Hi all,
First, I want to thank the Calais folks for a hugely enabling service and a great attitude. I'm building a user oriented desktop application and responsiveness / turn-around is key. I'm running into pretty slow performance though, and I'm wondering if perhaps there is something I can do to improve that. Please see the metrics that I've been getting below. I don't have the fastest / lowest latency connection, so the read numbers will reflect that. But as you can see, average response time is ~8 secs while Median time 3.5 secs, which is in itself interesting. I can't see any pattern except that one -- it seems that the response is heavily skewed by a few responses that take a long time. I'm wondering if perhaps I'm doing something really sub-optimal in my query, as my thinking was that performance should typically be much higher than this. I dunno, I've been doing a lot of testing, so maybe I've just been throttled. :)
Below are my basic param setup as well as some simple metrics. I'd appreciate any insight that people might have.
cheers,
Miles
<?xml version="1.0" encoding="UTF-8"?>
<c:params xmlns:c="http://s.opencalais.com/1/pred/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#">
<c:processingDirectives c:contentType="TEXT/HTML"
c:outputFormat="TEXT/SIMPLE" c:omitOutputtingOriginalText="TRUE" c:enableMetadataType="" c:calculateRelevanceScore="FALSE" c:docRDFaccessible="true">
</c:processingDirectives>
<c:userDirectives c:allowDistribution="true"
c:allowSearch="false" c:externalID="17cabs901" c:submitter="ABC">
</c:userDirectives>
<c:externalMetadata>
</c:externalMetadata>
</c:params>
Size Read Read / Size Response Response / Size Response/Read Parse Parse / Size URL
67299 541 0.0080 25908 0.3850 47.8891 25 0.0004 http://www.baroquemusic.org/bqxjsbach.html
136939 468 0.0034 22113 0.1615 47.2500 29 0.0002 http://www.youtube.com/watch?v=ipzR9bhei_o
67299 658 0.0098 26904 0.3998 40.8875 18 0.0003 http://www.baroquemusic.org/bqxjsbach.html
67299 652 0.0097 25518 0.3792 39.1380 19 0.0003 http://www.baroquemusic.org/bqxjsbach.html
67299 623 0.0093 23806 0.3537 38.2119 20 0.0003 http://www.baroquemusic.org/bqxjsbach.html
130820 499 0.0038 13320 0.1018 26.6934 1 0.0000 http://www.youtube.com/watch?v=dZn_VBgkPNY
47840 903 0.0189 19759 0.4130 21.8815 5 0.0001 http://techcrunch.com/2010/05/25/microsofts-j-allard-and-robbie-bach-are-leaving-the-company/
352134 662 0.0019 13467 0.0382 20.3429 10 0.0000 http://knol.google.com/k/my-classical-music-collection
123116 269 0.0022 4919 0.0400 18.2862 2 0.0000 http://www.youtube.com/watch?v=S6yuR8efotI
135997 369 0.0027 6231 0.0458 16.8862 3 0.0000 http://www.youtube.com/watch%3Fv%3DipzR9bhei_o
135811 412 0.0030 4484 0.0330 10.8835 4 0.0000 http://www.youtube.com/watch%3Fv%3DipzR9bhei_o
69671 546 0.0078 4513 0.0648 8.2656 12 0.0002 http://www.boingboing.net/2009/05/27/bach-played-by-dance.html
221353 1855 0.0084 14873 0.0672 8.0178 18 0.0001 http://www.engadget.com/2010/05/25/microsofts-robbie-bach-j-allard-leaving-as-part-of-broader-sha/
134 341 2.5448 2651 19.7836 7.7742 0 0.0000 http://tile.net/bach/index.html
207566 1462 0.0070 9656 0.0465 6.6047 11 0.0001 http://www.engadget.com/2010/01/11/microsofts-bach-says-there-are-other-places-where-zune-logical/
136087 545 0.0040 3454 0.0254 6.3376 4 0.0000 http://www.youtube.com/watch%3Fv%3DipzR9bhei_o
118662 445 0.0038 2819 0.0238 6.3348 2 0.0000 http://www.youtube.com/watch?v=xUHQ2ybTejU
113274 523 0.0046 2957 0.0261 5.6539 5 0.0000 http://www.youtube.com/watch?v=qB76jxBq_gQ
248250 1555 0.0063 8534 0.0344 5.4881 8 0.0000 http://www.huffingtonpost.com/david-bach/start-over-finish-rich-da_b_401075.html
193860 1612 0.0083 7819 0.0403 4.8505 35 0.0002 http://www.engadget.com/2009/10/06/robbie-bach-sits-down-for-a-roundtable-discussion-engadget-is-t/
116912 550 0.0047 2645 0.0226 4.8091 3 0.0000 http://www.youtube.com/watch?v=H8mjikxCBQA
119425 514 0.0043 2456 0.0206 4.7782 0 0.0000 http://www.youtube.com/watch%3Fv%3DLU_QR_FTt3E
5825 593 0.1018 2629 0.4513 4.4334 4 0.0007 http://www.kottke.org/09/05/bach-bach-revolution
66741 1182 0.0177 4848 0.0726 4.1015 24 0.0004 http://www.techflash.com/seattle/2010/05/bach_allard_leaving_microsoft_in_upheaval_of_consumer_business.html
4643 366 0.0788 1420 0.3058 3.8798 1 0.0002 http://www.jsbach.net/bcs/
10606 516 0.0487 1488 0.1403 2.8837 1 0.0001 http://www2.nau.edu/tas3/bachindex.html
56284 1273 0.0226 2861 0.0508 2.2474 17 0.0003 http://www.openculture.com/2010/11/a_big_bach_download.html
26206 679 0.0259 1520 0.0580 2.2386 2 0.0001 http://www.allmusic.com/artist/johann-sebastian-bach-q6980
4363 523 0.1199 1066 0.2443 2.0382 1 0.0002 http://www.bachdigital.org/
66916 1512 0.0226 2849 0.0426 1.8843 20 0.0003 http://www.bookabach.co.nz/
15895 681 0.0428 1251 0.0787 1.8370 2 0.0001 http://www.bach.org/
23256 858 0.0369 1462 0.0629 1.7040 2 0.0001 http://www.bachcentre.com/
20021 1030 0.0514 1755 0.0877 1.7039 2 0.0001 http://www.bachflower.com/
411 455 1.1071 535 1.3017 1.1758 1 0.0024 http://www.sebastianbach.com/
11784 4332 0.3676 1707 0.1449 0.3940 11 0.0009 http://www.selmer.com/
91143 857 0.1349 7834 0.7299 12.2225 9 0.0002 Average
67299 593 0.0093 3454 0.0672 5.6539 4 0.0001 Median

Thanks for the quick reply. It does seem about 2-3x faster currently. I'll update with any changes.
cheers,
Miles
Hello Miles,
Sorry to hear of your response time challenges. We just performed some maintenance on our servers, and think this may help to reduce such issues.
Let us know if you continue to experience this.
Regards,