anyone who has usedAWSIt took a few minutes to probably create an EC2 instance -- a remote virtual machine that can be created with just a few mouse clicks or via script. You choose a machine image (containing the operating system plus some business applications), an instance type (representing CPU and memory and other hardware), some further settings (including network, disk size and firewall), and you're ready to go for a few minutes within.
In this blog, we'll learn about:
- What is EC2? What is an instance?
- The logic behind AWS instance type names
- The cost of running an EC2 instance
- How to choose the right EC2 instance type
Demystification 1: What is EC2?
EC2 stands for "Elastic Compute Cloud," a fancy name for a virtual machine and its supporting infrastructure. They didn't want to call it a virtual machine because that was only two letters and not unique enough. They can't call it ECC because that'smemory typeand is also a typeCryptography. But there are two C's in Elastic Compute Cloud, hence 2. Maybe it should be superscripted, like EC2。
Secret 2: What is an instance?
An EC2 instance is a single virtual machine that runs (or stops) within the EC2 infrastructure in the AWS cloud environment. It may run Linux or Windows, and you can use it like a real Linux or Windows server computer.
The EC2 infrastructure supports everything needed to create a running EC2 instance, including networking, storage, firewalls, and the base Amazon Machine Image (AMI) used to create an EC2 instance.
Why is this important?
One of the great things about EC2, and the reason cloud services have become so popular, is that you pay for what you use. You can lease EC2 instances by the hour or by the second. This different approach to business (opex instead of capex) fundamentally changes how IT infrastructure is used.
The logic behind AWS instance type names
The biggest challenge in understanding AWS is how they name things. Everything is TLA (the three-letter acronym), or it's "resilient" or "simple." Once you understand the nomenclature, the logic of AWS is quite sound. EC2 instance sizes are a good example.
The instance types are named as follows:
t2.micro
r5ad.Large
x1e.16x large
m5. metal
At first glance, "size" might seem to mean nothing, but there is a way to the nomenclature of AWS EC2 instance types.
This has several parts:
- The first letter is "Family". more on that later
- Number
- additional letter before dot
- text after point
number:
The numbers are the algebras in the family: M5 is newer than M4, reflectingMoore's Law, a new generation of instances with the same specifications is usually cheaper than the previous generation. For example:
- M5.large (2 vCPUs and 8GiB RAM on a 3.1 GHz Intel Xeon Platinum 8175M) costs $0.107/hour,
However - M4.large (2 vCPUs and 8GiB RAM on a 2.4 GHz Intel Xeon E5-2676 or E5-2686) costs $0.222/hour.
(Ireland prices, February 2020)
The first EC2 instance introduced was m1.small in 2006.Jeff Barr has produced a fascinating timeline of EC2 history。
Other letters:
The other letters before the dot are also (mostly) logical:
- a lowercase letterA(e.g. m5a.large) meansAMD processorinstead of Intel processors. Both Linux and Windows are supported.
- a lowercase letterG(e.g. m6g.large) meansARM graviton processorinstead of Intel processors. As of this writing (September 2020), only Linux is supported.
- a lowercase letterd(e.g. m5d.large) indicates the presence ofinstance store disk（see footnote）。
- a lowercase lettern(e.g. m5n.large) for fasternetworking. For example, m5 supports 25Gbps; m5n supports 100 Gbps.
- Sometimes these can be combined, for example m5ad.large has an AMD processor and an instance store disk.
After point:
The bits after the dot are also logical. Regardless of the instance type, there is a correlation between the name and the number of vCPUs. An xlarge has 4 vCPUs, any multiplier for the xlarge is also a multiplier for the number of vCPUs.
| Name | # vCPU |
|---|---|
| . Moderate | 1 (but see below) |
| . big | 2 |
| .xlarge | 4 |
| .2x large | 8 |
| .4x Larger | 16 |
| .9x large | 36 etc. |
The system is a bit broken at the bottom: A1, M1 and M3.medium instances have 1 vCPU, but some medium instances have 2 vCPUs (C3, T2, T3, T3a). Smaller than medium instances mostly have 1 vCPU (including t2.nano, t2.micro, and t2.small), with the exception of the newer T3 and T3a, which have at least 2 vCPU.
There are some exceptions to this pattern in older generation instances, notably C1 and M2.
family:
Let's go back to "family". Now that there are many families in the instance type tribe, we group them into five categories:
- General purpose(start here)
- Computing optimization(more CPU, less memory)
- memory optimization(less CPU, more memory)
- storage optimization(large number of local instance store volumes)
- accelerated computing(GPU, FPGA, and other specialized processing)
In general, these series tell you the ratio of compute (vCPU) to memory (RAM). Starting with a .large instance:
Compute-optimized instances have 2 GiB RAM per vCPU (C instances)
General purpose instances have 4 GiB RAM per vCPU (T and M instances)
Memory-optimized instances provide 8 GiB RAM per vCPU (R5 and Z1d instances)
Storage-optimized instances provide (approximately) 8 GiB of RAM per vCPU (i3, D2, i3, H1)
Some instances have more memory:
x1 instances: 15.25 GiB per vCPU (64 vCPUs with 976 GiB RAM, or 128 vCPUs with 1952 GiB RAM)
x1e instances: 30.5 GiB per vCPU (4 to 128 vCPUs, 122 to 3904 GiB RAM)
In older generation instances, the correlation of vCPUs to RAM is less logical. A lot of memory is slightly smaller than specified above, which I'm guessing is due to less efficient virtualization.
In accelerated computing instances, vCPUs are related differently to RAM, where the GPU may be the primary performance characteristic. But for completeness:
G4: 4.000 GiB per vCPU
G3: 8.125 GiB per vCPU
P3: 8.125 GiB per vCPU
P2: 15.250 GiB per vCPU
F1: 15.250 GiB per vCPU
Exceptions:
Other exceptions to the above pattern are as follows:
- Smaller instances (less than .large) do not obey the rules, mainly because they cannot have less than 1 vCPU.
- The vCPU to RAM ratio of the A instances (first generation Graviton ARM machines) should put them together with other compute optimized instances. Obviously, they compare well to C instances in terms of price and performance.
- The relevance of older generation instances is not so easy to see. For example, 2 vCPUs of an R4 instance has 15.25 GiB, which is "nearly" an 8:1 ratio.
- Accelerated computing instances (emphasis on GPUs rather than CPUs).
- High Memory Instances (u*.metal) - Both have 448 logical processors but vary in amount of RAM up to 24 TiB.
The cost of running an EC2 instance
The cost of running an EC2 instance is divided into several factors:
- The cost of EC2 itself
- The cost of EBS volumes (virtual "disks" that have a separate lifecycle from the EC2 instance)
- Costs associated with certain parts of the network, such as static IP addresses
- Bandwidth Costs for Data Outside of AWS
The cost of an EC2 instance depends on how long it is running: when you stop an instance, it incurs no cost. When you start again, you start paying again. thispricing pageShows hourly costs for different instance sizes. For most operating systems (RedHat, Windows, SuSE), you will be charged for the entire billable hour, even if the machine is only running for a few minutes. For Amazon Linux and Ubuntu, billing is per second (minimum one minute).
A machine with twice the vCPUs will cost twice as much. Check out the m5 instances with Ubuntu in the London area at hourly prices (published September 2020):
m5.large $0.111 2 vCPUs 8 GiB RAM
m5.xlarge $0.222 4 vCPUs 16 GiB RAM
m5.2xlarge $0.444 8 vCPUs 32 GiB RAM
m5.4xlarge $0.888 16 vCPUs 64 GiB RAM
m5.8xlarge $1.776 32 vCPUs 128 GiB RAM
This means that it is equally cost-effective to choose a small number of large machines or a large number of small machines, depending on usage.
The graph below shows the correlation of machine cost (USD/hour) to machine size (GiB RAM) for different M instances in the Ireland region. AMD instances (M5a) are less expensive than Intel instances (M5). M5ad and M5d instances also have local instance store volumes, and M5n and M5dn instances have enhanced networking capabilities: additional capabilities have an associated additional cost.
Side note: Like everything else in AWS, prices are also available via API, so you canWrite your own application to calculate costsif you want.
Additionally, EC2 instances will have one or more EBS volumes attached. They are used for EC2's startup disk (the C:\ drive on a Windows machine). EBS volumes have their own life cycle, andCalculate the cost separately。
Additionally, there may be network costs associated with running an EC2 instance, including usingload balancer, a static IP address (called the"Elastic" IP addresses) andNAT gateway。
In addition, any data transferred from AWS to the Internet or across regions also generatesCost per GB。
How to choose an instance type
With all this information, choosing an instance type might seem like rocket science. The following points may help:
- Start with an M instance type, then useCloud monitoring monitoringto understand utilization.
- Changing the instance type involves stopping the instance, changing to the new instance type, and restarting the instance. This can be achieved with just a few clicks. Some downtime is required.
- For databases with more memory, the R instance may not have enough memory. Check out X1 and X1e memory optimized examples.
- For big data clusters, or anything that requires a lot of local storage, consider the H or D storage-optimized instance types.
- If your workload has bursty CPU usage, considerA T2 or T3 instance, which I've blogged about before。
- For bedtime reading, find the white paperwell-structured frameworkAnd read about best practices for building applications in the cloud.
further help
To get the most out of AWS, consider taking a training course.QA offers a full range of AWS training coursesand technical guidance in many other areas.
footnote:
EBS and instance storage:
EBS volumes are the storage of choice for EC2: they behave like physically attached disks, but are network-attached close to the host machine running the EC2 instance. EBS volumes are ideal for boot disks (operating system disks) and most applications, including high-performance databases.
instance store volumeis a disk (magnetic HDD or solid-state SSD) attached locally to the physical host machine running EC2. They are much faster than EBS volumes, but they are ephemeral: when the machine stops, the data is lost (actually, the physical device is actively zeroed to prevent data leakage). For boot disks, you must use EBS. Instance store volumes are ideal for buffers, caches, staging data, and other temporary content, or data that is replicated across instance groups (such as in big data clusters).
justin watkins
Justin is an experienced instructor with a passion for bringing technical content to life. He currently offers training courses on AWS (Amazon Web Services) cloud architecture, and his training portfolio also includes cluster management, big data, and math-intensive applications.
FAQs
What is the size of the m5 2xlarge instance type? ›
The m5. 2xlarge instance is in the general purpose family with 8 vCPUs, 32.0 GiB of memory and up to 10 Gibps of bandwidth starting at $0.384 per hour.
What is M5a large? ›large. The m5a. large instance is in the general purpose family with 2 vCPUs, 8.0 GiB of memory and up to 10 Gibps of bandwidth starting at $0.086 per hour.
What type of instance is m5d metal? ›metal. The m5d. metal instance is in the general purpose family with 96 vCPUs, 384.0 GiB of memory and 25 Gibps of bandwidth starting at $5.424 per hour.
What is an m5 instance? ›M5 instances offer a balance of compute, memory, and networking resources for a broad range of workloads. This includes web and application servers, small and mid-sized databases, cluster computing, gaming servers, caching fleets, and app development environments.
How much memory does c5 2xlarge have? ›2xlarge. The c5. 2xlarge instance is in the compute optimized family with 8 vCPUs, 16.0 GiB of memory and up to 10 Gibps of bandwidth starting at $0.34 per hour.
How much storage does DB m5 Xlarge have? ›| Model | vCPU | Storage |
|---|---|---|
| db.m5d.xlarge | 4 | 1 x 150 NVMe SSD |
| db.m5d.2xlarge | 8 | 1 x 300 NVMe SSD |
| db.m5d.4xlarge | 16 | 2 x 300 NVMe SSD |
| db.m5d.8xlarge | 32 | 2 x 600 NVMe SSD |
For example, an m5. xlarge instance type has two CPU cores and two threads per core by default—four vCPUs in total.
What processor does the m5 Xlarge have? ›| Compute | Value |
|---|---|
| Physical Processor | Intel Xeon Platinum 8175 |
| Clock Speed (GHz) | 3.1 |
| CPU Architecture | x86_64 |
| GPU | 0 |
M5ad instances are designed for general purpose workloads such as web servers, app servers, dev/test environments, gaming, logging, and media processing. They are available in 6 sizes: Instance Name. vCPUs. RAM.
What is the best EC2 instance for heavy CPU load? ›Amazon EC2 M6in and M6idn instances are ideal for network-intensive workloads such as backend servers, enterprise, gaming servers, and caching fleets applications.
What are the 3 types of EC2? ›
The AWS EC2 Instance Types are as follows:
General Purpose Instances. Compute Optimized Instances. Memory-Optimized Instances.
metal. The m5. metal instance is in the general purpose family with 96 vCPUs, 384.0 GiB of memory and 25 Gibps of bandwidth starting at $4.608 per hour.
What are the five instance categories? ›- General Purpose.
- Compute-Optimized.
- Memory-Optimized.
- Storage-Optimized.
- Accelerated Computing.
For example, an m5. 8xlarge instance has 32 vCPUs and 10 Gbps network bandwidth, and an m5. 16xlarge instance has 64 vCPUs and 20 Gbps network bandwidth.
How much does M5 xlarge cost? ›large pricing: $81.76 monthly - AWS EC2.
What is the difference between M5 large and C5 large? ›Amazon EC2 M5 and C5 instances offer high performance for both compute- and memory-intensive applications. A C5 instance delivers more compute capacity than memory, whereas a M5 instance balances compute, memory, and network bandwidth.
Is an instance the same as a VM? ›An instance is a virtual machine (VM) hosted on Google's infrastructure. You can create an instance or create a group of managed instances by using the Google Cloud console, the Google Cloud CLI, or the Compute Engine API.
How much is C5 2xlarge per hour? ›c5. 2xlarge. 2xlarge machine type has 8 vCPUs and 16 GB of memory. Pricing for this instance starts at $0.42 per hour and $309.52 monthly.
How much RAM does a t2 2xlarge have? ›| Size | vCPUs | Memory (GiB) |
|---|---|---|
| t2.large | 2 | 8 |
| t2.medium | 2 | 4 |
| t2.xlarge | 4 | 16 |
| t2.2xlarge | 8 | 32 |
4xlarge instance is in the memory optimized family and has 16 vCPUs, 128 GiB of memory and up to 10 Gibps of bandwidth starting at $2.0 per hour.
What is the purpose of EBS? ›
Amazon EBS allows you to create storage volumes and attach them to Amazon EC2 instances. Once attached, you can create a file system on top of these volumes, run a database, or use them in any other way you would use block storage.
How much storage does DB t2 Micro have? ›The db. t2. micro instance is in the general purpose family and has 1 vCPUs, 1 GiB of memory and low to moderate network performance starting at $0.018 per hour.
What is the max RDS Oracle size? ›MySQL, MariaDB, PostgreSQL, and Oracle RDS DB instances can be created with up to 64TB of storage, and SQL Server RDS DB instances with up to 16TB of storage when using the Provisioned IOPS and General Purpose (SSD) storage types.
What size is M5? ›| Thread Size | Nominal Diameter (mm) | Effective Diameter (mm) |
|---|---|---|
| M4 | 4.0 | 3.545 |
| M4.5 | 4.5 | 4.013 |
| M5 | 5.0 | 4.480 |
| M6 | 6.0 | 5.350 |
M5X18. M- Indicates this bolt uses metric threads. 5- Refers to the nominal diameter of the bolt shaft in millimeters. X- Separates the Metric size of the threads and how long the bolt is. 18- This means the threaded part of the bolt is 18mm long.
What are the standard sizes of M5? ›Size M5 Shank diameter 5mm Pitch 0.80mm Length (measured from under the head) 12mm Thread length Fully threaded Material Stainless...
What is the difference between M6i and M5? ›M6i instances provide up to 20% higher memory bandwidth per vCPU compared to M5 instances. You can also enable Elastic Fabric Adapter (EFA) in M6i, M6id, M6in, and M6idn instances on the 32xlarge and metal sizes.
How many cores does M5 8xlarge have? ›For example, a m5. 8xlarge has 16 physical CPU cores, which corresponds to 32 vCPUs. For clusters, Cloud Center only supports Linux On-demand instances.
What is the difference between compute optimized and memory optimized? ›EC2 Instance Type Breakdown
Compute Optimized: Good for compute-intensive applications such as some scientific modeling or high-performance web servers. Memory Optimized: Used for anything that needs memory-intensive applications, such as real-time big data analytics, or running Hadoop or Spark.
M5 instances have Intel while M5a instances have the AMD processor. As you may have guessed by now, 'a' in the instance name (eg: M5a. xlarge) refers to AMD. M5a instances offer same performance but are about 10% cheaper than their M5 counterparts.
How do I know my EC2 instance type? ›
To find an instance type using the console
Open the Amazon EC2 console at https://console.aws.amazon.com/ec2/ . From the navigation bar, select the Region in which to launch your instances. You can select any Region that's available to you, regardless of your location. In the navigation pane, choose Instance Types.
Disk and tiering limits by EC2 instance
Cloud Volumes ONTAP uses EBS volumes as disks, with a maximum disk size of 16 TiB.
What is the best EC2 instance for heavy CPU load? Consider running one of the instances of the compute optimized EC2 instance type. The compute optimized Amazon EC2 instance type contains two families: C and Hpc.
What is the maximum file size for EC2? ›By default, the maximum file upload size on an EC2 Ubuntu instance is set to 2 MB. This means that you can upload files that are up to 2 MB in size using the default settings.
What is the most popular EC2 instance? ›Among the most popular and widely used EC2 instance types, the General Purpose instance is a good choice if you are new to cloud computing or AWS in general. As this instance type offers a wide balance of computing power, memory, and storage, it is suited for a majority of AWS workloads.
What is better than EC2? ›Lambda is ideal for short-term tasks. EC2, in contrast, is intended for long-term, steady-state operations, such as running websites and applications or supporting resilient high-performance computing. Even so, both platforms can work together.
Is EC2 just a VM? ›Amazon Elastic Compute Cloud (EC2) is the web service you use to create and run virtual machines in the cloud. Amazon Web Services calls these virtual machines 'instances'.
What size is a M5 rod? ›8, M5 Metric STAINLESS STEEL Fully Threaded Rod, all-thread Diameter: 5mm Pitch: 0.8 Coarse Thread Overall Length: 1 meter / 1000mm / 100cm / 39.37" DIN 975 Please note, the thread profile of this metric threaded rod is common ISO metric screw threading.
How much RAM does EC2 instance have? ›The Most Memory of any EC2 Instance, SAP-Certified
AWS offers a wide range of memory optimized instances ranging from 8 GiB on R6g instances up to 24 TiB High Memory instances. Amazon EC2 High Memory instances have the highest amount of memory of any EC2 instance with up to 24 TiB of memory.
Go to the CloudWatch dashboard in your AWS console and click on Metrics. You'll see a new CWAgent card. After you click on this card, you should see your EC2 memory usage in a graph.
What does EC2 stand for? ›
Amazon Elastic Compute Cloud (Amazon EC2) is a web service that provides secure, resizable compute capacity in the cloud. It is designed to make web-scale cloud computing easier for developers.
Which AMI is best for EC2? ›We recommend that you use the Amazon ECS-optimized Amazon Linux 2 AMI for your Amazon EC2 instances unless your application requires a specific operating system or a Docker version that is not yet available in that AMI.
What are the three instances? ›"Three Instances" is a series of hand-blown hourglasses that respectively fail to contain their time-measuring substances.
What is the size of the M5 2xlarge instance type? ›The m5. 2xlarge instance is in the general purpose family with 8 vCPUs, 32.0 GiB of memory and up to 10 Gibps of bandwidth starting at $0.384 per hour.
How much network bandwidth does c5n 18xlarge have? ›| Storage | Value |
|---|---|
| EBS Optimized | True |
| Max Bandwidth (Mbps) on (EBS) | 4750 |
| Max Throughput (MB/s) on EBS | 593.75 |
| Max I/O Operations/second (IOPS) | 20000 |
An AWS vCPU is a single hyperthread of a two-thread Intel Xeon core for M5, M4, C5, C4, R4, and R4 instances. A simple way to think about this is that an AWS vCPU is equal to half a physical core.
How do I find out the size of my EC2 instance? ›- df = Disk Free command.
- -h = Human Readable, print sizes in human readable format (e.g., 1k, 234M, 2G)
- -T = Print Type, print file system type.
You can create MySQL, MariaDB, Oracle, and PostgreSQL RDS DB instances with up to 64 tebibytes (TiB) of storage. You can create SQL Server RDS DB instances with up to 16 TiB of storage. For this amount of storage, use the Provisioned IOPS SSD and General Purpose SSD storage types.
How do I check disk size in AWS? ›To view information about a volume using the console
Open the Amazon EC2 console at https://console.aws.amazon.com/ec2/ . In the navigation pane, choose Volumes. To reduce the list, you can filter your volumes using tags and volume attributes.
The size of the backup vault can be found in RDS -> Snapshots -> Backup service. If you have more than 1 backup in the vault, the size of the vault would be the sum of the size of all the backups.
How do I increase the size of my instance? ›
- Open the EC2 console.
- Select the instance you wish to resize, and stop the instance.
- With the selected instance, choose Actions > Instance Settings > Change Instance Type.
- From the Change Instance Type dialog box, choose which instance you would like to resize to.
- Open the AWS Service Quotas dashboard.
- Choose Amazon Elastic Compute Cloud (Amazon EC2).
- Choose the required service quota.
- Choose Request quota increase.
- First, stop your EC2 instance.
- Go to EC2 > Volumes and find your instance's volume.
- Go to Actions > Create Snapshot. ...
- Go to Actions > Modify Volume, and increase the storage size. ...
- It'll take a while for the volume resize operation to complete.
4xlarge instance is in the memory optimized family and has 16 vCPUs, 128 GiB of memory and up to 10 Gibps of bandwidth starting at $2.0 per hour.
How big is the storage on DB R5 Xlarge? ›The db. r5. xlarge instance is in the memory optimized family and has 4 vCPUs, 32 GiB of memory and up to 10 Gibps of bandwidth starting at $0.5 per hour.
What is the maximum RDS volume size? ›MySQL, MariaDB, PostgreSQL, and Oracle RDS DB instances can be created with up to 64TB of storage, and SQL Server RDS DB instances with up to 16TB of storage when using the Provisioned IOPS and General Purpose (SSD) storage types.
What is scaling an EC2 instance from R4 large to R4 4xlarge called? ›Scaling an instance from an r4.large to an r4.4xlarge is called. Vertical Scalability.
How many vCPUs are in EC2? ›However, a total of 8 CPU cores (16 AWS vCPUs) and 128 GB RAM are strongly recommended for a single production Amazon EC2 instance. An AWS vCPU is a single hyperthread of a two-thread Intel Xeon core for M5, M4, C5, C4, R4, and R4 instances.