Amazon Announces Free Monitoring for Amazon EC2 Instances
aws.amazon.comI've been a Rackspace Cloud customer for years, and happily so, but it's this constant drumbeat of innovation and forward momentum at Amazon that makes me seriously reconsider my choice.
I really want to see Rackspace step up its game over the next year, if for no other reason than to keep Amazon on its toes.
Let us know when/if Rackspace decides to step up to the plate (those publicly traded bastards).
What will be the impact of this on third party monitoring apps like CloudKick.?May be they have to provide more value addition on other areas of cloud computing like custom building of servers , automating sys-admin etc.
CloudKick is nicer, but then free is free. I use CloudKick for my stuff and I don't think I'll stop. Some customers however need to cut their costs for many-Ec2 instance deployments, so the free monitoring is good.
I just checked most of the large instances my customer runs, and noticed some are very underutilized; wonder if free monitoring might reduce Amazon's revenue?
Their developers will have more time to write pithy responses on HN?
Better info in the email they sent out:
Dear Amazon EC2 Customer,
We're excited to let you know that as of today, all Amazon EC2 instances come with free Basic Monitoring metrics from our Monitoring service, Amazon CloudWatch. You don't need to do anything to make this happen. It's there for you to use. Simply sign in to the AWS Management Console and select one of your active instances. You will immediately be able to view graphs and track performance on metrics such as CPU utilization, disk reads and writes, and network traffic.
Basic Monitoring for Amazon EC2 provides metric data on instance performance at five-minute frequency. Customers can optionally choose to enable Detailed Monitoring, which provides metric data at one-minute frequency for an additional $0.015 per instance-hour.
In addition, starting today you can now set alarms for any metric that Amazon CloudWatch monitors. You can configure these alarms via API call to send notifications or initiate Auto Scaling actions when metrics cross certain thresholds. Alarm pricing starts at $0.10 per alarm per month.
All the features mentioned above are available immediately in all regions (US-N. Virginia, US-N. California, EU-Ireland, APAC-Singapore), and we invite you to try them today! Learn more about Amazon CloudWatch at: aws.amazon.com/cloudwatch.
Sincerely,
The Amazon Web Services Team
P.S. Amazon CloudWatch monitoring remains free of charge for Amazon EBS volumes, Elastic Load Balancers, and Amazon RDS DB instances.
Wow! Amazon just keeps getting better and better (and cheaper and cheaper). Hmmm...what are they up to?
It is a race to the bottom. Amazon is attempting to commoditize all aspects of their Cloud offerings.
If something becomes a standard feature, a competitor or service cannot pick up the slack and use it as a differentiating feature.
I agree. This is not the first time that Amazon have targeted the companies that can/could be used to add value to their offerings.
There was a company (cannot remember the same sorry) that provided MySQL in amazons cloud before they launched their RDS service. And an much better service it was too. The startup did not ask you to give them a 4 hour window where they can bring down your DB. But still it was an Amazon service so the brand awareness was there.
This seems like a direct attack on the likes of cloudkick. I do not understand the motivation to be honest. They were making some money from their own monitoring and not they are making none, but they are pushing companies like cloudkick away from survival and innovating more.
While the free micro instances offer was obviously a push to put the boot into Rackspace and attempt to take the remaining % of the IaaS market this I understand less.
It seems Amazon have ignored the old rules where in a new market there is less direct competition as there is plenty to go around and have gone straight for total market domination.
Disclosure: My project has(had) as a highlight feature IaaS monitoring, metrics and alerting. So if the above sounds a little unreasoned please forgive me while I contemplate a pivot. :-(
It was FathomDB (I worked there). FathomDB is still going actually, but I'd basically warn anyone away from building a convenience service on top of Amazon's stack - they're using the Microsoft playbook.
Hank Rearden disagrees.
perhaps they are trying to counteract the inevitable backlash from the wikileaks story?
Somehow I doubt they had this plan "on standby" in case of some controversy.
On topic - This combined with the free year of a micro instance will allow you to put a proof of concept together for only the cost of your time.
Strictly speaking, the monitoring isn't essential to put a proof of concept together. It's nice, yes, but not essential.
a direct link to the post being referenced: http://aws.amazon.com/about-aws/whats-new/2010/12/03/announc...
This definitely sweetens the deal. AMZN's monitoring facilities have been expensive and not really all that good.
I agree with you, it very expensive for small startups not taken off yet. For example Cloudkicks charges $99/month for their entry level offering. May be their target customers are startups/companies which are taken off (profitable, VC funded). Also as suggested by other commenter (troels) , monitoring may be not essential for all startups at the early stage.
I've often been unhappy with off-the-shelf monitoring systems, because every major site I've worked on has had some unique kind of problem.
For instance, there was one site where user abuse was the real problem... Bad enough that I built something that detected possible abusive behavior and would beep my pager.
For another project we had about 20 geographically distributed mirror sites, and we had to monitor network connections to all of them and make sure they were all alive and staying synchronized.
Right now I've got a site where the caching system screws up periodically and then I start getting 500 errors. Sooner or later I'm going to really fix the problem, in the short term what I really need is something that gets in my face whenever the 500 error rate spikes.
Cop up, who's startup just died ?
I hope it was more like a "startup idea" though.
Amazon has had monitoring for their instances for a year now. The only change is in the way that they price it.
It used to be expensive enough that you wouldn't leave it on at all times, but would turn it on to diagnose something or check up on one of your boxes, then turn it back off when you were done. Now you get a baseline low-fi version for free.
I think they just realized that nobody was paying for this service, and that it really didn't cost them anything to provide it. So they just made it free.
As far as Startup Killing goes, I doubt this will hurt any "Cloud Monitoring" startups significantly. If you look at the 5 basic charts they give you, you can see it's in line with their stand on reporting across all their services: They give you the basics, but leave plenty of room for somebody else to step in and deliver something better.
After the introduction of micro instances, cloudwatch costs were simply unjustified. Way to go Amazon – best news I've heard this week!
Kudos! But this should have been done long ago.
Free monitoring... right, yes, I suppose you get monitoring by the NSA for free :)
If you are not a US citizen, of course.
If you are a US citizen, there's also an agency for that...
NSA, CIA and FBI "free monitoring" is not free - the billing is via another 3-letter agency: IRS ;)
Free Monitoring, it's like when they kicked out Wikileaks from their cloud? Nice feature.