I am running a small server on AWS that the networking and other thing is not that much important. Threfore, hosting an EC2 in any region does not matter to me. I view cost as the biggest factor in choosing the region.\
My server :
t3.small with 2 Gb ram, 2 CPU, 8 Gb storage.
I wonder if anyone has this kind of experience and which region is the cheapest one?
It is likely that an Amazon Lightsail package will be a lower-cost option for you rather than using Amazon EC2.
It offers a fixed-price for compute, storage and data transfer.
Related
I'm trying to train a model using AWS SageMaker notebooks and am disappointed with how slowly the model is training. I think my bottleneck lies with the IOPS speed to the persistent storage (EFS and EBS) my SageMaker notebooks are accessing for the dataset.
First, I tried training on a SageMaker Studio ml.g4dn.xlarge instance, then moved everything over to a SageMaker notebook ml.g4dn.xlarge instance through Jupyter. Even though g4dn.xlarge instances come with a physically wired 125GB SSD, I'm unable to access it because SageMaker Studio automatically creates an EFS store, and SageMaker notebook instances automatically create an EBS store. How could I store my dataset on the 125GB SSD instead of EFS or EBS to speed up the IOPS?
It is clear that there are instances with memory optimised for large amounts of data. In your case, if the dataset is given as input to the model with exactly that size (so there is no upstream preprocessing to lighten this amount of data), you must know that the g4dn is EBS optimised.
The most obvious answer i can think of is to use an S3 bucket
From "Maximum transfer speed between Amazon EC2 and Amazon S3":
Traffic between Amazon EC2 and Amazon S3 can leverage up to 100 Gbps
of bandwidth to VPC endpoints and public IPs in the same region.
Besides being very fast and performant, it is also the best solution in terms of design for all components of your project on AWS. Clearly, it entails different costs and a different architecture, but you will enjoy the maximum speed that the set of AWS services can offer you (and possibly require special configurations for even better performance).
My advice is to follow the AWS guidelines for developing a complex project from scratch: Build, training and deployment of machine learning models.
We run Couchbase in Kubernetes platform in AWS cloud. As per the 'Couchbase on AWS' best practices, it is suggested to use EBS 'gp3' or EBS 'io1' based on the following link. (https://docs.couchbase.com/server/current/cloud/couchbase-cloud-deployment.html#aws-deployment-methods)
But it seems AWS has introduced a new EFS storage type, known as, "Amazon EFS Elastic Throughput" (https://aws.amazon.com/blogs/aws/new-announcing-amazon-efs-elastic-throughput/). It gives much better throughput. Is it suggested to use EFS with Elastic Throughput for Couchbase storage?
Throughput claimed by AWS: Elastic Throughput allows you to drive throughput up to a limit of 3 GiB/s for read operations and 1 GiB/s for write operations per file system in all Regions.
I created a t3.micro EC2 instance on aws being costed at an hourly rate of $0.0065/hr. It's got 2 vCPUs and 1 GiB Memory. I manged to run a 128 tick CS:GO server on it, but the data transfer out charges are killing it. The estimated cost of this server per month is around $43, considering I only play 5 scrims (5v5 competitives) per day, and data transfer out alone costs me $38 in this case. However, some individuals are offering me a server for as low as $10 per month. What am I doing wrong? How do they do it?
You might consider moving from Amazon EC2 to Amazon Lightsail.
Lightsail has pricing plans that include volumes of Data Transfer traffic and it is designed for people who just want to launch a small number of virtual computers (eg WordPress instances) rather than configure a whole cloud infrastructure.
See: Amazon Lightsail Pricing | Virtual Private Server (VPS) | AWS
We are currently runs in-house hardware that we would like to potentially move to AWS. Our main application uses MySQL on a Linux machine (200GB Disk, 32GB RAM, 4 Cores) serving content to customers through a hardware load balancer (around 1 million unique users per month).
We also use a 500 GB CDN hosted by a third party that we would like to move to AWS potentially. What AWS services would you recommend we look at to achieve comparable functionality and would you have a rough monthly cost estimate?
The main reason we would like to move to AWS would be for cost reduction in hardware and better backup strategies.
Thanks!
1.You can host your application using two or more EC2 instances and you can use elastic load balancer to distribute load amongst these EC2 instances.
2.You could use amazon aurora MySQL(server less) which offers you pay as you go service which will allow to get maximum benefits minimising your cost.It is the very best option for your MySQL database as your users are very high and so as the load on the database.
3.For CDN you could go for aws cloudfront and s3. it offers higher availability to your application and it also has less costing.You just need to make some proper configuration and you are ready to go.
AWS is the best cloud option for you as it provides service for your each problem so can use services according to your use and make most of it.
It also provides very good costing options whcih makes your tasks easy.
Please go through aws docs and costing before you choose aws.
Comparable functionality would be to use AWS RDS to replace your MySQL database, one or more EC2 instances to run your application, and then AWS Load Balancer to distribute the load amongst those EC2 web instances. A combination of S3 and Cloudfront to use as a CDN.
Cost is going to depend on how many ec2 instances you use, and the size and options you use for RDS database(s) plus storage and bandwidth - impossible for me to estimate for you
But you can do your own estimates here: https://awstcocalculator.com/
There is only one apparently abstract upside of using Lightsail, simplicity, or significantly simplified interface.
Also, the first page of Lightsail talks about lower charges.
My question is how is it considered to reduce charges compared to EC2? Consider $5 Lightsail plan which charges $0.0067/hour of an instance (which is the cheapest) where EC2's same type of instance (t2.nano) costs just $0.0059/hour.
What am I missing? A detailed price comparison would be much appreciated showing how Lightsail costs lower as advertised.
The $5 Amazon LightSail plan includes:
A CPU that appears similar to a t2.nano ($0.0059c/hr in US regions) = approximately $4.25/month
20GB SSD storage, similar to Amazon EBS general purpose SSD (10c/GB/month) = $2/month
1TB data transfer (9c/GB = approximately $92 in US regions)
So, the real saving appears to be in Data Transfer.
Also with lightsail you get a static IP built into the price, with ec2 it’s about $4/month