepreneur' but loves discussions about business. You can follow him on Twitter.

matt October 12, 2011 at 4:33 pm

Thanks for posting. Was just running a test on this today. Bevo vs my zensix vps which is by no means high-performance (e.g. no memcache)

* serial testing option to eliminate the sides interfering w/ each other
* 3 tests at 30 loads on whatloadsfaster

test 1 – vps 29% faster1470ms vs bevo 1900ms
test 2 – vps 25% faster1457ms vs bevo 1827ms
test 3 – vps 25% faster1450ms vs bevo 1826ms

Weird my results were so much different than yours w/ p202 vs bevo. Big difference between the criteria of our tests was I was using cloak on both & it looked like you were running standard links. Anyone else w/ a test?

Mike Chiasson October 12, 2011 at 7:02 pm

Wow, thanks for the results. Yes the links we both using cloaking. Uncloaked my Prosper hits under 400ms on that test.

CTRtard October 12, 2011 at 6:15 pm

Nice post Mike. Are you using uncloaked redirects in Prosper? Also there seems to be a huge difference in your old test vs your new one. Bevo is way faster in this second test. So is 202. Is this a function of the destination LP?

Mike Chiasson October 12, 2011 at 7:05 pm

The destination LP is that simple page with just those 2 lines on it. It is hosted on the same server as my 202 install and using the same DNS, etc. Both links were using cloaking. This particular test looks like it uses your own connection to run, I am now on a wireless lan as before I was hard wired FIOS. Perhaps that is the difference as each were considerably faster. Both programs claim to have made speed upgrade though 🙂

Mike Chiasson October 12, 2011 at 7:07 pm

I also moved about 2 miles away from where I was, that much closer to Chicago where my hosting is. That must be the reason!!!! haha.

Marlon October 14, 2011 at 12:34 pm

Weird. I just installed 1.7.2 by copying the files into my prosper install directory (keeping the config file intact) and I still show 1.6.x !!!! Has anyone seen this as well?

Mike Chiasson October 14, 2011 at 1:53 pm

I did the install a few times and didn't get that problem. That's weird, you may want to reach out to Nana.

Comments on this entry are closed.

Previous post:

Next post: