Hello Forum,
I wanted to be sure to give back before I started taking. My background is in Computer Science and I wanted to share something that may help those attempting to tackle large technology based challenges. This forum is intended for anyone looking to learn about the technologies present for scalability and those interested in an automated software architecture infrastructure.
To Start:
Most software follows a three tier system.
It is important to understand the different levels of the stack because you will need to use multiple technologies in AWS to support them all.
Note: As I type this up on 6/12/17, I have no working models of this technology stack (besides what I work with at work, and I cannot share that of course), and therefor cannot go into great detail of the configuration or use of these technologies with absolute certainty. This guide to date will require research and trial and error on your part to get going. As I find a need to play with this technology stack, I do plan on releasing a more detailed guide on the setup.
The Tech!!!
For those unfamiliar, AWS stand for Amazon Web Services. There are multiple cloud based computing companies, but I like AWS for a few different reasons. Most of them are selfish (like the fact I already work with AWS on a daily basis), but there are other reasons to use it; like the fact that if AWS goes down, no one will really care that your application is down, because roughly 70% of the internet will also be down. (Ref: How a typo took down S3, the backbone of the internet)
S3 is an AWS feature btw
Inside of AWS, there are multiple features to be utilized. I will categorize them based upon which piece of the software architecture tier they fall under.
Outside of the stack:
VPC
UI: EC2 "Elastic Compute Cloud"
This piece of AWS allows you to launch stand alone cloud based servers for automation. You can publish both your UI code, and Business layer code into these cloud based servers to service requests. EC2 also let you place servers in multiple regions. (Like Oregon, California, N. Virginia, etc.) This is helpful for failover planning, as well as allowing your users to experience faster load speeds as the location of content is closer.
Things to expand on at a later date:
Data Access Layer
Lambda
If you are interested in the future of computing power and technology, pay attention to this part.
When factories were first opened up, and business owners discovered how much better electricity was at doing work rather than steam, they hired on employees who would build, maintain, and run a generator to power the factory. There were no utilities companies then; no one had a way to ensure electricity to their factory without these employees. When the first power plants and Utilities companies opened up, factory owners were hesitant to fire their employees who ran the generators, as they wanted to ensure their work and productivity was unaffected. We look back on this with the perk of hindsight and realize how silly those factory owner's fears were. Today, we simply plug our electronics into the wall and things magically light up. No one cares which power plant sent our energy, because it just happens. We take it for granted. I bring this analogy up, because it is exactly what is occurring in computing power today.
AWS Lambda – Run Code in the Cloud | AWS Blog
Lambda itself is not the application that is changing the future of computing, but the ideas and processes around it are. Lambda is micro-computing.
Micro-Computing occurs inside of a server. The server has software installed that will allow the physical server to be split up into multiple "theoretical" computers. (Meaning that all hardware of 1 physical machine, are shared among 4 "Virtual" machines.) Virtualization (VM or VMWare) has been around for a while now. Micro-Computing steps that game up by launching only enough data and information that is necessary to 1) Launch the Machine. 2) Preform whatever action is located in the request. (Lambda Call). Lambda also allows these requests to be preformed asynchronously which means that if 150 requests are sent to your service, 150 of these Micro-Computers can be spun up at once to preform the work. These Micro-Computers can additionally be spun up on any server that is also a VM Host.
In a sense, Lambda sparks the birth of the online Utility Company. Soon, developers and companies will be writing code that can be utilized on any server. No more downtime, no more service spike errors, and other impacts are sure to follow.
Ok, I am starving, and have to take a break. I have a lot to expand on Lambda and do not want to do it injustice. If you have question or critique, please fire away
Cheers!
Matt
I wanted to be sure to give back before I started taking. My background is in Computer Science and I wanted to share something that may help those attempting to tackle large technology based challenges. This forum is intended for anyone looking to learn about the technologies present for scalability and those interested in an automated software architecture infrastructure.
To Start:
Most software follows a three tier system.
- User Interface (UI) This is the pretty webpage you see before you. This level of the technology stack is responsible for receiving data from a server, and displaying it in a fashion that makes sense to humans. Newer tech has also allowed the UI to interact with the Database layer, and update the UI without a page refresh needing to be preformed. (HTML 5, knockout.js, angular.js, ect)
- Data Access Layer / Business Layer: This layer of the stack is responsible for transmitting data submitted on the UI and sending it to the Database; and vice versa. This layer is also responsible of validating internal business rules before allowing transactions. (For example: Chase will not allow you to withdraw $100 when you account balance is $0. You have the Data Access Layer to thank for this.
- Database: This layer of the stack is responsible for storing your data and accessing it when necessary in your application. Sounds simple enough, but anyone who is well versed in databases will tell you its easier than it sounds.
It is important to understand the different levels of the stack because you will need to use multiple technologies in AWS to support them all.
Note: As I type this up on 6/12/17, I have no working models of this technology stack (besides what I work with at work, and I cannot share that of course), and therefor cannot go into great detail of the configuration or use of these technologies with absolute certainty. This guide to date will require research and trial and error on your part to get going. As I find a need to play with this technology stack, I do plan on releasing a more detailed guide on the setup.
The Tech!!!
For those unfamiliar, AWS stand for Amazon Web Services. There are multiple cloud based computing companies, but I like AWS for a few different reasons. Most of them are selfish (like the fact I already work with AWS on a daily basis), but there are other reasons to use it; like the fact that if AWS goes down, no one will really care that your application is down, because roughly 70% of the internet will also be down. (Ref: How a typo took down S3, the backbone of the internet)
S3 is an AWS feature btw
Inside of AWS, there are multiple features to be utilized. I will categorize them based upon which piece of the software architecture tier they fall under.
Outside of the stack:
VPC
- A VPC stands for your "Virtual Private Cloud". You can think of it as your own little slice of AWS. The VPC allows your code to be hosted along side other companies code on the same server, with no interference. There is not a huge need to go into detail on this one, but realize that all of the tech listed below this point will need to be maintained inside of a VPC.
- Note: your VPC will need a: Subnet, Route Table, Internet Gateway, Network ACL, and Security Group. Again, these pieces of the technology will need to be researched to get off of the ground, but most of these pieces of the puzzle are not to hard. You will want to take extra care note when setting up the Security Group though, as incorrect setting there can leave your environment open to the world for access.
- Noteworthy: Depending on your business needs, you can pay a premium to reserve your own servers in AWS. (Not Shared)
- Think of this as Amazon's version of 'Go Daddy'. While Route 53 contains many more uses than just purchasing domain names, such as natively managing SSL certs (if necessary), and the ability to load balance your web routing; the primary function for new businesses will be to buy your domain name, and point it to the correct server / load balancer in your EC2 (to be covered...).
UI: EC2 "Elastic Compute Cloud"
This piece of AWS allows you to launch stand alone cloud based servers for automation. You can publish both your UI code, and Business layer code into these cloud based servers to service requests. EC2 also let you place servers in multiple regions. (Like Oregon, California, N. Virginia, etc.) This is helpful for failover planning, as well as allowing your users to experience faster load speeds as the location of content is closer.
Things to expand on at a later date:
- Nat Gateway
- Remote RDP
- Availability Zone
- Region
- Load Balancers
- Auto Scaling
- AMIs (Images of the VMs)
- Instances
Data Access Layer
Lambda
If you are interested in the future of computing power and technology, pay attention to this part.
When factories were first opened up, and business owners discovered how much better electricity was at doing work rather than steam, they hired on employees who would build, maintain, and run a generator to power the factory. There were no utilities companies then; no one had a way to ensure electricity to their factory without these employees. When the first power plants and Utilities companies opened up, factory owners were hesitant to fire their employees who ran the generators, as they wanted to ensure their work and productivity was unaffected. We look back on this with the perk of hindsight and realize how silly those factory owner's fears were. Today, we simply plug our electronics into the wall and things magically light up. No one cares which power plant sent our energy, because it just happens. We take it for granted. I bring this analogy up, because it is exactly what is occurring in computing power today.
AWS Lambda – Run Code in the Cloud | AWS Blog
Lambda itself is not the application that is changing the future of computing, but the ideas and processes around it are. Lambda is micro-computing.
Micro-Computing occurs inside of a server. The server has software installed that will allow the physical server to be split up into multiple "theoretical" computers. (Meaning that all hardware of 1 physical machine, are shared among 4 "Virtual" machines.) Virtualization (VM or VMWare) has been around for a while now. Micro-Computing steps that game up by launching only enough data and information that is necessary to 1) Launch the Machine. 2) Preform whatever action is located in the request. (Lambda Call). Lambda also allows these requests to be preformed asynchronously which means that if 150 requests are sent to your service, 150 of these Micro-Computers can be spun up at once to preform the work. These Micro-Computers can additionally be spun up on any server that is also a VM Host.
In a sense, Lambda sparks the birth of the online Utility Company. Soon, developers and companies will be writing code that can be utilized on any server. No more downtime, no more service spike errors, and other impacts are sure to follow.
Ok, I am starving, and have to take a break. I have a lot to expand on Lambda and do not want to do it injustice. If you have question or critique, please fire away
Cheers!
Matt
Dislike ads? Remove them and support the forum:
Subscribe to Fastlane Insiders.
