ec2baseAmazon EC2 by Robot and Me |
|
http://www.mnxsolutions.com/blog/linux/amazon-ec2-benchmark-pystone.html
The pricing model for Amazon EC2 looks attractive from the surface, but when you get down to it — monthly pricing for similar performing hardware can be much cheaper at a dedicated server provider like Softlayer.com. The use of EC2 is climbing, but I am concerned that many of the current uses of EC2 are better served using leased hardware.
http://alan.blog-city.com/has_amazon_ec2_become_over_subscribed.htm
One of our aw2.0 portfolio companies, has been a long term user of Amazon EC2 running a sizable 24×7 of core instances with a number of instances going up and down as scale demands it. Our monthly bill gets us the dubious honor of a first point of contact with an Amazon Account Manager (not that that has been much use). We’ve pushed the limits of many of their services and continue to do so.
After 3 years of production usage what we can tell you is this .. Amazon do have a breaking point.
Visual evidence of Amazon EC2 network issues
https://www.cloudkick.com/blog/2010/jan/12/visual-ec2-latency/
http://news.ycombinator.com/item?id=1048873
http://news.ycombinator.com/item?id=1048694
As my mother would say; “misery enjoys company”
http://alan.blog-city.com/amazon_ec2_latency_the_pretty_graphs.htm
Amazon: We Don’t Have Cloud Capacity Issues
PLEASE SAY IT AIN’T SO, CLOUD OVER-SUBSCRIPTION?
http://vuksan.com/blog/2009/12/04/cloud-cartography-load-co-residence-detection/
I was actually quite surprised at the magnitude of degradation. I’d say this may be even a more successful co-residence detection attack than network probing since you could generate legitimate HTTP traffic to a site of interest (or a node of interest), throw tons of load at it and see if you notice response degradation.
What Clouds Can Learn From Airlines
http://www.infoq.com/news/2010/01/ec2-oversubscribed
There have been various reports from the community of Amazon EC2 users, that their instances are suffering poor performance, as the result of high internal network latency. This has led to speculations that Amazon’s Cloud might be getting oversubscribed.
http://tech.slashdot.org/story/10/01/15/1350213/Amazon-EC2-May-Be-Experiencing-Growing-Pains
“Some developers using Amazon EC2 are wondering aloud whether the popularity of the cloud computing service is beginning to affect its performance. Amazon this week denied speculation that it was experiencing capacity problems after a veteran developer reported performance issues and suggested that EC2 might be oversubscribed. Meanwhile, a cloud monitoring service published charts showing increased latency on EC2 in recent weeks. The reports follow an incident over the holidays in which a DDoS on a DNS provider slowed Amazon’s retail and cloud operations.”
I have observed “noisy neighbor” effects.
If you are using HBase EC2 scripts, which run HBase region servers on all of the slaves colocated with tasktrackers and user tasks, I do not recommend using other than c1.xlarge instances.
Our scripts use c1.medium instances for the separate Zookeeper quorum ensemble as they need fewer resources in terms of RAM but are still sensitive to io and cpu scheduling latencies.
http://www.jonathanboutelle.com/mt/archives/2010/01/hadoophackday_w.html
-Hadoop is very resource-intensive! We started out using 1-node clusters to run our jobs against small subsets of data. Very quickly teams started upgrading to 5-node clusters due to the amount of time they were having to wait for results. Final runs against full data sets were powered by 10-node clusters of “medium” ec2 servers. You have no choice but to use cloud computing for these kinds of jobs, because it seems to me that production use could easily require 100s of nodes, and no one would want to buy that many servers for machines that they only use one hour a day.
Teich also said that all of Heroku’s m2.2xlarge instances were running in a single availability zone, which was a mistake. He stressed that Heroku had failover built in already — if 21 instances had failed instead of 22, or if it had spread instances across several zones, “we wouldn’t be talking [about the outage],” he said.
Nevertheless, on Friday, January 2, every m2.2xlarge instance in that availability zone suddenly vanished, despite all other types of EC2 instances running as normal. That’s unheard of in traditional hosting. It would be like every server with a given amount of RAM suddenly shutting down, regardless of operating system, age, brand, hardware or location in the data center, with no effect on its neighbors.
“For us, there’s the stuff you plan for and then there’s the stuff you don’t even know about,” Teich said.
An event like this was an “unknown unknown” that nobody planned for because nobody imagined it. He chalked it up to the learning process and pointed out that everybody in Amazon Web Services was flying by the seat of the pants at least part of the time.
HN thread:
http://news.ycombinator.com/item?id=1046500
http://moussadao.com/2009/12/30/amazon-ec2-pricings-demystified/
Amazon Elastic Compute Cloud (Amazon EC2) is the leading provider for cloud based computes resources. Amazon EC2 allows users to provision compute resources on-demand. With On-demand Instances, Amazon EC2 does not require any commitment, sign-up fees, and upfront payments. Amazon Web Services (AWS) uses a monthly billing cycle; therefore users get charged at end of each month for their EC2 usages.
Online backups for the truly paranoid
Tarsnap is a secure online backup service for BSD, Linux, OS X, Solaris, Cygwin, and can probably be compiled on many other UNIX-like operating systems. The tarsnap client code provides a flexible and powerful command-line interface which can be used directly or via shell scripts.
At the present time, tarsnap does not support Windows (except via Cygwin) and does not have a graphical user interface.
http://aws.typepad.com/aws/2009/12/thirdparty-aws-tracking-sites.html
A couple of really cool third-party AWS tracking sites have sprung up lately. Some of these sites make use of AWS data directly and others measure it using their own proprietary methodologies. I don’t have any special insight in to the design or operation of these sites, but at first glance they appear to be reasonably accurate.
http://zuzara.com/blog/2009/11/01/tpcc-mysql-rough-benchmark-for-amazon-rds/
I tried to do tpcc-mysql benchmark for Amazon RDS. Before do that, did the same test for EC2 small instance.
This is a pretty rough benchmark, but I can say EC2 small instance and RDS small instance have the same performances as CPU and memory are the same spec. RDS is about 30% expensive. (EC2=$0.085, RDS=$0.11/h)
http://img.skitch.com/20091212-8mfp3jihiuts3diifk5e62py56.jpg
Totally. Need to know what u’re doing, man. And how.
http://www.allthingsdistributed.com/2009/12/amazon_ec2_spot_instances.html
Spot instances. Yeah.
# On-Demand Instances – On-Demand Instances let you pay for compute capacity by the hour with no long-term commitments or upfront payments. You can increase or decrease your compute capacity depending on the demands of your application and only pay the specified hourly rate for the instances you use. These instances are used mostly for short term workloads and for workloads with unpredictable resource demand characteristics.
# Reserved Instances – Reserved Instances let you make a low, one-time, upfront payment for an instance, reserve it for a one or three year term, and pay a significantly lower rate for each hour you run that instance. You are assured that your Reserved Instance will always be available in the Availability Zone in which you purchased it. These instances are used for longer running workloads with predictable resource demands.
# Spot Instances – Spot Instances allow you to specify the maximum hourly price that you are willing to pay to run a particular instance type. We set a Spot Price for each instance type in each region, which is the price all customers will pay to run a Spot Instance for that given hour. The Spot Price fluctuates based on supply and demand for instances, but customers will never pay more than the maximum price they have specified. These instances are used for workloads with flexible completion times.
http://news.ycombinator.com/item?id=984691
Using EC2 for anything but overflow is silly. The costs are extraordinarily high if you are doing decent volume ( read- if you have more than 3-5 servers )
Bandwidth is ridiculously expensive with them. You can get 3-4x cheaper per megabit going dedicated.
Servers are crazy expensive. Compare the most powerful machine they have vs something on 10tb / gigenet / theplanet for the price. You will definitely end up with a more powerful machine for half the price on either.
The only real advantage to using ec2 is the hourly billing. Make its perfect for overflow, but thats about it. I read about alot of startups that use ec2 for things like webservers or other servers that have 100% reliance.
http://aws.typepad.com/aws/2009/12/the-economics-of-aws.html
For the past several years, many people have claimed that cloud computing can reduce a company’s costs, improve cash flow, reduce risks, and maximize revenue opportunities. Until now, prospective customers have had to do a lot of leg work to compare the costs of a flexible solution based on cloud computing to a more traditional static model. Doing a genuine “apples to apples” comparison turns out to be complex — it is easy to neglect internal costs which are hidden away as “overhead”.
