I've written a short blog post on how to keep your heroku dyno from
'timing out' which can lead to a slow loading site for those with a
low amount of traffic.
Once you pay for a dyno, its up all the time, so this strategy is only necessary when you are not yet paying to host your site on heroku.
So I think it makes sense to take the idea of "ping the server every X min", and add a bit more to the logic of asserting that the response is correct and timely. Implementing even a basic strategy for monitoring your application's uptime, performance, and consistent functionality is an important component in how you manage your site, and there are a variety of tools out there that can help.
New relic can ping your app and alert you when it is not responding and passing your assertion test. (specific words on a page)
BrowserMob.com and others can do multistep flows like logging in and navigating to a page.
StillAlive.com and the heroku add on is great for defining a full cucumber-like test suite to ensure that not only is your site up, but it is still functional.
But I hope that people use some restraint when deciding which of their apps they want to 'artificially keep awake', and find a nice balance that uses resources wisely.
Using the free plan on cloudflare.com CDN might be useful, particularly if your main pages and most of the assets can be cached. (the page will load quickly in the browser, from the cdn, then fire an uncacheable request to wake up the server, without the user feeling like its slow).
Actually using cloudflare is a win-win even after you're off the free heroku plan, because it allows you to handle more concurrent users and load things faster.
On Tue, Feb 14, 2012 at 11:34 AM, Nick <ntaberna...@gmail.com> wrote: > I've written a short blog post on how to keep your heroku dyno from > 'timing out' which can lead to a slow loading site for those with a > low amount of traffic.
> -- > You received this message because you are subscribed to the Google Groups > "Heroku" group. > To post to this group, send email to heroku@googlegroups.com. > To unsubscribe from this group, send email to > heroku+unsubscribe@googlegroups.com. > For more options, visit this group at > http://groups.google.com/group/heroku?hl=en.
On Thu, Feb 16, 2012 at 02:34:09PM -0600, John McCaffrey wrote: > Once you pay for a dyno, its up all the time, so this strategy is only > necessary when you are not yet paying to host your site on heroku.
> So I think it makes sense to take the idea of "ping the server every X > min", and add a bit more to the logic of asserting that the response is > correct and timely. Implementing even a basic strategy for monitoring your > application's uptime, performance, and consistent functionality is an > important component in how you manage your site, and there are a variety of > tools out there that can help.
> New relic can ping your app and alert you when it is not responding and > passing your assertion test. (specific words on a page)
> BrowserMob.com and others can do multistep flows like logging in and > navigating to a page.
> StillAlive.com and the heroku add on is great for defining a full > cucumber-like test suite to ensure that not only is your site up, but it is > still functional.
> But I hope that people use some restraint when deciding which of their apps > they want to 'artificially keep awake', and find a nice balance that uses > resources wisely.
> Using the free plan on cloudflare.com CDN might be useful, particularly if > your main pages and most of the assets can be cached. (the page will load > quickly in the browser, from the cdn, then fire an uncacheable request to > wake up the server, without the user feeling like its slow).
> Actually using cloudflare is a win-win even after you're off the free > heroku plan, because it allows you to handle more concurrent users and load > things faster.
> On Tue, Feb 14, 2012 at 11:34 AM, Nick <ntaberna...@gmail.com> wrote:
> > I've written a short blog post on how to keep your heroku dyno from > > 'timing out' which can lead to a slow loading site for those with a > > low amount of traffic.
> > -- > > You received this message because you are subscribed to the Google Groups > > "Heroku" group. > > To post to this group, send email to heroku@googlegroups.com. > > To unsubscribe from this group, send email to > > heroku+unsubscribe@googlegroups.com. > > For more options, visit this group at > > http://groups.google.com/group/heroku?hl=en.
> -- > Thanks, > -John
> -- > You received this message because you are subscribed to the Google Groups "Heroku" group. > To post to this group, send email to heroku@googlegroups.com. > To unsubscribe from this group, send email to heroku+unsubscribe@googlegroups.com. > For more options, visit this group at http://groups.google.com/group/heroku?hl=en.
As a database guy at Heroku, I'm not one to speak authoritatively on this, so please take this as the personal thoughts of someone and not an official statement.
We idle apps in order to avoid having to charge for them. The more people who prevent this behaviour, the more expensive our "free" apps become to run, and the more likely we are to have to change our policies about what we can offer in a free app.
While I admire the ingenuity in this post, I would suggest that you reduce the amount of time your application takes to boot, or simply accept that a few seconds of lag on the first request after a period of idleness is a reasonable trade-off for free web hosting.
I take your points well. I don't mean to try and 'do one over' on
Heroku. I appreciate the service you offer very much. My thinking
behind it was that you would never exceed the 450 hours of dyno time
allocated to each app so there wouldn't be a problem and if you did
you would be charged anyway.
Is there a paid for solution from Heroku to achieve the same result?
The cost jump between 1 free dyno and paying for a dyno is quite large
for small applications. So perhaps you could offer a $10 package which
essentialy works the same way? If i'm honest I don't feel I pay Heroku
enough but I have too many small apps (10 or so) to pay for each one
to have a dedicated dyno.
?
On Feb 16, 8:39 pm, Peter van Hardenberg <p...@heroku.com> wrote:
> As a database guy at Heroku, I'm not one to speak authoritatively on
> this, so please take this as the personal thoughts of someone and not
> an official statement.
> We idle apps in order to avoid having to charge for them. The more
> people who prevent this behaviour, the more expensive our "free" apps
> become to run, and the more likely we are to have to change our
> policies about what we can offer in a free app.
> While I admire the ingenuity in this post, I would suggest that you
> reduce the amount of time your application takes to boot, or simply
> accept that a few seconds of lag on the first request after a period
> of idleness is a reasonable trade-off for free web hosting.
On Fri, Feb 17, 2012 at 4:44 AM, Nick <ntaberna...@gmail.com> wrote: > Is there a paid for solution from Heroku to achieve the same result? > The cost jump between 1 free dyno and paying for a dyno is quite large > for small applications. So perhaps you could offer a $10 package which > essentialy works the same way? If i'm honest I don't feel I pay Heroku > enough but I have too many small apps (10 or so) to pay for each one > to have a dedicated dyno.
Fair question. I don't know that there is a good solution today, but I can definitely imagine how one might feel as though a number of smaller, low-traffic apps which you'd be willing to pay something for might not add up to making sense to pay $350/mo to run.
In the shortest term, addressing the boot time of your app is the fastest way to make initial route times faster, but that's a crummy solution overall. Wish I had better advice or ideas. In the mean time, if you do set up something like this, just keep that in mind. It's not a big deal, but it does put gradual pressure on our systems.
I'll take your question to our people who are thinking about these problems. We'll have to find a solution eventually.
I thought the Cedar dynos didn't turn off when idle. Can't remember where I got that idea from though, so I'm probably entirely wrong.
But there's also http://devcenter.heroku.com/articles/dyno-idling which suggest that you could scale your web dynos to zero (on Cedar) and use a different process name to run the webserver (I'm assuming this would work, haven't tried it). This is because the other processes (non web) could be used for other fun things, like running redis (there was an example in a presentation that showed some TCP routing) which you don't want to turn off.
Also, in your post you say 450 hours, which is incorrect. It's 750.
I think this is a good discussion to have, in terms of appropriate use of 'free resources' to keep the plan sustainable, and "reasonable pricing for the 'little guy'" with tons of small apps that don't get daily usage.
Those topics are important, but not something that will be resolved any time soon, the more tractable computing problem is how to best optimize a site on the free plan, so as to minimize the impact of spinning up a dyno.
A few strategies: 1. use browser and http caching for as much as you can 2. use a cdn like cloudflare.com or others 3. reduce dependencies for your app? 4. figure out what is slow, and how to improve it....
I'm not really sure what is happening in the spin up of a dyno, and what I could do to make it better. I just hit an idle app of mine and looked in the logs to see what it was doing. 2012-02-17T18:22:35+00:00 heroku[web.1]: Unidling 2012-02-17T18:22:35+00:00 heroku[web.1]: State changed from down to created 2012-02-17T18:22:35+00:00 heroku[web.1]: State changed from created to starting 2012-02-17T18:22:39+00:00 heroku[web.1]: Starting process with command `bundle exec unicorn -p 56060 -c ./config/unicorn.rb` 2012-02-17T18:22:41+00:00 app[web.1]: I, [2012-02-17T18:22:41.412560 #1] INFO -- : listening on addr=0.0.0.0:56060 fd=3 2012-02-17T18:22:41+00:00 app[web.1]: I, [2012-02-17T18:22:41.413000 #1] INFO -- : worker=0 spawning... 2012-02-17T18:22:41+00:00 app[web.1]: I, [2012-02-17T18:22:41.416644 #1] INFO -- : worker=1 spawning... 2012-02-17T18:22:41+00:00 app[web.1]: I, [2012-02-17T18:22:41.417177 #9] INFO -- : worker=0 spawned pid=9 2012-02-17T18:22:41+00:00 app[web.1]: I, [2012-02-17T18:22:41.417390 #9] INFO -- : Refreshing Gem list 2012-02-17T18:22:41+00:00 app[web.1]: I, [2012-02-17T18:22:41.418434 #1] INFO -- : worker=2 spawning... 2012-02-17T18:22:41+00:00 app[web.1]: I, [2012-02-17T18:22:41.420059 #1] INFO -- : worker=3 spawning... 2012-02-17T18:22:41+00:00 app[web.1]: I, [2012-02-17T18:22:41.420874 #12] INFO -- : worker=1 spawned pid=12 2012-02-17T18:22:41+00:00 app[web.1]: I, [2012-02-17T18:22:41.421081 #12] INFO -- : Refreshing Gem list 2012-02-17T18:22:41+00:00 app[web.1]: I, [2012-02-17T18:22:41.425084 #1] INFO -- : master process ready 2012-02-17T18:22:41+00:00 app[web.1]: I, [2012-02-17T18:22:41.426771 #14] INFO -- : worker=2 spawned pid=14 2012-02-17T18:22:41+00:00 app[web.1]: I, [2012-02-17T18:22:41.426987 #14] INFO -- : Refreshing Gem list 2012-02-17T18:22:41+00:00 app[web.1]: I, [2012-02-17T18:22:41.446178 #18] INFO -- : worker=3 spawned pid=18 2012-02-17T18:22:41+00:00 app[web.1]: I, [2012-02-17T18:22:41.446496 #18] INFO -- : Refreshing Gem list 2012-02-17T18:22:42+00:00 heroku[web.1]: State changed from starting to up 2012-02-17T18:22:48+00:00 app[web.1]: Cannot find or read /app/config/newrelic.yml 2012-02-17T18:22:55+00:00 app[web.1]: Cannot find or read /app/config/newrelic.yml 2012-02-17T18:22:56+00:00 app[web.1]: Cannot find or read /app/config/newrelic.yml 2012-02-17T18:22:56+00:00 app[web.1]: worker=1 ready 2012-02-17T18:22:56+00:00 app[web.1]: 2012-02-17T18:22:56+00:00 app[web.1]: 2012-02-17T18:22:56+00:00 app[web.1]: Started GET "/" for 67.162.61.119 at 2012-02-17 18:22:56 +0000 2012-02-17T18:22:57+00:00 app[web.1]: Cannot find or read /app/config/newrelic.yml
So if I'm reading this right, a request came in at 18:22:35, and woke up the dyno, and that request passed through to the app at 18:22:56, meaning there was about a 21 second delay.
There seems to be some time in between each " Cannot find or read /app/config/newrelic.yml" I wonder if putting that config in there would make the spin up faster?
I'm also spinning up 4 unicorn workers, would it be faster to start with only 1 or 2, and then add the others later (is that something that could/should be done programmatically?)
I'm curious what other people are seeing for dyno spin up times, and any strategies for making things go faster. (Wonder if I'm including any gems that are not needed in prod, and if that has an impact)
Run heroku logs for one of your idled dynos, then hit the url and see how long it takes, is there a lot of variation? (bamboo vs cedar, thin vs. unicorn, rails vs. sinatra, etc)
On Fri, Feb 17, 2012 at 12:27 PM, Daniel Huckstep <
darkhel...@darkhelmetlive.com> wrote: > I thought the Cedar dynos didn't turn off when idle. Can't remember where > I got that idea from though, so I'm probably entirely wrong.
> But there's also http://devcenter.heroku.com/articles/dyno-idling which > suggest that you could scale your web dynos to zero (on Cedar) and use a > different process name to run the webserver (I'm assuming this would work, > haven't tried it). This is because the other processes (non web) could be > used for other fun things, like running redis (there was an example in a > presentation that showed some TCP routing) which you don't want to turn off.
> Also, in your post you say 450 hours, which is incorrect. It's 750.
> To post to this group, send email to heroku@googlegroups.com. > To unsubscribe from this group, send email to > heroku+unsubscribe@googlegroups.com. > For more options, visit this group at > http://groups.google.com/group/heroku?hl=en.
I would like to +1 the idea to come up with a way to get an always responsive app on Heroku for a reasonable price.
The current free plan makes the site so unresponsive on the first request that it's really only good for testing and toy sites, the next level for $36/month and 2 permanent dynos is a big step away for many people. I think there is a sweet spot in between there, in which Heroku could reach an monetize a large market segment.
Just my 2 cents. :)
Kevin
On Feb 17, 2012, at 10:27 AM, Daniel Huckstep <darkhel...@darkhelmetlive.com> wrote:
> I thought the Cedar dynos didn't turn off when idle. Can't remember where I got that idea from though, so I'm probably entirely wrong.
> But there's also http://devcenter.heroku.com/articles/dyno-idling which suggest that you could scale your web dynos to zero (on Cedar) and use a different process name to run the webserver (I'm assuming this would work, haven't tried it). This is because the other processes (non web) could be used for other fun things, like running redis (there was an example in a presentation that showed some TCP routing) which you don't want to turn off.
> Also, in your post you say 450 hours, which is incorrect. It's 750. > -- > You received this message because you are subscribed to the Google Groups "Heroku" group. > To view this discussion on the web visit https://groups.google.com/d/msg/heroku/-/h5MWNmpn3JIJ. > To post to this group, send email to heroku@googlegroups.com. > To unsubscribe from this group, send email to heroku+unsubscribe@googlegroups.com. > For more options, visit this group at http://groups.google.com/group/heroku?hl=en.
The 'starter' package is only ~$35/mo which isn't exactly monumentally expensive. Are you suggesting something between that and free? What you're suggesting sounds like your charged by the CPU cycle rather than the hours?
To be brutally honest, I host loads of apps on 1 web dyno and just make sure that the spin up time is short enough that it's not a problem. If I ever need to run more than 1 web dyno it's generally because the traffic levels require it, in which case $35 becomes less of a problem.
Personally, I think that having a single dyno, which can still serve hundreds of thousands of requests a day /for free/ is a pretty good deal. I'm happy to pay $35 to double it.
On Friday, 17 February 2012 at 12:44, Nick wrote: > Peter,
> I take your points well. I don't mean to try and 'do one over' on > Heroku. I appreciate the service you offer very much. My thinking > behind it was that you would never exceed the 450 hours of dyno time > allocated to each app so there wouldn't be a problem and if you did > you would be charged anyway.
> Is there a paid for solution from Heroku to achieve the same result? > The cost jump between 1 free dyno and paying for a dyno is quite large > for small applications. So perhaps you could offer a $10 package which > essentialy works the same way? If i'm honest I don't feel I pay Heroku > enough but I have too many small apps (10 or so) to pay for each one > to have a dedicated dyno.
> ? > On Feb 16, 8:39 pm, Peter van Hardenberg <p...@heroku.com (http://heroku.com)> wrote: > > As a database guy at Heroku, I'm not one to speak authoritatively on > > this, so please take this as the personal thoughts of someone and not > > an official statement.
> > We idle apps in order to avoid having to charge for them. The more > > people who prevent this behaviour, the more expensive our "free" apps > > become to run, and the more likely we are to have to change our > > policies about what we can offer in a free app.
> > While I admire the ingenuity in this post, I would suggest that you > > reduce the amount of time your application takes to boot, or simply > > accept that a few seconds of lag on the first request after a period > > of idleness is a reasonable trade-off for free web hosting.
> > Peter
> -- > You received this message because you are subscribed to the Google Groups "Heroku" group. > To post to this group, send email to heroku@googlegroups.com (mailto:heroku@googlegroups.com). > To unsubscribe from this group, send email to heroku+unsubscribe@googlegroups.com (mailto:heroku+unsubscribe@googlegroups.com). > For more options, visit this group at http://groups.google.com/group/heroku?hl=en.
I think having a single permanent (i.e. always instantly responding) dyno for $10-20/month would make sense for a lot of people. It definitely provides for a much better user experience than the current free plan, especially for low traffic scenarios.
It could work like this: free plan as before, for testing etc. Once in a low-traffic production environment, upgrade the free dyno to a permanent dyno for $10-20/month. Then later, buy a second dyno for $35/month.
Kevin
On Feb 17, 2012, at 4:48 AM, Neil Middleton <neil.middle...@gmail.com> wrote:
> The 'starter' package is only ~$35/mo which isn't exactly monumentally expensive. Are you suggesting something between that and free? > What you're suggesting sounds like your charged by the CPU cycle rather than the hours?
> To be brutally honest, I host loads of apps on 1 web dyno and just make sure that the spin up time is short enough that it's not a problem. If I ever need to run more than 1 web dyno it's generally because the traffic levels require it, in which case $35 becomes less of a problem.
> Personally, I think that having a single dyno, which can still serve hundreds of thousands of requests a day /for free/ is a pretty good deal. I'm happy to pay $35 to double it.
> Neil
> On Friday, 17 February 2012 at 12:44, Nick wrote:
>> Peter,
>> I take your points well. I don't mean to try and 'do one over' on >> Heroku. I appreciate the service you offer very much. My thinking >> behind it was that you would never exceed the 450 hours of dyno time >> allocated to each app so there wouldn't be a problem and if you did >> you would be charged anyway.
>> Is there a paid for solution from Heroku to achieve the same result? >> The cost jump between 1 free dyno and paying for a dyno is quite large >> for small applications. So perhaps you could offer a $10 package which >> essentialy works the same way? If i'm honest I don't feel I pay Heroku >> enough but I have too many small apps (10 or so) to pay for each one >> to have a dedicated dyno.
>> ? >> On Feb 16, 8:39 pm, Peter van Hardenberg <p...@heroku.com> wrote: >>> As a database guy at Heroku, I'm not one to speak authoritatively on >>> this, so please take this as the personal thoughts of someone and not >>> an official statement.
>>> We idle apps in order to avoid having to charge for them. The more >>> people who prevent this behaviour, the more expensive our "free" apps >>> become to run, and the more likely we are to have to change our >>> policies about what we can offer in a free app.
>>> While I admire the ingenuity in this post, I would suggest that you >>> reduce the amount of time your application takes to boot, or simply >>> accept that a few seconds of lag on the first request after a period >>> of idleness is a reasonable trade-off for free web hosting.
>>> Peter
>> -- >> You received this message because you are subscribed to the Google Groups "Heroku" group. >> To post to this group, send email to heroku@googlegroups.com. >> To unsubscribe from this group, send email to heroku+unsubscribe@googlegroups.com. >> For more options, visit this group at http://groups.google.com/group/heroku?hl=en.
> -- > You received this message because you are subscribed to the Google Groups "Heroku" group. > To post to this group, send email to heroku@googlegroups.com. > To unsubscribe from this group, send email to heroku+unsubscribe@googlegroups.com. > For more options, visit this group at http://groups.google.com/group/heroku?hl=en.
> The 'starter' package is only ~$35/mo which isn't exactly monumentally expensive. Are you suggesting something between that and free?
> What you're suggesting sounds like your charged by the CPU cycle rather than the hours?
> To be brutally honest, I host loads of apps on 1 web dyno and just make sure that the spin up time is short enough that it's not a problem. If I ever need to run more than 1 web dyno it's generally because the traffic levels require it, in which case $35 becomes less of a problem.
> Personally, I think that having a single dyno, which can still serve hundreds of thousands of requests a day /for free/ is a pretty good deal. I'm happy to pay $35 to double it.
> Neil
> On Friday, 17 February 2012 at 12:44, Nick wrote:
> > Peter,
> > I take your points well. I don't mean to try and 'do one over' on
> > Heroku. I appreciate the service you offer very much. My thinking
> > behind it was that you would never exceed the 450 hours of dyno time
> > allocated to each app so there wouldn't be a problem and if you did
> > you would be charged anyway.
> > Is there a paid for solution from Heroku to achieve the same result?
> > The cost jump between 1 free dyno and paying for a dyno is quite large
> > for small applications. So perhaps you could offer a $10 package which
> > essentialy works the same way? If i'm honest I don't feel I pay Heroku
> > enough but I have too many small apps (10 or so) to pay for each one
> > to have a dedicated dyno.
> > ?
> > On Feb 16, 8:39 pm, Peter van Hardenberg <p...@heroku.com (http://heroku.com)> wrote:
> > > As a database guy at Heroku, I'm not one to speak authoritatively on
> > > this, so please take this as the personal thoughts of someone and not
> > > an official statement.
> > > We idle apps in order to avoid having to charge for them. The more
> > > people who prevent this behaviour, the more expensive our "free" apps
> > > become to run, and the more likely we are to have to change our
> > > policies about what we can offer in a free app.
> > > While I admire the ingenuity in this post, I would suggest that you
> > > reduce the amount of time your application takes to boot, or simply
> > > accept that a few seconds of lag on the first request after a period
> > > of idleness is a reasonable trade-off for free web hosting.
> > > Peter
> > --
> > You received this message because you are subscribed to the Google Groups "Heroku" group.
> > To post to this group, send email to heroku@googlegroups.com (mailto:heroku@googlegroups.com).
> > To unsubscribe from this group, send email to heroku+unsubscribe@googlegroups.com (mailto:heroku+unsubscribe@googlegroups.com).
> > For more options, visit this group athttp://groups.google.com/group/heroku?hl=en.
To be brutally honest, there is no hard and fast rule, but I have found the following:
- Rails is really fast at slowing down, so ensure that your dependencies and initialisers are at an absolute minimum.
- Keep things small and lightweight, break your application up into lots of smaller applications. Sinatra, for instance, spins up almost instantly on Heroku and can take you a fair distance in terms of simple functionality
- Generally speaking, the newer the Rails version, the quicker the spinup time, especially if you're not using ActiveRecord
- If you're feeling beta-y, try running on Ruby 1.9.3, this makes a massive difference to Rails.
Other than this, I'm not really aware of any silver bullet, but I do know that lots of smaller more focused applications spin up better than one monolithic Rails app, which is why Unix is so good I guess.
On Monday, 20 February 2012 at 21:56, Ed Jones wrote: > Neil said: > > I host loads of apps on 1 web dyno and just make sure that the spin up time is short enough that it's not a problem.
> Neil and others, thanks.
> I just launched a new (beta) app, and the response time is just > horrible.
> Could you pass a few tips on how to "make sure that the spin up time > is short enough"? > Thanks!!!
> ed
> On Feb 17, 7:48 am, Neil Middleton <neil.middle...@gmail.com (http://gmail.com)> wrote: > > I'm confused here.
> > The 'starter' package is only ~$35/mo which isn't exactly monumentally expensive. Are you suggesting something between that and free? > > What you're suggesting sounds like your charged by the CPU cycle rather than the hours?
> > To be brutally honest, I host loads of apps on 1 web dyno and just make sure that the spin up time is short enough that it's not a problem. If I ever need to run more than 1 web dyno it's generally because the traffic levels require it, in which case $35 becomes less of a problem.
> > Personally, I think that having a single dyno, which can still serve hundreds of thousands of requests a day /for free/ is a pretty good deal. I'm happy to pay $35 to double it.
> > Neil
> > On Friday, 17 February 2012 at 12:44, Nick wrote: > > > Peter,
> > > I take your points well. I don't mean to try and 'do one over' on > > > Heroku. I appreciate the service you offer very much. My thinking > > > behind it was that you would never exceed the 450 hours of dyno time > > > allocated to each app so there wouldn't be a problem and if you did > > > you would be charged anyway.
> > > Is there a paid for solution from Heroku to achieve the same result? > > > The cost jump between 1 free dyno and paying for a dyno is quite large > > > for small applications. So perhaps you could offer a $10 package which > > > essentialy works the same way? If i'm honest I don't feel I pay Heroku > > > enough but I have too many small apps (10 or so) to pay for each one > > > to have a dedicated dyno.
> > > ? > > > On Feb 16, 8:39 pm, Peter van Hardenberg <p...@heroku.com (http://heroku.com)> wrote: > > > > As a database guy at Heroku, I'm not one to speak authoritatively on > > > > this, so please take this as the personal thoughts of someone and not > > > > an official statement.
> > > > We idle apps in order to avoid having to charge for them. The more > > > > people who prevent this behaviour, the more expensive our "free" apps > > > > become to run, and the more likely we are to have to change our > > > > policies about what we can offer in a free app.
> > > > While I admire the ingenuity in this post, I would suggest that you > > > > reduce the amount of time your application takes to boot, or simply > > > > accept that a few seconds of lag on the first request after a period > > > > of idleness is a reasonable trade-off for free web hosting.
> > > > Peter
> > > -- > > > You received this message because you are subscribed to the Google Groups "Heroku" group. > > > To post to this group, send email to heroku@googlegroups.com (mailto:heroku@googlegroups.com). > > > To unsubscribe from this group, send email to heroku+unsubscribe@googlegroups.com (mailto:heroku+unsubscribe@googlegroups.com). > > > For more options, visit this group athttp://groups.google.com/group/heroku?hl=en.
> -- > You received this message because you are subscribed to the Google Groups "Heroku" group. > To post to this group, send email to heroku@googlegroups.com (mailto:heroku@googlegroups.com). > To unsubscribe from this group, send email to heroku+unsubscribe@googlegroups.com (mailto:heroku+unsubscribe@googlegroups.com). > For more options, visit this group at http://groups.google.com/group/heroku?hl=en.
These recommandations for Rails apps looks very good,
But in my case i use Scala and the netty server and spin-up time is
not good at all. And I am not sure it can be tuned so much as the
sinatra case when the platform is java.
Do anyone have any suggestions for my combo?
Kind regards
Steffen
On Feb 20, 11:15 pm, Neil Middleton <neil.middle...@gmail.com> wrote:
> To be brutally honest, there is no hard and fast rule, but I have found the following:
> - Rails is really fast at slowing down, so ensure that your dependencies and initialisers are at an absolute minimum.
> - Keep things small and lightweight, break your application up into lots of smaller applications. Sinatra, for instance, spins up almost instantly on Heroku and can take you a fair distance in terms of simple functionality
> - Generally speaking, the newer the Rails version, the quicker the spinup time, especially if you're not using ActiveRecord
> - If you're feeling beta-y, try running on Ruby 1.9.3, this makes a massive difference to Rails.
> Other than this, I'm not really aware of any silver bullet, but I do know that lots of smaller more focused applications spin up better than one monolithic Rails app, which is why Unix is so good I guess.
> -Neil
> On Monday, 20 February 2012 at 21:56, Ed Jones wrote:
> > Neil said:
> > > I host loads of apps on 1 web dyno and just make sure that the spin up time is short enough that it's not a problem.
> > Neil and others, thanks.
> > I just launched a new (beta) app, and the response time is just
> > horrible.
> > Could you pass a few tips on how to "make sure that the spin up time
> > is short enough"?
> > Thanks!!!
> > ed
> > On Feb 17, 7:48 am, Neil Middleton <neil.middle...@gmail.com (http://gmail.com)> wrote:
> > > I'm confused here.
> > > The 'starter' package is only ~$35/mo which isn't exactly monumentally expensive. Are you suggesting something between that and free?
> > > What you're suggesting sounds like your charged by the CPU cycle rather than the hours?
> > > To be brutally honest, I host loads of apps on 1 web dyno and just make sure that the spin up time is short enough that it's not a problem. If I ever need to run more than 1 web dyno it's generally because the traffic levels require it, in which case $35 becomes less of a problem.
> > > Personally, I think that having a single dyno, which can still serve hundreds of thousands of requests a day /for free/ is a pretty good deal. I'm happy to pay $35 to double it.
> > > Neil
> > > On Friday, 17 February 2012 at 12:44, Nick wrote:
> > > > Peter,
> > > > I take your points well. I don't mean to try and 'do one over' on
> > > > Heroku. I appreciate the service you offer very much. My thinking
> > > > behind it was that you would never exceed the 450 hours of dyno time
> > > > allocated to each app so there wouldn't be a problem and if you did
> > > > you would be charged anyway.
> > > > Is there a paid for solution from Heroku to achieve the same result?
> > > > The cost jump between 1 free dyno and paying for a dyno is quite large
> > > > for small applications. So perhaps you could offer a $10 package which
> > > > essentialy works the same way? If i'm honest I don't feel I pay Heroku
> > > > enough but I have too many small apps (10 or so) to pay for each one
> > > > to have a dedicated dyno.
> > > > ?
> > > > On Feb 16, 8:39 pm, Peter van Hardenberg <p...@heroku.com (http://heroku.com)> wrote:
> > > > > As a database guy at Heroku, I'm not one to speak authoritatively on
> > > > > this, so please take this as the personal thoughts of someone and not
> > > > > an official statement.
> > > > > We idle apps in order to avoid having to charge for them. The more
> > > > > people who prevent this behaviour, the more expensive our "free" apps
> > > > > become to run, and the more likely we are to have to change our
> > > > > policies about what we can offer in a free app.
> > > > > While I admire the ingenuity in this post, I would suggest that you
> > > > > reduce the amount of time your application takes to boot, or simply
> > > > > accept that a few seconds of lag on the first request after a period
> > > > > of idleness is a reasonable trade-off for free web hosting.
> > > > > Peter
> > > > --
> > > > You received this message because you are subscribed to the Google Groups "Heroku" group.
> > > > To post to this group, send email to heroku@googlegroups.com (mailto:heroku@googlegroups.com).
> > > > To unsubscribe from this group, send email to heroku+unsubscribe@googlegroups.com (mailto:heroku+unsubscribe@googlegroups.com).
> > > > For more options, visit this group athttp://groups.google.com/group/heroku?hl=en.
> > --
> > You received this message because you are subscribed to the Google Groups "Heroku" group.
> > To post to this group, send email to heroku@googlegroups.com (mailto:heroku@googlegroups.com).
> > To unsubscribe from this group, send email to heroku+unsubscribe@googlegroups.com (mailto:heroku+unsubscribe@googlegroups.com).
> > For more options, visit this group athttp://groups.google.com/group/heroku?hl=en.
> These recommandations for Rails apps looks very good,
> But in my case i use Scala and the netty server and spin-up time is
> not good at all. And I am not sure it can be tuned so much as the
> sinatra case when the platform is java.
> Do anyone have any suggestions for my combo?
> Kind regards
> Steffen
> On Feb 20, 11:15 pm, Neil Middleton <neil.middle...@gmail.com> wrote:
> > To be brutally honest, there is no hard and fast rule, but I have found the following:
> > - Rails is really fast at slowing down, so ensure that your dependencies and initialisers are at an absolute minimum.
> > - Keep things small and lightweight, break your application up into lots of smaller applications. Sinatra, for instance, spins up almost instantly on Heroku and can take you a fair distance in terms of simple functionality
> > - Generally speaking, the newer the Rails version, the quicker the spinup time, especially if you're not using ActiveRecord
> > - If you're feeling beta-y, try running on Ruby 1.9.3, this makes a massive difference to Rails.
> > Other than this, I'm not really aware of any silver bullet, but I do know that lots of smaller more focused applications spin up better than one monolithic Rails app, which is why Unix is so good I guess.
> > -Neil
> > On Monday, 20 February 2012 at 21:56, Ed Jones wrote:
> > > Neil said:
> > > > I host loads of apps on 1 web dyno and just make sure that the spin up time is short enough that it's not a problem.
> > > Neil and others, thanks.
> > > I just launched a new (beta) app, and the response time is just
> > > horrible.
> > > Could you pass a few tips on how to "make sure that the spin up time
> > > is short enough"?
> > > Thanks!!!
> > > ed
> > > On Feb 17, 7:48 am, Neil Middleton <neil.middle...@gmail.com (http://gmail.com)> wrote:
> > > > I'm confused here.
> > > > The 'starter' package is only ~$35/mo which isn't exactly monumentally expensive. Are you suggesting something between that and free?
> > > > What you're suggesting sounds like your charged by the CPU cycle rather than the hours?
> > > > To be brutally honest, I host loads of apps on 1 web dyno and just make sure that the spin up time is short enough that it's not a problem. If I ever need to run more than 1 web dyno it's generally because the traffic levels require it, in which case $35 becomes less of a problem.
> > > > Personally, I think that having a single dyno, which can still serve hundreds of thousands of requests a day /for free/ is a pretty good deal. I'm happy to pay $35 to double it.
> > > > Neil
> > > > On Friday, 17 February 2012 at 12:44, Nick wrote:
> > > > > Peter,
> > > > > I take your points well. I don't mean to try and 'do one over' on
> > > > > Heroku. I appreciate the service you offer very much. My thinking
> > > > > behind it was that you would never exceed the 450 hours of dyno time
> > > > > allocated to each app so there wouldn't be a problem and if you did
> > > > > you would be charged anyway.
> > > > > Is there a paid for solution from Heroku to achieve the same result?
> > > > > The cost jump between 1 free dyno and paying for a dyno is quite large
> > > > > for small applications. So perhaps you could offer a $10 package which
> > > > > essentialy works the same way? If i'm honest I don't feel I pay Heroku
> > > > > enough but I have too many small apps (10 or so) to pay for each one
> > > > > to have a dedicated dyno.
> > > > > ?
> > > > > On Feb 16, 8:39 pm, Peter van Hardenberg <p...@heroku.com (http://heroku.com)> wrote:
> > > > > > As a database guy at Heroku, I'm not one to speak authoritatively on
> > > > > > this, so please take this as the personal thoughts of someone and not
> > > > > > an official statement.
> > > > > > We idle apps in order to avoid having to charge for them. The more
> > > > > > people who prevent this behaviour, the more expensive our "free" apps
> > > > > > become to run, and the more likely we are to have to change our
> > > > > > policies about what we can offer in a free app.
> > > > > > While I admire the ingenuity in this post, I would suggest that you
> > > > > > reduce the amount of time your application takes to boot, or simply
> > > > > > accept that a few seconds of lag on the first request after a period
> > > > > > of idleness is a reasonable trade-off for free web hosting.
> > > > > > Peter
> > > > > --
> > > > > You received this message because you are subscribed to the Google Groups "Heroku" group.
> > > > > To post to this group, send email to heroku@googlegroups.com (mailto:heroku@googlegroups.com).
> > > > > To unsubscribe from this group, send email to heroku+unsubscribe@googlegroups.com (mailto:heroku+unsubscribe@googlegroups.com).
> > > > > For more options, visit this group athttp://groups.google.com/group/heroku?hl=en.
> > > --
> > > You received this message because you are subscribed to the Google Groups "Heroku" group.
> > > To post to this group, send email to heroku@googlegroups.com (mailto:heroku@googlegroups.com).
> > > To unsubscribe from this group, send email to heroku+unsubscribe@googlegroups.com (mailto:heroku+unsubscribe@googlegroups.com).
> > > For more options, visit this group athttp://groups.google.com/group/heroku?hl=en.
On Wed, Feb 22, 2012 at 4:17 AM, Nick <ntaberna...@gmail.com> wrote: > An option for a single site is http://www.pingdom.com/
> They can 'ping' your site every hour thereby keeping it awake. It's > free for one site.
> On Feb 21, 8:38 am, Sejensen <sejen...@gmail.com> wrote: > > Hi Neil and others
> > These recommandations for Rails apps looks very good,
> > But in my case i use Scala and the netty server and spin-up time is > > not good at all. And I am not sure it can be tuned so much as the > > sinatra case when the platform is java.
> > Do anyone have any suggestions for my combo?
> > Kind regards > > Steffen
> > On Feb 20, 11:15 pm, Neil Middleton <neil.middle...@gmail.com> wrote:
> > > To be brutally honest, there is no hard and fast rule, but I have > found the following:
> > > - Rails is really fast at slowing down, so ensure that your > dependencies and initialisers are at an absolute minimum.
> > > - Keep things small and lightweight, break your application up into > lots of smaller applications. Sinatra, for instance, spins up almost > instantly on Heroku and can take you a fair distance in terms of simple > functionality
> > > - Generally speaking, the newer the Rails version, the quicker the > spinup time, especially if you're not using ActiveRecord
> > > - If you're feeling beta-y, try running on Ruby 1.9.3, this makes a > massive difference to Rails.
> > > Other than this, I'm not really aware of any silver bullet, but I do > know that lots of smaller more focused applications spin up better than one > monolithic Rails app, which is why Unix is so good I guess.
> > > -Neil
> > > On Monday, 20 February 2012 at 21:56, Ed Jones wrote: > > > > Neil said: > > > > > I host loads of apps on 1 web dyno and just make sure that the > spin up time is short enough that it's not a problem.
> > > > Neil and others, thanks.
> > > > I just launched a new (beta) app, and the response time is just > > > > horrible.
> > > > Could you pass a few tips on how to "make sure that the spin up time > > > > is short enough"? > > > > Thanks!!!
> > > > ed
> > > > On Feb 17, 7:48 am, Neil Middleton <neil.middle...@gmail.com ( > http://gmail.com)> wrote: > > > > > I'm confused here.
> > > > > The 'starter' package is only ~$35/mo which isn't exactly > monumentally expensive. Are you suggesting something between that and free? > > > > > What you're suggesting sounds like your charged by the CPU cycle > rather than the hours?
> > > > > To be brutally honest, I host loads of apps on 1 web dyno and just > make sure that the spin up time is short enough that it's not a problem. > If I ever need to run more than 1 web dyno it's generally because the > traffic levels require it, in which case $35 becomes less of a problem.
> > > > > Personally, I think that having a single dyno, which can still > serve hundreds of thousands of requests a day /for free/ is a pretty good > deal. I'm happy to pay $35 to double it.
> > > > > Neil
> > > > > On Friday, 17 February 2012 at 12:44, Nick wrote: > > > > > > Peter,
> > > > > > I take your points well. I don't mean to try and 'do one over' on > > > > > > Heroku. I appreciate the service you offer very much. My thinking > > > > > > behind it was that you would never exceed the 450 hours of dyno > time > > > > > > allocated to each app so there wouldn't be a problem and if you > did > > > > > > you would be charged anyway.
> > > > > > Is there a paid for solution from Heroku to achieve the same > result? > > > > > > The cost jump between 1 free dyno and paying for a dyno is quite > large > > > > > > for small applications. So perhaps you could offer a $10 package > which > > > > > > essentialy works the same way? If i'm honest I don't feel I pay > Heroku > > > > > > enough but I have too many small apps (10 or so) to pay for each > one > > > > > > to have a dedicated dyno.
> > > > > > ? > > > > > > On Feb 16, 8:39 pm, Peter van Hardenberg <p...@heroku.com ( > http://heroku.com)> wrote: > > > > > > > As a database guy at Heroku, I'm not one to speak > authoritatively on > > > > > > > this, so please take this as the personal thoughts of someone > and not > > > > > > > an official statement.
> > > > > > > We idle apps in order to avoid having to charge for them. The > more > > > > > > > people who prevent this behaviour, the more expensive our > "free" apps > > > > > > > become to run, and the more likely we are to have to change our > > > > > > > policies about what we can offer in a free app.
> > > > > > > While I admire the ingenuity in this post, I would suggest > that you > > > > > > > reduce the amount of time your application takes to boot, or > simply > > > > > > > accept that a few seconds of lag on the first request after a > period > > > > > > > of idleness is a reasonable trade-off for free web hosting.
> > > > > > > Peter
> > > > > > -- > > > > > > You received this message because you are subscribed to the > Google Groups "Heroku" group. > > > > > > To post to this group, send email to heroku@googlegroups.com(mailto: > heroku@googlegroups.com). > > > > > > To unsubscribe from this group, send email to > heroku+unsubscribe@googlegroups.com (mailto: > heroku+unsubscribe@googlegroups.com). > > > > > > For more options, visit this group athttp:// > groups.google.com/group/heroku?hl=en.
> > > > -- > > > > You received this message because you are subscribed to the Google > Groups "Heroku" group. > > > > To post to this group, send email to heroku@googlegroups.com(mailto: > heroku@googlegroups.com). > > > > To unsubscribe from this group, send email to > heroku+unsubscribe@googlegroups.com (mailto: > heroku+unsubscribe@googlegroups.com). > > > > For more options, visit this group athttp:// > groups.google.com/group/heroku?hl=en.
> -- > You received this message because you are subscribed to the Google Groups > "Heroku" group. > To post to this group, send email to heroku@googlegroups.com. > To unsubscribe from this group, send email to > heroku+unsubscribe@googlegroups.com. > For more options, visit this group at > http://groups.google.com/group/heroku?hl=en.
On Wednesday, February 22, 2012 at 10:04 AM, Kevin Goslar wrote: > With that, you keep Heroku unnecessarily busy by having them start a dyno 24 times every day, and your app still sleeps 55 min per hour.
> On Wed, Feb 22, 2012 at 4:17 AM, Nick <ntaberna...@gmail.com (mailto:ntaberna...@gmail.com)> wrote: > > An option for a single site is http://www.pingdom.com/
> > They can 'ping' your site every hour thereby keeping it awake. It's > > free for one site.
> > On Feb 21, 8:38 am, Sejensen <sejen...@gmail.com (mailto:sejen...@gmail.com)> wrote: > > > Hi Neil and others
> > > These recommandations for Rails apps looks very good,
> > > But in my case i use Scala and the netty server and spin-up time is > > > not good at all. And I am not sure it can be tuned so much as the > > > sinatra case when the platform is java.
> > > Do anyone have any suggestions for my combo?
> > > Kind regards > > > Steffen
> > > On Feb 20, 11:15 pm, Neil Middleton <neil.middle...@gmail.com (mailto:neil.middle...@gmail.com)> wrote:
> > > > To be brutally honest, there is no hard and fast rule, but I have found the following:
> > > > - Rails is really fast at slowing down, so ensure that your dependencies and initialisers are at an absolute minimum.
> > > > - Keep things small and lightweight, break your application up into lots of smaller applications. Sinatra, for instance, spins up almost instantly on Heroku and can take you a fair distance in terms of simple functionality
> > > > - Generally speaking, the newer the Rails version, the quicker the spinup time, especially if you're not using ActiveRecord
> > > > - If you're feeling beta-y, try running on Ruby 1.9.3, this makes a massive difference to Rails.
> > > > Other than this, I'm not really aware of any silver bullet, but I do know that lots of smaller more focused applications spin up better than one monolithic Rails app, which is why Unix is so good I guess.
> > > > -Neil
> > > > On Monday, 20 February 2012 at 21:56, Ed Jones wrote: > > > > > Neil said: > > > > > > I host loads of apps on 1 web dyno and just make sure that the spin up time is short enough that it's not a problem.
> > > > > Neil and others, thanks.
> > > > > I just launched a new (beta) app, and the response time is just > > > > > horrible.
> > > > > Could you pass a few tips on how to "make sure that the spin up time > > > > > is short enough"? > > > > > Thanks!!!
> > > > > ed
> > > > > On Feb 17, 7:48 am, Neil Middleton <neil.middle...@gmail.com (mailto:neil.middle...@gmail.com) (http://gmail.com)> wrote: > > > > > > I'm confused here.
> > > > > > The 'starter' package is only ~$35/mo which isn't exactly monumentally expensive. Are you suggesting something between that and free? > > > > > > What you're suggesting sounds like your charged by the CPU cycle rather than the hours?
> > > > > > To be brutally honest, I host loads of apps on 1 web dyno and just make sure that the spin up time is short enough that it's not a problem. If I ever need to run more than 1 web dyno it's generally because the traffic levels require it, in which case $35 becomes less of a problem.
> > > > > > Personally, I think that having a single dyno, which can still serve hundreds of thousands of requests a day /for free/ is a pretty good deal. I'm happy to pay $35 to double it.
> > > > > > Neil
> > > > > > On Friday, 17 February 2012 at 12:44, Nick wrote: > > > > > > > Peter,
> > > > > > > I take your points well. I don't mean to try and 'do one over' on > > > > > > > Heroku. I appreciate the service you offer very much. My thinking > > > > > > > behind it was that you would never exceed the 450 hours of dyno time > > > > > > > allocated to each app so there wouldn't be a problem and if you did > > > > > > > you would be charged anyway.
> > > > > > > Is there a paid for solution from Heroku to achieve the same result? > > > > > > > The cost jump between 1 free dyno and paying for a dyno is quite large > > > > > > > for small applications. So perhaps you could offer a $10 package which > > > > > > > essentialy works the same way? If i'm honest I don't feel I pay Heroku > > > > > > > enough but I have too many small apps (10 or so) to pay for each one > > > > > > > to have a dedicated dyno.
> > > > > > > ? > > > > > > > On Feb 16, 8:39 pm, Peter van Hardenberg <p...@heroku.com (mailto:p...@heroku.com) (http://heroku.com)> wrote: > > > > > > > > As a database guy at Heroku, I'm not one to speak authoritatively on > > > > > > > > this, so please take this as the personal thoughts of someone and not > > > > > > > > an official statement.
> > > > > > > > We idle apps in order to avoid having to charge for them. The more > > > > > > > > people who prevent this behaviour, the more expensive our "free" apps > > > > > > > > become to run, and the more likely we are to have to change our > > > > > > > > policies about what we can offer in a free app.
> > > > > > > > While I admire the ingenuity in this post, I would suggest that you > > > > > > > > reduce the amount of time your application takes to boot, or simply > > > > > > > > accept that a few seconds of lag on the first request after a period > > > > > > > > of idleness is a reasonable trade-off for free web hosting.
> > > > > > > > Peter
> > > > > > > -- > > > > > > > You received this message because you are subscribed to the Google Groups "Heroku" group. > > > > > > > To post to this group, send email to heroku@googlegroups.com (mailto:heroku@googlegroups.com) (mailto:heroku@googlegroups.com). > > > > > > > To unsubscribe from this group, send email to heroku+unsubscribe@googlegroups.com (mailto:heroku%2Bunsubscribe@googlegroups.com) (mailto:heroku+unsubscribe@googlegroups.com (mailto:heroku%2Bunsubscribe@googlegroups.com)). > > > > > > > For more options, visit this group athttp://groups.google.com/group/heroku?hl=en.
> > > > > -- > > > > > You received this message because you are subscribed to the Google Groups "Heroku" group. > > > > > To post to this group, send email to heroku@googlegroups.com (mailto:heroku@googlegroups.com) (mailto:heroku@googlegroups.com). > > > > > To unsubscribe from this group, send email to heroku+unsubscribe@googlegroups.com (mailto:heroku%2Bunsubscribe@googlegroups.com) (mailto:heroku+unsubscribe@googlegroups.com (mailto:heroku%2Bunsubscribe@googlegroups.com)). > > > > > For more options, visit this group athttp://groups.google.com/group/heroku?hl=en.
> > -- > > You received this message because you are subscribed to the Google Groups "Heroku" group. > > To post to this group, send email to heroku@googlegroups.com (mailto:heroku@googlegroups.com). > > To unsubscribe from this group, send email to heroku+unsubscribe@googlegroups.com (mailto:heroku%2Bunsubscribe@googlegroups.com). > > For more options, visit this group at http://groups.google.com/group/heroku?hl=en.
> -- > You received this message because you are subscribed to the Google Groups "Heroku" group. > To post to this group, send email to heroku@googlegroups.com (mailto:heroku@googlegroups.com). > To unsubscribe from this group, send email to heroku+unsubscribe@googlegroups.com (mailto:heroku+unsubscribe@googlegroups.com). > For more options, visit this group at http://groups.google.com/group/heroku?hl=en.
> They can 'ping' your site every hour thereby keeping it awake. It's
> free for one site.
> On Feb 21, 8:38 am, Sejensen <sejen...@gmail.com> wrote:
> > Hi Neil and others
> > These recommandations for Rails apps looks very good,
> > But in my case i use Scala and the netty server and spin-up time is
> > not good at all. And I am not sure it can be tuned so much as the
> > sinatra case when the platform is java.
> > Do anyone have any suggestions for my combo?
> > Kind regards
> > Steffen
> > On Feb 20, 11:15 pm, Neil Middleton <neil.middle...@gmail.com> wrote:
> > > To be brutally honest, there is no hard and fast rule, but I have > found the following:
> > > - Rails is really fast at slowing down, so ensure that your > dependencies and initialisers are at an absolute minimum.
> > > - Keep things small and lightweight, break your application up into > lots of smaller applications. Sinatra, for instance, spins up almost > instantly on Heroku and can take you a fair distance in terms of simple > functionality
> > > - Generally speaking, the newer the Rails version, the quicker the > spinup time, especially if you're not using ActiveRecord
> > > - If you're feeling beta-y, try running on Ruby 1.9.3, this makes a > massive difference to Rails.
> > > Other than this, I'm not really aware of any silver bullet, but I do > know that lots of smaller more focused applications spin up better than one > monolithic Rails app, which is why Unix is so good I guess.
> > > -Neil
> > > On Monday, 20 February 2012 at 21:56, Ed Jones wrote:
> > > > Neil said:
> > > > > I host loads of apps on 1 web dyno and just make sure that the > spin up time is short enough that it's not a problem.
> > > > Neil and others, thanks.
> > > > I just launched a new (beta) app, and the response time is just
> > > > horrible.
> > > > Could you pass a few tips on how to "make sure that the spin up time
> > > > is short enough"?
> > > > Thanks!!!
> > > > ed
> > > > On Feb 17, 7:48 am, Neil Middleton <neil.middle...@gmail.com (
> http://gmail.com)> wrote:
> > > > > I'm confused here.
> > > > > The 'starter' package is only ~$35/mo which isn't exactly > monumentally expensive. Are you suggesting something between that and free?
> > > > > What you're suggesting sounds like your charged by the CPU cycle > rather than the hours?
> > > > > To be brutally honest, I host loads of apps on 1 web dyno and just > make sure that the spin up time is short enough that it's not a problem. > If I ever need to run more than 1 web dyno it's generally because the > traffic levels require it, in which case $35 becomes less of a problem.
> > > > > Personally, I think that having a single dyno, which can still > serve hundreds of thousands of requests a day /for free/ is a pretty good > deal. I'm happy to pay $35 to double it.
> > > > > Neil
> > > > > On Friday, 17 February 2012 at 12:44, Nick wrote:
> > > > > > Peter,
> > > > > > I take your points well. I don't mean to try and 'do one over' on
> > > > > > Heroku. I appreciate the service you offer very much. My thinking
> > > > > > behind it was that you would never exceed the 450 hours of dyno > time
> > > > > > allocated to each app so there wouldn't be a problem and if you > did
> > > > > > you would be charged anyway.
> > > > > > Is there a paid for solution from Heroku to achieve the same > result?
> > > > > > The cost jump between 1 free dyno and paying for a dyno is quite > large
> > > > > > for small applications. So perhaps you could offer a $10 package > which
> > > > > > essentialy works the same way? If i'm honest I don't feel I pay > Heroku
> > > > > > enough but I have too many small apps (10 or so) to pay for each > one
> > > > > > to have a dedicated dyno.
> > > > > > ?
> > > > > > On Feb 16, 8:39 pm, Peter van Hardenberg <p...@heroku.com (
> http://heroku.com)> wrote:
> > > > > > > As a database guy at Heroku, I'm not one to speak > authoritatively on
> > > > > > > this, so please take this as the personal thoughts of someone > and not
> > > > > > > an official statement.
> > > > > > > We idle apps in order to avoid having to charge for them. The > more
> > > > > > > people who prevent this behaviour, the more expensive our > "free" apps
> > > > > > > become to run, and the more likely we are to have to change our
> > > > > > > policies about what we can offer in a free app.
> > > > > > > While I admire the ingenuity in this post, I would suggest > that you
> > > > > > > reduce the amount of time your application takes to boot, or > simply
> > > > > > > accept that a few seconds of lag on the first request after a > period
> > > > > > > of idleness is a reasonable trade-off for free web hosting.
> > > > > > > Peter
> > > > > > --
> > > > > > You received this message because you are subscribed to the > Google Groups "Heroku" group.
> > > > > > To post to this group, send email to heroku@googlegroups.com(mailto:
> heroku@googlegroups.com).
> > > > > > To unsubscribe from this group, send email to > heroku+unsubscribe@googlegroups.com (mailto:
> heroku+unsubscribe@googlegroups.com).
> > > > > > For more options, visit this group athttp://
> groups.google.com/group/heroku?hl=en.
> > > > --
> > > > You received this message because you are subscribed to the Google > Groups "Heroku" group.
> > > > To post to this group, send email to heroku@googlegroups.com(mailto:
> heroku@googlegroups.com).
> > > > To unsubscribe from this group, send email to > heroku+unsubscribe@googlegroups.com (mailto:
> heroku+unsubscribe@googlegroups.com).
> > > > For more options, visit this group athttp://
> groups.google.com/group/heroku?hl=en.
> --
> You received this message because you are subscribed to the Google Groups > "Heroku" group.
> To post to this group, send email to heroku@googlegroups.com.
> To unsubscribe from this group, send email to > heroku+unsubscribe@googlegroups.com.
> For more options, visit this group at > http://groups.google.com/group/heroku?hl=en.
> -- > You received this message because you are subscribed to the Google Groups > "Heroku" group.
> To post to this group, send email to heroku@googlegroups.com.
> To unsubscribe from this group, send email to > heroku+unsubscribe@googlegroups.com.
> For more options, visit this group at > http://groups.google.com/group/heroku?hl=en.
I would answer your question by yes: something between free and 36$ for less resources.
There're plenty of hosting options for 5-10$/month. In another market (.NET), compare http://www.winhost.com/ with 5$/mo dynamic up that's up all the time. I have an old blog running there for 10$/mo, great. I would love to rewrite it in Ruby and push it to Heroku, but price is certainly not competitive for something small.
On Friday, February 17, 2012 7:48:46 AM UTC-5, Neil Middleton wrote:
> I'm confused here.
> The 'starter' package is only ~$35/mo which isn't exactly monumentally > expensive. Are you suggesting something between that and free? > What you're suggesting sounds like your charged by the CPU cycle rather > than the hours?
> To be brutally honest, I host loads of apps on 1 web dyno and just make > sure that the spin up time is short enough that it's not a problem. If I > ever need to run more than 1 web dyno it's generally because the traffic > levels require it, in which case $35 becomes less of a problem.
> Personally, I think that having a single dyno, which can still serve > hundreds of thousands of requests a day /for free/ is a pretty good deal. > I'm happy to pay $35 to double it.
> Neil
> On Friday, 17 February 2012 at 12:44, Nick wrote:
> Peter,
> I take your points well. I don't mean to try and 'do one over' on > Heroku. I appreciate the service you offer very much. My thinking > behind it was that you would never exceed the 450 hours of dyno time > allocated to each app so there wouldn't be a problem and if you did > you would be charged anyway.
> Is there a paid for solution from Heroku to achieve the same result? > The cost jump between 1 free dyno and paying for a dyno is quite large > for small applications. So perhaps you could offer a $10 package which > essentialy works the same way? If i'm honest I don't feel I pay Heroku > enough but I have too many small apps (10 or so) to pay for each one > to have a dedicated dyno.
> ? > On Feb 16, 8:39 pm, Peter van Hardenberg <p...@heroku.com> wrote:
> As a database guy at Heroku, I'm not one to speak authoritatively on > this, so please take this as the personal thoughts of someone and not > an official statement.
> We idle apps in order to avoid having to charge for them. The more > people who prevent this behaviour, the more expensive our "free" apps > become to run, and the more likely we are to have to change our > policies about what we can offer in a free app.
> While I admire the ingenuity in this post, I would suggest that you > reduce the amount of time your application takes to boot, or simply > accept that a few seconds of lag on the first request after a period > of idleness is a reasonable trade-off for free web hosting.
> Peter
> -- > You received this message because you are subscribed to the Google Groups > "Heroku" group. > To post to this group, send email to heroku@googlegroups.com. > To unsubscribe from this group, send email to > heroku+unsubscribe@googlegroups.com. > For more options, visit this group at > http://groups.google.com/group/heroku?hl=en.
On Friday, February 17, 2012 1:47:39 PM UTC-5, Kevin wrote:
> I would like to +1 the idea to come up with a way to get an always > responsive app on Heroku for a reasonable price.
> The current free plan makes the site so unresponsive on the first request > that it's really only good for testing and toy sites, the next level for > $36/month and 2 permanent dynos is a big step away for many people. I think > there is a sweet spot in between there, in which Heroku could reach an > monetize a large market segment.
> Just my 2 cents. :)
> Kevin
> On Feb 17, 2012, at 10:27 AM, Daniel Huckstep <darkh...@darkhelmetlive.com<javascript:>> > wrote:
> I thought the Cedar dynos didn't turn off when idle. Can't remember where > I got that idea from though, so I'm probably entirely wrong.
> But there's also http://devcenter.heroku.com/articles/dyno-idling which > suggest that you could scale your web dynos to zero (on Cedar) and use a > different process name to run the webserver (I'm assuming this would work, > haven't tried it). This is because the other processes (non web) could be > used for other fun things, like running redis (there was an example in a > presentation that showed some TCP routing) which you don't want to turn off.
> Also, in your post you say 450 hours, which is incorrect. It's 750.
> -- > You received this message because you are subscribed to the Google Groups > "Heroku" group. > To view this discussion on the web visit > https://groups.google.com/d/msg/heroku/-/h5MWNmpn3JIJ. > To post to this group, send email to her...@googlegroups.com <javascript:> > . > To unsubscribe from this group, send email to > heroku+un...@googlegroups.com <javascript:>. > For more options, visit this group at > http://groups.google.com/group/heroku?hl=en.
