Reviews (1-25 of 59)
1. We host multiple backend and cloud solutions on EC2.
2. When we want to train Machine Learning Models that need heavy computation and GPU power, we go with EC2.
3. Some solutions are not mandatory to keep up all the time. We use cloud formation script which spins up EC2 - host the solution and thrash it down when not needed.
- EC2 has wide variety of machine configurations. If the intended solutions are memory heavy, CPU heavy, GPU heavy or IO heavy, EC2 will provide proper machine configurations as per the requirements.
- EC2 has lot of Machine Images to setup OS and required softwares. It also allows you to create the image of your own disk. This facilitates user to stop the EC2 instance without loosing the work. It helps to reduce the bill. The image can be attached again to EC2 to start from the same place from where it was left.
- Amazon allows different way to obtain instances like on-demand, spot and reserved. Depending upon the need, one can take wise decision to save cost and address the situation in the best possible way.
- This service is a bit difficult to consume. New users need a big learning curve to use this service effectively.
- UI for EC2 service is a little complex and at many places, it misses detailed explanation.
- Sometimes it takes too long to create images of EC2 instances. This keeps your EC2 up for that extra time. When instances are heavy, it penalizes a lot of money.
Since EC2 is a complex service, it requires proper monitoring of usage. While users are a novice, it requires a bit more examination for proper usage.
- It is very good as a mainframe, for web projects, webapps, databases, etc.
- It is the ideal tool for scalable projects, because it is pay-as-you-go.
- It has a high-speed internet connection.
- It has a high level of security, and a very useful control panel from the web that allows you to use the console without having to start the instance.
- It is very versatile and the mainstay for many other functions offered by AWS.
- Having so many options makes it complex to understand how to start an instance, or configure it.
- If the key generated to enter the VPS is lost, all the information is lost and there is no way to access it again.
- Having multiple instances at the same time is really expensive.
- The commands to use the console are somewhat complex.
- no hardware purchases
- ease of setup and use
- regulations require Tier1 or 0 availability
- Low-impact virtual servers. If you need a server that consumes very little resources/disk space, EC2 is extremely economical.
- Testing. Spinning up an EC2 virtual machine to test applications, services, etc is invaluable.
- Dealing with small businesses, the recurring cost of a high-performance EC2 instance (or one requiring a ton of S3 storage) might be more than an on-premise server for the same task.
- Even with the AWS cost calculator, it can be difficult to accurately estimate the recurring cost of EC2 instances.
- Very cost effective
- Easily scalable. Can increase or decrease servers in minutes.
- Very easy to use. Amazing admin console giving you full control of your servers.
- You have the option to do 1 or 3 year reserved instances, but nothing in between.
- AWS CLI (command line interface) can be tricky to learn and use.
- There are a very large amount of services and configuration options, it's sometimes hard to keep track and understand them all.
- It is possible to create automated backup schedules so your servers will be backed up with less tweaking and work from employees. It is also super quick to restore those backups if necessary. Finally, they back up the server in an efficient manner so you don't have redundant sized backups, but rather just backups of the files changed over the period of time you make the backups.
- The ability to launch instances of any size and then change the size later is incredibly useful and has saved us both time and money in multiple instances. For example, we launched a large instance to make sure we had enough power for an application we launched and later we realized it didn't need as much power so we just turned the instance off and changed the instance type and turned it back on and started saving money!
- The ability to create snapshots of instances is fantastic. We have had some issues launching new instances from snapshots and being unable to access the instance for some reason. It wasn't ever fully resolved but we found our own workarounds.
DevOps supports a fleet of reserved EC2 instances to host backend services and pipelines using Kubernetes.
Data Science team uses EC2 instances to run Jupytor notebooks to do feature exploration on pre-loaded data, and sometimes spot instances to support ad-hoc feature generation and model training. Data team uses the spot instances to run backfill jobs whenever needed.
- A great variety of choices in Amazon Machine Image (AMI) types. Users can select a more basic type to run generic workloads, but also have the choice to pick an AMI pre-installed with specific services in the AWS Marketplace.
- The range of instance types can support the usage from a student's exploration (inexpensive general-purpose nano instances) to an enterprise's most intense workloads (memory or storage-optimized instances with terabytes of memory and ultra-fast network connection).
- The pricing options, from regular instances, reserved instances to spot instances allow users to get the job done and make smart choices about how much they want to pay and when they want to pay.
- The choices on AMIs, instance types and additional configuration can be overwhelming for any non-DevOps person.
- The pricing information should be more clear (than only providing the hourly cost) when launching the instance. AWS DynamoDB gives an estimated monthly cost when creating tables, and I would love to see similar cost estimation showing on EC2 instances individually, as not all developers gets access to the actual bills.
- The term for reserving instances are at least 12 months. With instance types changing so fast and better instances coming out every other day, it's really hard to commit to an existing instance type for 1 or more years at a time.
For users who want to use a managed service, for example a Hadoop platform, I would recommend going with Cloudera and similar companies to get the best support possible.
- It's cost-effective because you pay for only what you use, this is imperative in not-for-profit higher ed, and there are also no long term commitments or upfront fees. This is only one of the pricing options available (the one that best serves us) so there is flexibility.
- Which is my next point, the flexibility of EC2 is what sets is apart from anything else. It is designed to not let the technology get in the way of your ideas and work.
- It is VERY quick and easy to get up and running. Someone brand new to EC2 could have this up and running very fast.
- Performance..EC2 is somewhat slow by comparison and is getting worse as faster processors and hard drives come in the to market.
- Reserved instances.. while they eliminate the price-gap, they do nothing about the performance gap.
- Platform Specific Language that is of any AWS product. The more you need to do with your system and the further you go "inside", the further you go from traditionally recognized languages and their operations. There is a learning curve and it's oftentimes necessary to have support.
- Customizable computing is here to stay with EC2. From a minimal 2-core machine for a minimal microservice to the biggest 16-core Xeon with Nvidia Quadros for a machine-learning model, compute power can be immediately up or downsized to your current need at any given moment.
- Cost savings with EC2 is incredible. By putting computer hardware in a communal bidding system, you minimize your cost per server with every other company who also wants those resources. If you have time-independent processes that you need to process, you can even bid on leftover server contracts that have gone unfilled for pennies on the dollar.
- EC2 storage options are lengthy, with EBS volumes, ephemeral storage, and multiple options to customize throughput and storage cost for each one.
- The UI of AWS is quite hard to familiarize with, along with the infrastructure setup. It's a conglomeration of hundreds of acronyms specific to AWS that must be understood including their minutiae to effectively run a cloud deployment. While AWS' documentation is extensive, their beginner-focused guides could use work.
- Instances become incredibly hard to manage after a critical mass, forcing companies to create their own management applications to fill the void that AWS leaves. Along with this, AWS' SDKs can be very poorly documented making this task exponentially more difficult.
- If you lose your SSH key that is released with the creation of the instance, you lose complete access to the server. While I understand the reasoning behind this decision, MFA recovery would be a nice touch.
- It provides you with static IP addresses.
- Auto-scaling feature.
- Easy to configure and set up your instance.
- You can always change the type of your instances (allocation of more or less CPU/memory for your instance).
- Securely log in to your environment with PEM files.
- I think that AWS Console should have a terminal screen through which you can access your EC2 instances easily in the browser.
- Sometimes you cannot have any clue why the instance is auto-scaled, when you may be pretty sure that there is no high traffic in that particular time.
- The ;earning curve is a bit high in order to make your instances fully configured, and the community is still weak.
- Quick setup: Once you understand the process, the AWS console makes standing up an EC2 instance a breeze.
- Config options: there are plenty of different types of EC2 instances, all geared for specific use cases.
- Documented processes: Amazon White papers are such a great resource when questions arise.
- Default limit: In an EC2 Instance the default limit is 20 per region, you must request for more per region.
- User knowledge: since it as a new technology, getting our admins trained quickly and efficiently has slowed our efforts.
- Cost of support: if you need to engage AWS support the cost can hurt.
- Quick and easy way to deploy resources based on business needs.
- Vast array of AMI to select based on project requirements.
- Allows for quick access control and scalability based on storage and compute needs.
- Pricing for resources can be expanded on more.
- IP allocation for virtual machines should be sticky even though they are not static.
- AMI pricing should be more under control.
- The ability to expand resources for your cloud deployment depending on demand makes it highly appealing.
- The cloud hosting offers excellent backup capabilities so reverting to an earlier version is a breeze.
- Cloud infrastructure avoids the need for us to maintain hardware resources locally. Security and software patching is all handled remotely.
- The whole process of configuration and spinning up instances requires thorough technical knowledge. The learning curve for it seems a bit daunting.
- The training documentation and support is a bit lacking. We tried using Lambda expressions for program flow execution and did not find great resources on that topic.
It frequently addressed the issue of having reliable web servers or virtual machines without having to actually acquire physical servers, rely on resellers of the service or deal with other providers who I've had technical issues with in the past.
- Variety of sizes, you can fine-tune your instance quite a lot rather than being tied into specific tiers like some resellers offer.
- Easy to provision, either using an Amazon tool or AMI, Terraform and/or Ansible I've found it easy to get set up and going on a new EC2 instance.
- With the rise of tools such as Ansible it would be good to see AWS provide similar standardised tooling for EC2.
It's also possible to fine-tune your AWS spending whereas I've found this difficult with AWS resellers in the past.
- It's a very easy task to fire up an EC2, even for first-timer education. Launch a free account, pop into EC2, and follow the prompts.
- Snapshots and Images are particularly nice, fast and easy to work with. I've had my entire catalog erased by mistake, and been able to restore content within minutes.
- Access to data is right there - you have root access to your virtual machine, no clunky interfaces, no negotiation with hosts for resources.
- EC2 is quick to launch, but ends up being an unnecessarily complex rabbit hole. Most users on EC2 are attempting to accomplish the same goals: host content with close to 0 fault tolerance. I want people to be able to buy my stuff. I should be able to one-click a load balancing and autoscaling package for my existing EC2 instance that will scale resources in proportion to my incoming sessions and corresponding usage. Instead, setting up advanced EC2 features ends up requiring an expert to accomplish functions which should be readily available.
- Usage-based billing sounds like you may be getting the best value, but understand that Amazon is not losing money on hosting, and costs are no longer entirely predictable. Cents for data piles up quickly and once you get load balancing up with a influx of customers, your finance department comes at you waving your budget around...
- EC2 makes right-sizing your servers a breeze. You can quickly spin up a server in the cloud and if it turns out the vCPU, RAM, or storage space is wrong, within minutes you can change all of that.
- EC2 makes backups and restores a breeze. We actually had a client that allowed a hacker to remote into their production server. We were able to shutdown the EC2 instance, spin up a backup from an AMI, and then attach the existing elastic IP. This was all done within a 15 minute window.
- EC2 makes quickly deploying multiple servers a very easy. Within minutes, you can deploy a whole fleet of cloud servers.
- EC2 is easy to script. We are able to save our clients a lot of money by scripting their EC2 instance to shutdown/startup at predetermined times so they are only paying for the server when they are using it.
- EC2 in my opinion, is lacking the ability to connect to a console from within the AWS console. I sometimes miss how I can connect to the console with VMware and Hyper-V but not with EC2. You have to utilize RDP or SSH to connect to an instance.
- Sometimes EC2 instances lockup due to reasons with the underlying hardware and need to be shutdown and the started back up so the instance can spin up on new hardware. This is sometimes a problem because unless you set up proper alerting/scripting, you don't know there is an issue until a user reports it.
- EC2 can be a bit daunting for the beginning user. You really do need some kind of training before you dive in.
- Easy to maintain. We can elastically grow the instances as we need.
- Can be distributed among several regions, hence it performs well.
- Can be configured to restrict the access to instances outside specific IPs.
- Can be tied to load balancer.
- Spot instances available to bid for cheaper price.
- There should be an option to upgrade to only CPU and memory, instead of getting overall big instances.
- Sometime we are forced to upgrade or terminate old instances. They could support old instances.
- Launch time of the instances has room for improvement. Could be faster.
EC2 instances allow us to abstract all the management of servers and concentrate on what really generates value in our business: building solutions for our clients.
- Quick learning curve and ease of acquisition for new learners due to their 12 month free trial.
- Connection to the entire AWS ecosystem, such as RDS service for database management.
- Dynamic scaling of instance resources allows you to achieve the performance you are looking for without having to pay more than necessary.
- Hot swap of volumes and other resources.
- You can't easily know the end of free trial period, which can generate monthly costs for unused services (even so, the support for these isolated cases is very good!)
- The default configuration of resource usage alerts could be better. Even so, there are alternatives to control these cases outside of AWS.
- While you're still learning how to handle instances, one can make some serious mistakes, such as leaving open ports or deleting an instance without realizing it. Again, is not a core AWS responsability but a few alerts could be great (or you can leave infraestructure experts handle all the management).
But if virtualization is an alternative, or you hear good comments among your managers about IaaS or PaaS, then AWS EC2 is the way to go.
And for start-ups it is definitely the best alternative.
- Cost effictiveness is great, they only charge for what you use so you do not have to pay for what you dont use
- Intuitive interface, makes setting up and deploying new and existing projects an ease
- Secure. Have not had server attacks since we migrated to them, so the uptime has been phenomenal
- The PEM keys are a bit confusing if you are not accustomed to it
- A dummy version/starter guide would be great. Once you have it configured its easy to use and makes sense, but my first interaction with it was a bit to grasp
- Add easily installable cPanel or Plesk or equivalent as an option for customers just wanting to host sites, who don't need the more in-depth options
- Best in performance
- Offers so much customization and security measures
- Very cost effective
- Upgrading of EC2 core components should be pre-announced to the customer, rather than crashing the server (which happened to me in 2017 more than 8 times.)
- Customer support is very costly and not that effective on cases, if bought
- For choosing the Mumbai region, I had to get authorization from AWS support. It was a very awkward and lengthy process: it took 2 days for the request (this happened to us in Dec 2017.)
Blogs and Personal website can be hosted at a very cheap cost (approx. $20) and in a way that's more secure than any other cloud providers. And for high performance EC2, go for Reserved Instances which is more affordable than unreserved instances
- Full control over the software and settings.
- Instant availability of a new server with the power you require.
- Thorough permission support to ensure only those who have the rights to monitor or configure the servers can do so.
- Many world wide locations to make sure it's closer to the country your users are in.
- Huge learning curve. To get a basic instance up with default settings is very easy, but there's hundreds or perhaps thousands of settings without explanations of what they do.
- Multiple ways to do the same thing, like the browser console, the command line, and APIs, means finding answers on how to do something may be provided only in one way and not the way you have to do it.
- Lack of documentation on best practices in many scenarios. AWS assumes you have devops experience and makes it too easy for you to make mistakes and follow bad practices.
Amazon Elastic Compute Cloud (EC2) Scorecard Summary
Feature Scorecard Summary
About Amazon Elastic Compute Cloud (EC2)
Amazon Elastic Compute Cloud (Amazon EC2) is a web service that provides secure, resizable compute capacity in the cloud. Users can launch instances with a variety of OSs, load them with custom application environments, manage network access permissions, and run images on multiple systems.
Bare metal instances
Amazon EC2 Fleet (fleet management)
Pause and resume instances
GPU compute instances
GPU graphics instances
High I/O instances
Dense HDD storage instances
Optimized CPU configurations
Flexible storage options
Place instances in multiple locations
Elastic IP addresses
Auto-scale capacity up or down
Elastic Fabric Adapter
Available on AWS PrivateLink
Amazon Time Sync Service
Amazon Elastic Compute Cloud (EC2) Technical Details