On Fri, Jul 10, 2015 at 1:20 PM, Ted Roche <[email protected]> wrote: >> >> That being said, it does seem unlikely that in the ordinary course of >> business we would have over 200 people connecting to their site. It may or >> may not depend on definition of "connection": are there not some >> circumstances under which software enhances the total bandwidth of an >> instance of communication between two internet nodes by means of multiple >> connections? >>
If you had to wait to download all the resources from a typical website in a single queue of "request one jpg... wait... receive the jpg. Request the next CSS file... wait.. " the internet boom would have been over long ago. All modern browsers run requests in parallel. The good ones let you specify how many threads to run, typically a minimum of 4. Smart websites serve static resources from a different host name so they can be further "paralleled." Some internet web sites tell you the magic settings to soup up your browser for more connections, as do some "performance add-ons" Some browsers are overly-aggressive about caching, poking around web sites for what might be the next page you're looking to download, so-called "look-ahead" caching which is too much, imo. I've got some traffic on one of my sites that has all the hallmarks of malware; I traced it down to a couple of clients who have browsers on their desktops, and matching browsers on their "smart" devices. When one device requests a web page, the other will, too, to update its cache. This is the opposite of "smart" imo. When they are talking about "connections" they may mean raw TCP/IP communications blocked at the firewall, http requests blocked at the web server, database connections, or something else. It's a pretty fuzzy term, and may be specific to the application you share with them. There should also be caching going on. Ideally, all "static" resources in a website should be served with a specified lifetime, so the browser caches them all locally. A good web server and/or front end cache should also be serving these from caches and not having to return the same files off disk each time. Smart web browsers ask if cached resources have changed and don't request them again unless they have. If you are getting "expire" headers showing that everything expires immediately, and all have to be requested each time, it's no surprise you're bogging down their web site. It's possible you could rig up a caching server on your side of the relationship, but I wouldn't recommend it unless you have nothing else better to do, which is pretty unlikely. I'd query the vendor further about the "connections" explaining as you have to us that there are only 100 users on your site and you wonder how 200 connections could be made and let them explain to you what's going on. Then, perhaps the two of you can work out how to work better together. If you can't get satisfactory answers, since you have a contractual relationship with the site, perhaps it's time to revisit the contract. -- Ted Roche Ted Roche & Associates, LLC http://www.tedroche.com _______________________________________________ Post Messages to: [email protected] Subscription Maintenance: http://mail.leafe.com/mailman/listinfo/profox OT-free version of this list: http://mail.leafe.com/mailman/listinfo/profoxtech Searchable Archive: http://leafe.com/archives/search/profox This message: http://leafe.com/archives/byMID/profox/cacw6n4uptkwj623j9x32qjqm5kw3omo-6p-jv3ixgv72oy+...@mail.gmail.com ** All postings, unless explicitly stated otherwise, are the opinions of the author, and do not constitute legal or medical advice. This statement is added to the messages for those lawyers who are too stupid to see the obvious.
