I'm using URL Fetch Java API
http://code.google.com/appengine/docs/java/urlfetch/overview.html#Fetching_URLs_with_java_net
to get information from graph.facebook.com, however, i keep getting
the error:
java.lang.IllegalArgumentException
at java.net.URI.create(URI.java:842)
...
Caused
I am using JPA, but that's irrelevant.
--
You received this message because you are subscribed to the Google Groups
Google App Engine for Java group.
To post to this group, send email to google-appengine-java@googlegroups.com.
To unsubscribe from this group, send email to
It has to do with limits that are set for the API your using. If you need
longer transit times, I would use a task or something like the the Channel
API. Or split up the traffic.
Brandon Donnelson
http://gwt-examples.googlecode.com
--
You received this message because you are subscribed to
I've been able to lower the impact of these datastore
DeadlineExceededException by setting a global datastore deadline using :
http://code.google.com/appengine/docs/java/javadoc/com/google/appengine/api/datastore/DatastoreServiceConfig.html
I just catch the ApiProxy$ApiDeadlineExceededException
I just found something rather strange :
After setting a 10 seconds deadline on my datastore service and setting a
page reload when deadline exceptions are thrown, I got caught in an infinite
loop : pages were reloading for several minutes until a background task
using a datastore service
Hi,
Can anyone please help me ?
My app works fine in my computer, but after uploaded it to gae server, i
received an error.
* See my log from appengine logs.
[19/May/2011:18:09:41 -0700] GET /agendaxyzdms HTTP/1.1 500 0 - Mozilla/5.0
(Macintosh; Intel Mac OS X 10.6; rv:2.0.1) Gecko/20100101
Hi my name is Daniel, I'm new to coding, so everything literally feels like
a foreign language at this point, but I decided to start learning because
I'm interested in developing a news app for smartphones and tablets.
Anyway i'm using the Java version of the GAE SDK and was wondering how
Amount of concurrency / instance is critical for deciding whether to
use AppEngine as a platform.
Although AppEngine has it's disadvantages (vendor lock-in, restricted
architecture, etc),
the key reason to choose AppEngine is scalability.
The datastore scales.
Now we need to focus the other
Q: What is the time granularity of the instance pricing? ie if I
have an
instance up for 5 minutes, what am I charged, $0.08 / 60*5?
A: Instances are charged for their uptime and until they are idle for
15
minutes (when the scheduler takes them down). So if you have an on-
demand
Instance only
On Wed, May 18, 2011 at 5:03 PM, Brandon Wirtz drak...@digerat.com wrote:
Put some ads on it and it should get to the $9 a month. Put a donate button
on it.
It's never that simple, and certainly isn't in this case.
That said, Google certainly doesn't owe anyone a free service layer,
nor a
Anything getting enough traffic to cross the free barrier should be able to
have a revenue stream. Especially if you thought it was going to be a
commercial venture. And this is cold hearted but if you have loyal users
who are using the app that much and none of them will pony up $.50 a
month...
What's driving the requirements? Are you building a consumer app with
a clear vision of what the user experience should be? Top down. Or
are you starting with an API and trying to figure out how to sanely
present that to an audience? Bottom up.
It sounds like your ZoeOS was the later and
I still have 2.5x as much billed usage for HR Storage as I have reported by
the datastore admin.
I don't have any defined indexes. But I Vacuumed (more than once) and that
didn't reduce the number.
It's only 10 cents a day on most my apps, but that's $3 a month. (per app)
which isn't a
You presume that my app has a web front end. Most of the users of
this particular app are iPhone clients, and the developer responsible
for that code has moved on to other projects.
...and actually, there does seem to be a vast chasm between successful
apps and unsuccessful ones. I'm not
Pay pal will do $5 a year annual subscriptions.
It has a front end... if there is a front end you can stick ads there...
Yeah iphone sucks for ad revenueHave you considered that Other mobile
OS, the one from Google?
;-)
-Original Message-
From: google-appengine@googlegroups.com
1. What I pay for appengine is not a function of my revenue. Whether or not
I will put adsense in my app is beyond the scope of this thread.
2. Exceeding free quota doesn't happen in as big step as $0 - $9. Mine is
an app that deals with images. I would cross free quota only in terms of
space
Greg,
Stressing a point made here by Mike Wesner: is the memory limit on
(frontend) instances being cut down to 128mb? As far as I know, there
was no formal documentation of an instance's memory limit before, but
the soft limit error was usually triggered at around 180mb. Should
be expect this
Thanks, it is really helpful!
On May 17, 7:03 pm, Robert Kluin robert.kl...@gmail.com wrote:
Hi Uri,
It looks like the remote_api doesn't support lease_tasks. You
should be able to enable leasing and deleting tasks by adding the
following lines to the taskqueue service list:
Hi,
Anyone has any clues on this?
Thanks for the help
Chimbu
On May 16, 1:43 pm, Chimbu Aravind chimbu.li...@gmail.com wrote:
Hi,
Need some help in uploading a file from Widows Mobile 6.5 device to
GAE blobstore
In the GAE server, I have the following python code; (Its still
crude.. )
How about new Google's In App payment system? Looks promising.
Thing is, people want too much for free. $9 is less than you pay for
internet at home, hosting anything at Google is cheaper than putting
old PC in the basement :)
I hope that going out of preview will mean even more tools and toys
I still get an exception:
[Remote API Server] [handler.py:308] ERROR Exception while handling
service_name: taskqueue
method: QueryAndOwnTasks
request: \n\011workitems\021\000\000\000\000\000 \254@\030\024
Traceback (most recent call last):
File C:\Program
While we may have varied from original topic of the thread, it is very
relevant.
For $9 who would you run instead? How much hosting would that buy you at
DreamHost? Or Amazon Or Rackspace. Same question at $50, and $500
Most solutions have sweet spots in pricing, and as one of the larger
I get 10 GB of EBS space for free on AWS. It is insane to think it is
alright to pay $9 to be able to consume more than 5GB.
For $10, you get 100 GB space on AWS. And for $10, you get ~59GB space on
appengine. Now, how does it make any sense to pay minimum of $9 to be able
to cross the 5GB free
While I'm at it. My Bytes Sent to Bytes Received ratio is about 10:1, but my
Outgoing to Incoming bandwidth ratio is like 10:9 this seems wrong as well.
I haven't set up a test case yet, but just wondering if more than just the
contents of my URLFetches counts in my Incoming Bandwidth. Like
It is insulting to suggest those opposed to $9/month are freeloaders and
expect google to do charity. The issue is not about having to pay 4x-5x. The
issue is about fairness.
For those who were already paying more than $4 a month, $9/month is a non
issue. For everybody else it is an issue. Nobody
Hi Greg after applying the patch I get this exception, please help
tasks = q.lease_tasks(3600, 20)
File C:\Program Files\Google\google_appengine\google\appengine\api
\taskqueue\taskqueue.py, line 1041, in lease_tasks
response)
File C:\Program
Hi
One of the disadvantages with GAE is that there is no CMS like Drupal och
Joomla to easily deploy new large projects. Is there a CMS for GAE in the
making? I found myself creating the basic functions of a CMS from scratch ie
adding, viewing, editing and deleting articles in a categorized
Hi,
I am working on a CMS, called claymus http://claymus.googlecode.com which
is under-development. I am planning to release a pre-release version asap
and a stable version by end of this year.
--
Prashant
On Thu, May 19, 2011 at 3:44 PM, Niklas Rosencrantz nikla...@gmail.comwrote:
Hi
One
On May 19, 8:34 am, Jeff Schnitzer j...@infohazard.org wrote:
What's driving the requirements? Are you building a consumer app with
a clear vision of what the user experience should be?
Pretty much yes - so top down then.
--
You received this message because you are subscribed to the
Thanks for everyone's input - it helped.
On May 19, 11:48 am, Paul C. Meehan paulm...@gmail.com wrote:
On May 19, 8:34 am, Jeff Schnitzer j...@infohazard.org wrote:
What's driving the requirements? Are you building a consumer app with
a clear vision of what the user experience should be?
Now, how does it make any sense to pay minimum of $9 to be able to
cross the 5GB free limit?
For me, it makes no sense.
I'm working with GAE since the beginning, and the mojo was pay for
what you use/need.
It's not only a question of $9, but they change the rules with no real
explanation.
I use
On Thu, May 19, 2011 at 10:10 AM, Brandon Wirtz drak...@digerat.com wrote:
For $9 who would you run instead? How much hosting would that buy you at
DreamHost? Or Amazon Or Rackspace. Same question at $50, and $500
At Amazon, $500 would buy $500 worth of hosting, $50 would buy $50
worth of
On Thu, May 19, 2011 at 8:36 AM, Brandon Wirtz drak...@digerat.com wrote:
I still have 2.5x as much billed usage for HR Storage as I have reported by
the datastore admin.
...
It’s only 10 cents a day on most my apps, but that’s $3 a month… (per app)
which isn’t a big deal yet, but it adds
If you use blobstore.create_upload_url() the Blobstore API file will
create the blob and call your
blobstore_handlers.BlobstoreUploadHandler once the upload is
completed.
The using self.get_uploads('file') in your handler you will be able to
get the BlobInfo corresponding to your file in the blob
Agreed.
Why should I pay for idle time, if it's idle?
That *does not make any sense*!
--
You received this message because you are subscribed to the Google Groups
Google App Engine group.
To post to this group, send email to google-appengine@googlegroups.com.
To unsubscribe from this group,
One reason is, its there ready to serve a new request.
If there is no instance running, one will have to be spun up, adding
(possibly significant) delay to the next request.
With 15 minutes Google appear to be offering a compromise. It's
probably not totally arbitrary number, but based on
You are using it because the instance is consuming memory. I think the idle
time should become lesser compared to the number instances.
1 (primary) 15 minutes idle or always up.
2 (secondary instances) 5 minutes or less.
Also if the scheduler starts a secondary instance which during its
On Thu, May 19, 2011 at 2:52 PM, Barry Hunter barrybhun...@gmail.com wrote:
With 15 minutes Google appear to be offering a compromise.
This is the problem: they haven't fixed anything they've merely
shifted the burden by compromising.
Under the current scheme, as Greg explained, full-fat Java
Thank you Prashant for informing about this exciting project that I
will follow.
Niklas
On Thu, May 19, 2011 at 12:44 PM, Prashant antsh...@gmail.com wrote:
Hi,
I am working on a CMS, called claymus which is under-development. I am
planning to release a pre-release version asap and a stable
Well yes, they have shifted the 'burden' but I see it in a different way.
The 'old' way, pretty much promoted low cpu use, even if that came at
the expense of latency. The slow requests, would necessitate lots of
instances - costing Google.
The 'new' way promotes keeping your latency down. Quick
I get that hashlib error a lot. It usually happens when a file that imports
hashlib gets recompiled while dev_appserver is running. Restarting
dev_appserver gets around the problem, and I've never experienced it while
running in production.
--
You received this message because you are
Nono..It's not the hashlib error.
(I see the hashlib warning all the time too.)
It's the error messages below that are new.
On 19 May 2011 23:20, Calvin calvin.r...@gmail.com wrote:
I get that hashlib error a lot. It usually happens when a file that
imports hashlib gets recompiled while
Oh, sorry. No I've never seen the EOFError exception before. I would check
any recently modified files for stray ctrl characters (ctrl-d?).
--
You received this message because you are subscribed to the Google Groups
Google App Engine group.
To post to this group, send email to
Hmm...very strange.
When I do the following, it's fine:
c:\program files (x86)\google\google_appengine\appcfg.py update
c:\appengine\keikaku
On 19 May 2011 23:36, Calvin calvin.r...@gmail.com wrote:
Oh, sorry. No I've never seen the EOFError exception before. I would
check any recently
Next theory: There could be garbage in your .appcfg_cookies or .appcfg_nag
files. Not sure where they live on windows systems.
--
You received this message because you are subscribed to the Google Groups
Google App Engine group.
To post to this group, send email to
Yes, Yes, Yes.
$9 was the amount of Mt. Dew Per day, not per month. Currently it's across
10 apps, but in the next month it is across 15k domains. Rounding Errors
could pay for another 3 interns at that scale.
Also because the number seems to be so close to 2.5:1 I wonder if this is a
bug.
I looked at creating a CMS on AppEngine, and quickly discovered that the
issue is that most users of a CMS want their plugins. The hill to get over
in terms of install base is formidable when you consider that Java, and
Python are not as commonly spoken languages as PHP, Ruby, and ASP. That's
That totally worked!
Thanks Calvin!
On 19 May 2011 23:59, Calvin calvin.r...@gmail.com wrote:
Next theory: There could be garbage in your .appcfg_cookies or .appcfg_nag
files. Not sure where they live on windows systems.
--
You received this message because you are subscribed to the
As a non Google person and someone was pretty shocked about the price
changes I have to say you're over-reacting.
If your application has a average latency of 250ms (which is quite high
imho) you can get 345,600 hits per day on the free tier, apparently forever.
If you're getting that kind of
Verify that none of your other Versions are getting traffic. I have had a
few issues where in testing I somehow let the Google Bot know about another
version of the site and it got indexed and trafficked and was running up a
bill, (and was a pain to get that moved over to the primary version)
On Thu, May 19, 2011 at 9:43 PM, Kenneth kennet...@aladdinschools.comwrote:
As a non Google person and someone was pretty shocked about the price
changes I have to say you're over-reacting.
If your application has a average latency of 250ms (which is quite high
imho) you can get 345,600 hits
It looks like your CPU went down 2 days ago and you are showing a bill from
3 days ago (usage report for 2011-05-15), wait another day or two and you
should see the drop, if not, let me know.
Thanks!
Greg D'Alesandre
Senior Product Manager, Google App Engine
On Thu, May 19, 2011 at 9:16 AM,
On Thu, May 19, 2011 at 4:17 PM, Barry Hunter barrybhun...@gmail.com wrote:
Well yes, they have shifted the 'burden' but I see it in a different way.
The 'old' way, pretty much promoted low cpu use, even if that came at
the expense of latency. The slow requests, would necessitate lots of
On 19 May 2011 17:31, Stephen sdeasey+gro...@gmail.com wrote:
On Thu, May 19, 2011 at 4:17 PM, Barry Hunter barrybhun...@gmail.com wrote:
Well yes, they have shifted the 'burden' but I see it in a different way.
The 'old' way, pretty much promoted low cpu use, even if that came at
the expense
As far as I know the Blobstore has never been available to free apps and now
they're giving you 5gigs free, seems like a pretty big win to me.
The 50k free database operations is a bit more nebulous since Google are
still scratching their heads over it. With correct memcache usage it really
Hi Brandon,
Different Steve here. Anyway, when I mentioned in a different thread
about indexes, you also have to count your property indexes as well.
For each property that you have indexed (and these wouldn't show up in
your configured indexes) you have two indexes defined: one ascending
and one
SLA + Infinitely scalable is what comes to mind.
It is not free, the ones who use a paid account have to cover the costs of
the free ones as well. It is how their and AWS business model works.
So if you want more for free, then the billing prices go up, not a good
idea..and not fair.
--
You
The 50k free database operations is a bit more nebulous since Google are
still scratching their heads over it. With correct memcache usage it really
should be enough, although that might be a bit of wishful thinking and
obviously depends significantly on your use case.
As reference.
I think I went out of my way to use non-indexable types, but I'll make sure
that is the case. I often store things as type blob to prevent indexing.
But I may have set some things to the wrong type.
You don't think Vacuuming would help with this?
Considering all the other data shown in Quota
Hi,
I've just started using AppEngine and am getting warned that my methods are
using a high amount of CPU and may exceed the quota. A typical log line:
200 332ms 2048cpu_ms 1951api_cpu_ms 0kb
Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/534.36 (KHTML, like Gecko)
Chrome/13.0.767.1
On Thu, May 19, 2011 at 9:58 AM, Barry Hunter barrybhun...@gmail.com wrote:
On 19 May 2011 17:31, Stephen sdeasey+gro...@gmail.com wrote:
No, the way it currently works is if your apps latency 1000ms
(figures of 800-900ms have also been mentioned) it won't scale. The
lower the latency the
On Thu, May 19, 2011 at 10:54 PM, Kenneth kennet...@aladdinschools.comwrote:
As far as I know the Blobstore has never been available to free apps and
now they're giving you 5gigs free, seems like a pretty big win to me.
They have had 2GB for free. You only had to enable billing.
For
On Thu, May 19, 2011 at 7:04 PM, Jeff Schnitzer j...@infohazard.org wrote:
There is a terrific amount of price signaling going on in this change.
It indicates that:
* Memory is precious, CPU time is almost irrelevant.
This is not too surprising; if you look at the rest of the cloud
To be clear, the 10GB of charity storage is only valid for the first 12
months for new customers.
http://aws.amazon.com/free/
--
You received this message because you are subscribed to the Google Groups
Google App Engine group.
To post to this group, send email to
Hey Greg,
Will we be able to adjust the scheduler for each version of our
application? Two use-cases come to mind: 1) different options for
Python / Go / Java versions of an app, and 2) different options for
'front-end' versions serving user traffic and processing tasks. Would
be nice so that
That does seem a high. Does it also say that a new instance was started for
this request? That would make it less surprising.
Are you storing/getting all the entities in one db call? That's the most
straightforward way to speed up requests. There's some overhead to
serializing and
Thanks, Brandon, this is great info!
This plus Stephen's info shows how to get scalability without concurrency:
No, the way it currently works is if your apps latency 1000ms
(figures of 800-900ms have also been mentioned) it won't scale. The
lower the latency the more they scale it. So the
Vacuuming just applies to your configured indexes not the
single-property indexes themselves. There was a post recently about
possibly making it more obvious somewhere in the dataviewer which
properties are indexed and which aren't because people can start off
with something indexed then unindex
Yeah, the people building small-scale apps shouldn't worry about the
instance quota. They should worry about the datastore operations quota.
If, on average, every request queries for 10 entities... your app is limited
to 5,000 requests per day. That's 3-4 requests per minute. Not much. And
if
Q: how big is AppEngine?
A: 4 x tumblr.com:
GAE reached 1 billion page views at the end of last year:
http://www.onebigfluke.com/2010/12/nuff-said.html
and tumblr.com reached 250 M page views this month:
http://staff.tumblr.com/post/5578802048/whoa-yesterday-we-jumped-past-250m-pageviews-a
I'm also getting really erratic behavior right now, high cpu, deadline
exceeded exceptions, 30 second long instance spinups. The status page
still says everything's fine though.
On May 19, 8:35 pm, Calvin calvin.r...@gmail.com wrote:
That does seem a high. Does it also say that a new instance
It still works out much cheaper than appengine.
10GB on AWS would cost: $1
on appengine: $9.85 (which would get 98.5GB on AWS.)
On Thu, May 19, 2011 at 11:54 PM, Calvin calvin.r...@gmail.com wrote:
To be clear, the 10GB of charity storage is only valid for the first 12
months for new
Ah, that is pretty damn cool.
No there wasn't an instance started, this is totally typical for an entire
day of requests.
I've taken a screenshot, http://burtonini.com/temp/appstats.png
Unsurprisingly the bulk of the time is spent in datastore_v3.Put(). So,
first rookie question. Bar has a
yes, it's a google interview question ... for google! ;-)
On Fri, May 20, 2011 at 2:45 AM, Dennis dennisf...@gmail.com wrote:
Q: how big is AppEngine?
A: 4 x tumblr.com:
GAE reached 1 billion page views at the end of last year:
http://www.onebigfluke.com/2010/12/nuff-said.html
and
Aha! A superset. That explains it! I've changed my call to pass 'cp1252'
instead of 'iso-8859-1' and gotten rid of the replace call, and it seems to be
working right now.
Thanks so much!!!
-Joshua
On May 18, 2011, at 11:58 PM, Geoffrey Spear wrote:
I'd guess the original encoding is
On Thu, May 19, 2011 at 11:15 AM, Stephen sdeasey+gro...@gmail.com wrote:
On Thu, May 19, 2011 at 7:04 PM, Jeff Schnitzer j...@infohazard.org wrote:
This is not too surprising; if you look at the rest of the cloud
industry, almost all services are priced by the amount of RAM used and
make no
Last few hours I've been getting random-looking deadline-exceeded
exceptions, 30second plus spinups, error code 104 whatever that
means. Here's a capture of what should be a 50ms request:
http://dl.dropbox.com/u/353497/gaewtf.jpg
Anyone else?
--
You received this message because you are
Yes and happy to see that I'm not alone.
Check these other posts :
https://groups.google.com/d/topic/google-appengine/b2TcgwcW6qs/discussion
https://groups.google.com/d/topic/google-appengine/gSLESWDEJOc/discussion
https://groups.google.com/d/topic/google-appengine-java/6tnyfJLBIXM/discussion
On Thu, May 19, 2011 at 7:55 PM, Jeff Schnitzer j...@infohazard.org wrote:
On Thu, May 19, 2011 at 11:15 AM, Stephen sdeasey+gro...@gmail.com wrote:
On Thu, May 19, 2011 at 7:04 PM, Jeff Schnitzer j...@infohazard.org wrote:
This is not too surprising; if you look at the rest of the cloud
I personally hate it when I have to do double puts, especially double puts
with no other RPC in between. Allocating ids for keys is an RPC call too,
but it would probably end up being faster.
Also, you can create a unique key for an object using uuid.uuid4().
key =
http://stackoverflow.com/questions/6063588/on-google-app-engine-in-python-how-do-you-detect-if-a-jpeg-uses-cmyk-color-space
The Images API in Google App Engine does not appear to handle CMYK
JPEG images properly when applying transforms. On the production
server the resulting image looks washed
But how much would it cost to have multiple EC2 instances serving the
content along with S3 hosting the content? A single micro instance running
for a month would be $14.
So what you could do is host static content in S3 and host the app in App
Engine and save both ways.
--
You received
I had some spots where I wasn't using non-indexed/able data types, hopefully
that will reduce this storage.
I haven't built a test case, do you happen to know if all types write/read
at the same speed assuming they are the same size?
-Brandon
-Original Message-
From:
Depends on the amount of static content. I serve static out of GAE because
it is faster than S3 and about the same price depending on your tier.
From: google-appengine@googlegroups.com
[mailto:google-appengine@googlegroups.com] On Behalf Of Calvin
Sent: Thursday, May 19, 2011 1:19 PM
To:
On Thursday, 19 May 2011 19:42:17 UTC+1, Jeff Schnitzer wrote:
Yeah, the people building small-scale apps shouldn't worry about the
instance quota. They should worry about the datastore operations quota.
Yeah - I'm writing a one-page-webapp (that incidentally rocks on Google
Chrome and
On Fri, May 20, 2011 at 1:48 AM, Calvin calvin.r...@gmail.com wrote:
But how much would it cost to have multiple EC2 instances serving the
content along with S3 hosting the content? A single micro instance running
for a month would be $14.
So what you could do is host static content in S3
Hi!
I set up the cron job and collecting data from internet. I'd like to
display some statistics about the data. It's pretty enough to update
the view once per day.
From my point of view the ideal implementation could be to re-generate
the index.html file once per day. Definitely I do not want
Use Memcache logic (not code is as follows)
Def(get)
MyIndex is None
Try MyIndex from memcache return my index
If myindex is none my index from datastore, add myindex to memcache, return
my index
-Original Message-
From: google-appengine@googlegroups.com
I would assume so but can't answer that definitely. I think it's based
on protobuf but not sure.
On Thu, May 19, 2011 at 1:20 PM, Brandon Wirtz drak...@digerat.com wrote:
I had some spots where I wasn't using non-indexed/able data types, hopefully
that will reduce this storage.
I haven't
You can put generated data to memcache with time parameter for expire your
data and datastore, then read it from memcache. If memcache empty - read
from datastore.
2011/5/20 PF petr.felzm...@gmail.com
Hi!
I set up the cron job and collecting data from internet. I'd like to
display some
On Thu, May 19, 2011 at 1:03 PM, Stephen sdeasey+gro...@gmail.com wrote:
On Thu, May 19, 2011 at 7:55 PM, Jeff Schnitzer j...@infohazard.org wrote:
http://www.linode.com/index.cfm
http://www.rackspace.com/cloud/cloud_hosting_products/servers/pricing/
http://www.slicehost.com/
I have found this in the release notes for the SDK 1.5.0 :
Fixed an issue where the API Deadlines were not being enforced in the SDK.
http://code.google.com/p/googleappengine/wiki/SdkForJavaReleaseNotes
It might be related.
Also found this older article dealing of consistency and deadlines :
@Jeff
I would question any comparisons that ends like If you’re going to
buy a VPS I’d appreciate it if you used my referral page for Linode or
for Slicehost when doing so.
I would prefer a donate button and most probably I would contribute.
Nick Milon
On May 19, 9:55 pm, Jeff Schnitzer
My app's been fine up until about 6 hours ago, so I'm pretty sure that
you won't find anything in the docs or literature that applies to me,
just some sort of infrastructure hiccup. Thanks though!
On May 19, 11:10 pm, Francois Masurel f.masu...@gmail.com wrote:
I have found this in the release
On my side, it happens periodically since 1.5.0 update. I'll keep an eye on
it.
Francois
On Thu, May 19, 2011 at 23:36, tempy fay...@gmail.com wrote:
My app's been fine up until about 6 hours ago, so I'm pretty sure that
you won't find anything in the docs or literature that applies to me,
I've never expected it to be a 'charity'. The first time I looked at GAE in
2008, it had a price tag.
Here is my experience:
1. I looked at GAE and AWS, GAE won hands down because of its lower price
and zero system administrative overhead.
2. I put my business on it seriously.
3. In this three
Lots of talk and fighting here with developers comparing App Engine vs
AWS vs Rackspace vs any_other_VPS_service_on_earth.
There was not such big talk before, coz App Engine looked different
from those products both in terms of pricing as well as features.
Now, thanks to latest news, it managed to
If you're worried about his objectivity, it's worth noting that
Slicehost performed very poorly - worse than EC2 in most tests. Most
likely, he mentions the referral page for Linode and Slicehost because
they are the only providers that offer an affiliate program.
In the absence of any reason to
@Jeff
No bad intentions neither for his objectivity nor your findings.
I just wanted to say that product - comparison publications should not
depend on products/services compared for an income, it is a matter of
principle.
I would be grateful if you could possibly share your own tests
Nick
On
250ms / request means you get MAX 4 request/seconds once it shift to
instance based pricing. Thats what 1 instance can serve. Your per day
concept only applies to CPU based pricing (the current model)
A single entity db.get + db.put can cost you more than 250ms sometimes.
Many people in this
1 - 100 of 112 matches
Mail list logo
