- speed (improve agility and time to market)
- scalability (scale up or down depending on resource requirements)
- lower maintenance costs (see previous bullet)
Showing posts with label aws. Show all posts
Showing posts with label aws. Show all posts
Digital Media in the AWS Cloud
One day, I'll take the hint and stop signing up for these webinars. Ten minutes into it, my brain glazed over because the content dramatically shifted from being marketer friendly to heavy hardcore IT. All I managed to ascertain from this one were the benefits for hosting media on Amazon's cloud. However, this isn't any different than prior AWS webinars.
Fast & Frugal Webinar - Running a Lean Startup with AWS
This was a multipart webinar series that Amazon AWS hosted. I thought that the speakers would shed some light into a few notions that are critically important to small business owners, primarily what a lean startup is, how these startups are using AWS for their business, and how to take this knowledge from webinar to execution. Some parts were way too technical for a mixed business audience.
"Lean" is a concept that is closely associated with manufacturing, and according to Wikipedia it is a production practice that considers the expenditure of resources for any goal other than the creation of value for the end customer to be wasteful, and thus a target for elimination. Working from the perspective of the customer who consumes a product or service, "value" is defined as any action or process that a customer would be willing to pay for. Essentially, lean is centered on preserving value with less work.
Recommendations for Lean Startups:
Use innovation accounting with three learning milestones. (1) establish the baseline to build a minimum viable product and measure how customers behave right now. (2) tune the engine and experiment to see if we can improve metrics from the baseline towards the deal. (3) pivot or persevere when experiments reach diminishing returns, it's time to pivot.
Leverage AWS. Why? Your server infrastructure needs will change. If you are able to run on the same infrastructure at launch and a year later, you overbuilt your launch infrastructure, or even worse, your business is not scaling. Goal: find a Scalable Business Model.
Build your Minimum Viable Product (MVP). Toss the vanity metrics (ones use in frilly reports to senior management, shareholders, or key stakeholders) and instead measure key assumptions that answer what we are trying to prove with this startup. Use 3rd party SaaS metrics or Amazon's elastic mapreduce. Goal: learn from the process, iterate or pivot in a new direction.
Use the building blocks AWS provides. It is on demand, available, elastic (use as much or as little as needed), pay as you go, low cost, no contracts, no upfront payments, and no cap-ex.
Vertical scaling - increase hardware, cpu, memory
Horizontal scaling - increases instance number, master/slave replications
Database sharding - indexes partial customer datasets
Goal: stay lean by using only what you need and reduce costs.
"A startup is a human institution designed to deliver a new product or service under conditions of extreme uncertainty." --Eric Ries, author of The Lean Startup
"Lean" is a concept that is closely associated with manufacturing, and according to Wikipedia it is a production practice that considers the expenditure of resources for any goal other than the creation of value for the end customer to be wasteful, and thus a target for elimination. Working from the perspective of the customer who consumes a product or service, "value" is defined as any action or process that a customer would be willing to pay for. Essentially, lean is centered on preserving value with less work.
 |
| The Toyota Way, an example of a lean process |
Use innovation accounting with three learning milestones. (1) establish the baseline to build a minimum viable product and measure how customers behave right now. (2) tune the engine and experiment to see if we can improve metrics from the baseline towards the deal. (3) pivot or persevere when experiments reach diminishing returns, it's time to pivot.
Leverage AWS. Why? Your server infrastructure needs will change. If you are able to run on the same infrastructure at launch and a year later, you overbuilt your launch infrastructure, or even worse, your business is not scaling. Goal: find a Scalable Business Model.
Build your Minimum Viable Product (MVP). Toss the vanity metrics (ones use in frilly reports to senior management, shareholders, or key stakeholders) and instead measure key assumptions that answer what we are trying to prove with this startup. Use 3rd party SaaS metrics or Amazon's elastic mapreduce. Goal: learn from the process, iterate or pivot in a new direction.
Use the building blocks AWS provides. It is on demand, available, elastic (use as much or as little as needed), pay as you go, low cost, no contracts, no upfront payments, and no cap-ex.
Vertical scaling - increase hardware, cpu, memory
Horizontal scaling - increases instance number, master/slave replications
Database sharding - indexes partial customer datasets
Goal: stay lean by using only what you need and reduce costs.
Amazon Simple Email Service (SES)
Attended an Amazon webinar today that covered their Simple Email Service. I have yet to test it out. Initial impressions suggest that you should stick with your existing email service provider for basic email marketing broadcasts because there aren't a whole lot of tools nor analytics built yet for SES. It is a basic as basic gets.
A few things to note:
There are two types of email sending options: formatted or raw message; both of which are also supported by API functions (SendEmail, SendRawEmail). The basic feedback analytics are reminiscent of basic web stats (GetSendStatistics API) which will tell you about delivery attempts, rejected messages, hard bounces, and spam complaints. And, everyone's lists are served up separately, so even if you have an overlap of customers with another business that also uses Amazon SES, their spam complaints do not affect your broadcasts (unlike how Gmail blocks spam).
Most ESPs have sending limits built into their pricing contracts. Amazon is no different. Every SES user is assigned a quota (max number of emails that can be sent in 24 hours) and an access level (everyone starts out in the sandbox to test SES features and can send up to 200 recipients a day), which is tiered. However, you can only send to and from a verified email address.
Pricing. ESPs that support small to medium sized businesses such as iContact, Bronto, Vertical Response, or ConstantContact, shouldn't fret about Amazon's pricing structure. Pricing is tiered and is based on the combination of two elements: data transfer and CPM per email message.
Current CPM is $0.10 per thousand messages sent. Data transfer pricing is as follows:
Pricing examples:
1000 email messages to one recipient per day with content size 10kb
= 31,000 recipients sent per month
= 3.1 GB in/out data transfer
= $3.73 for the month
1000 email messages outbound to one recipient per day with content size 100kb
= 31,000 recipients sent per month
= 31 GB in/out data transfer
= $10.70 for the month
Amazon EC2 users are already subscribed to the free pricing tier and are capped at 2000 messages for free each day.
To get started with Amazon SES:
Subscribe
Verify email addresses / create your own whitelist of verified addresses
Send Email
Request production access
Get Feedback
Limits on sandbox accounts: 100 whitelisted verified addresses, up to 10MB per message (because most popular email readers cannot handle more than that).
Limits on all accounts: Does not support SOAP or file attachments
A few things to note:
- Supported APIs
- Built-in Features
- Sending Limits
- Access Levels
- Pricing
There are two types of email sending options: formatted or raw message; both of which are also supported by API functions (SendEmail, SendRawEmail). The basic feedback analytics are reminiscent of basic web stats (GetSendStatistics API) which will tell you about delivery attempts, rejected messages, hard bounces, and spam complaints. And, everyone's lists are served up separately, so even if you have an overlap of customers with another business that also uses Amazon SES, their spam complaints do not affect your broadcasts (unlike how Gmail blocks spam).
Most ESPs have sending limits built into their pricing contracts. Amazon is no different. Every SES user is assigned a quota (max number of emails that can be sent in 24 hours) and an access level (everyone starts out in the sandbox to test SES features and can send up to 200 recipients a day), which is tiered. However, you can only send to and from a verified email address.
Pricing. ESPs that support small to medium sized businesses such as iContact, Bronto, Vertical Response, or ConstantContact, shouldn't fret about Amazon's pricing structure. Pricing is tiered and is based on the combination of two elements: data transfer and CPM per email message.
Current CPM is $0.10 per thousand messages sent. Data transfer pricing is as follows:
| Data Amt | Cost |
| First GB | Free |
| up to 10 TB | $0.15/GB |
| next 40 TB | $0.11/GB |
| next 100 TB | $0.09/GB |
| 150+ TB | $0.08/GB |
Pricing examples:
1000 email messages to one recipient per day with content size 10kb
= 31,000 recipients sent per month
= 3.1 GB in/out data transfer
= $3.73 for the month
1000 email messages outbound to one recipient per day with content size 100kb
= 31,000 recipients sent per month
= 31 GB in/out data transfer
= $10.70 for the month
Amazon EC2 users are already subscribed to the free pricing tier and are capped at 2000 messages for free each day.
To get started with Amazon SES:
Subscribe
Verify email addresses / create your own whitelist of verified addresses
Send Email
Request production access
Get Feedback
Limits on sandbox accounts: 100 whitelisted verified addresses, up to 10MB per message (because most popular email readers cannot handle more than that).
Limits on all accounts: Does not support SOAP or file attachments
Subscribe to:
Posts (Atom)