Skip to content

Document for Docker #2

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
MeghanaBandaru wants to merge 18 commits into
base: develop
Choose a base branch
from
Open

Document for Docker #2

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
18 commits
Select commit Hold shift + click to select a range
6401900
Add src folder
ashay-maheshwari Jun 16, 2016
dbe3957
Delete template files
Jun 28, 2016
9bd1ab5
Add files
Jun 28, 2016
b401517
Update file
ashay-maheshwari Jun 28, 2016
721b6f8
Update file
ashay-maheshwari Jun 28, 2016
0dad7b8
Update file
ashay-maheshwari Jun 28, 2016
98011fe
Add new cost study document
Jun 30, 2016
40927be
Add modified cost study document
ashay-maheshwari Jun 30, 2016
67af5a6
Update file
ashay-maheshwari Jun 30, 2016
6c50099
Update document
ashay-maheshwari Jun 30, 2016
732bcf2
Update aws-cost-study-new.org
ashay-maheshwari Jun 30, 2016
66c7a99
Update doc
Jun 30, 2016
6d7b0cf
Update document
ashay-maheshwari Jun 30, 2016
64dabed
Add document for Docker
Jul 13, 2016
12e8139
Delete old files
Jul 13, 2016
19ca0bb
Delete hour-cost-graph
ashay-maheshwari Jul 13, 2016
cea363b
SAR test memory utilization document
ashay-maheshwari Jul 14, 2016
652706f
added licensing file
vleadadmin Apr 23, 2018
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
192 changes: 192 additions & 0 deletions docs/aws-cost-study/aws-cost-study-new.org
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,192 @@
#+Title: Analysis of AWS cost for Virtual labs hosting
#+Author: Ashay Maheshwari
#+Date: June 28, 2016

* Introduction
+ This document represents the broken structure of AWS cost for the
month of May,2 016. Values obtained are estimated to be correct with
error fraction of + or - Rs 10,000.
+ All instances mentioned here are On-Demand instances and are charged on per-hour basis.
+ $1 = Rs70 (Approximated)
+ 1 month = 30 days

* Motivation
This document is to educate about our billings in AWS. It gives an idea of where are we spending and hence to explore docker to reduce
the cost by half. With Docker, we can have scalability with constant price.

* AWS instance cost
** AWS Rate chart (Then June 29, 2016)
+ Table given below show the AWS pricing for instance type used by VLEAD.
+ It represents the cost of an instance type, if up for 24 hours upto 30 days (24 hours * 30).

+ Table A.1
|-----------+-------------+-----------------------+-------------------------------+--------------------------------------|
| Type | No. of Core | Amount of RAM (in GB) | Rate in ($/hour/per-instance) | Cost for 30 days(in Rs per instance) |
|-----------+-------------+-----------------------+-------------------------------+--------------------------------------|
| t2.micro | 1 | 1 | 0.013 | 655.2 |
|-----------+-------------+-----------------------+-------------------------------+--------------------------------------|
| t2.small | 1 | 2 | 0.026 | 1310.4 |
|-----------+-------------+-----------------------+-------------------------------+--------------------------------------|
| t2.medium | 2 | 4 | 0.052 | 2620.80 |
|-----------+-------------+-----------------------+-------------------------------+--------------------------------------|
| t2.large | 2 | 8 | 0.136 | 6854.4 |
|-----------+-------------+-----------------------+-------------------------------+--------------------------------------|
| m3.large | 2 | 7.5 | 0.154 | 7761.6 |
|-----------+-------------+-----------------------+-------------------------------+--------------------------------------|

** Type and number of instances currently running(For the month of June)

+ Table A.2, given below represents the number of instances of specific type, which are currently running in VLEAD AWS architecture.

+ Table A.2
|-----------+------------------|
| Type | No. of Instances |
|-----------+------------------|
| t2.micro | 86 |
|-----------+------------------|
| t2.small | 13 |
|-----------+------------------|
| t2.medium | 1 |
|-----------+------------------|
| t2.large | 2 |
|-----------+------------------|
| | |
|-----------+------------------|
| TOTAL | 102 |
|-----------+------------------|

** AWS bill for instances(May, 2016)
+ AWS pricing for On-Demand instances is on the basis of $X/hour.
+ As soon as an instance is launched, it is charged for first hour. With the start of minute number 61, we are charged for second hour.
+ Table given below reflects the hour based run time of each type of instance, cost per month in $ and Rs.
+ Say that you launched 5 t2.micro instances and ran them of 5 hours each. Then you will be billed for 25 hours.

+ Table A.3
|------------------+-------------------------------------+----------------------+------------------------|
| Type of Instance | instance uptime(in hours per month) | Cost (in $per month) | Cost (in Rs per month) |
|------------------+-------------------------------------+----------------------+------------------------|
| t2.micro | 54,746 | 711.8 | 49,770.00 |
|------------------+-------------------------------------+----------------------+------------------------|
| t2.small | 7,341 | 191.03 | 13,372.10 |
|------------------+-------------------------------------+----------------------+------------------------|
| t2.medium | 704 | 37.33 | 2,613.10 |
|------------------+-------------------------------------+----------------------+------------------------|
| t2.large | 591 | 80.38 | 5,626.60 |
|------------------+-------------------------------------+----------------------+------------------------|
| m3.large | 746 | 150.12 | 10,508.40 |
|------------------+-------------------------------------+----------------------+------------------------|
| | | | |
|------------------+-------------------------------------+----------------------+------------------------|
| TOTAL COST | - | 1170.66 | 81,890.10 |
|------------------+-------------------------------------+----------------------+------------------------|


** Stopped instances
+ We have some stopped instances, for which costs are negligible (generally Storage cost)

+ Table A.4
|------------------+-----------+--------------+---------|
| No. of Instances | Type | Region | Remarks |
|------------------+-----------+--------------+---------|
| 1 | m3.large | N.California | - |
|------------------+-----------+--------------+---------|
| 1 | t2.medium | N.California | - |
|------------------+-----------+--------------+---------|
| 2 | m3.large | Sao. Paulo | - |
|------------------+-----------+--------------+---------|

* AWS Storage cost
** AWS Rate chart for Storage(then June 29, 2016)
Following table shows the pricing for type of Storage and Snapshot data used by VLEAD.

+ Table A.5
|-----------------+--------------------------|
| Storage Type | Rate (in $ per GB-month) |
|-----------------+--------------------------|
| SNAPSHOT DATA | 0.1299999911 |
|-----------------+--------------------------|
| SSD DATA | 0.1899999996 |
|-----------------+--------------------------|
| Magnetic Disk | 0.0499999997 |
|-----------------+--------------------------|

** AWS bill for Storage (For the month of May 2016)
+ Rate for every Storage type is different as mentioned in Table A.5
+ Charges are applied on the basis of per GB-month

+ Table A.6
|-----------+-----------------------------------+----------------------------------------+-----------------------------------+-----------------------+------------------------|
| Region | Snapshot data (GB used per month) | General Purpse SSD (GB used per month) | Magnetic Disk (GB used per month) | Cost (in $ per month) | Cost (in Rs per month) |
|-----------+-----------------------------------+----------------------------------------+-----------------------------------+-----------------------+------------------------|
| Sao Poulo | 54.040 | 417.639 | - | 86.38 | 6046.600 |
|-----------+-----------------------------------+----------------------------------------+-----------------------------------+-----------------------+------------------------|
| US East | 41.429 | 494.253 | 1,604.189 | 133.65 | 9355.500 |
|-----------+-----------------------------------+----------------------------------------+-----------------------------------+-----------------------+------------------------|
| US West | 42.490 | 91.042 | 226.389 | 34 | 2380 |
|-----------+-----------------------------------+----------------------------------------+-----------------------------------+-----------------------+------------------------|
| | | | | | |
|-----------+-----------------------------------+----------------------------------------+-----------------------------------+-----------------------+------------------------|
| TOTAL | - | - | - | 254.03 | 17,782.10 |
|-----------+-----------------------------------+----------------------------------------+-----------------------------------+-----------------------+------------------------|

* Graphs
Following graphs gives a visible statistics for values shown in tables.
+ Cost share between Storage and EC2 instance ::
This graphs shows that our major cost is from EC2 instances.
[[./storage-vs-instances.png]]


+ Number of instances of each type ::
This graphs gives the number of each type of instances currently running in our AWS architecture. Major amount of those instances
are t2.micro, which is the cheapest of all.
[[./no-of-instances.png]]


+ Number of hours of each instance we pay for ::
#+BEGIN_SRC formula
Total No. of hours an instance type was up * rate of instance type in $/hour
#+END_SRC
+ Graph given below share the number of hours an instance type is running for a month.
[[./no-of-hours.png]]

+ Graph given below share the hour-cost relationship of each instance type
[[./hour-cost-graph.png]]

+ Cost spent on each instance type ::
This graph show the share of cost spent on each instance type.
[[./cost-spent-share.png]]


* Estimated price for Docker
+ We can containerize our VM based AWS architecture using Docker and OpenVz.
+ Assuming that we take 32 GB of AWS instance ::
+ Table A.7
|---------------+------------------+------------------+------------------------+-------------------------|
| Instance type | CORE/RAM (in GB) | Rate (in $/hour) | Cost for 1 day (in Rs) | Cost for 30 days(in Rs) |
|---------------+------------------+------------------+------------------------+-------------------------|
| m4.2xlarge | 8/32 | $0.559 | 939.12 | 28,173.600 |
|---------------+------------------+------------------+------------------------+-------------------------|
| m4.xlarge | 4/16 | $0.279 | 468.72 | 14,061.600 |
|---------------+------------------+------------------+------------------------+-------------------------|

+ Expected cost for containerization ::
This can be the cost required for VM to run containers and Storage used. Probably Storage cost will remain
stable. For 32 GB VM used, monthly estimated cost -
#+BEGIN_SRC cost
$ Total Cost = Cost of VM for 30 days + Cost of Storage used
$ Total Cost = 28,173.600 + 17,781 = 45,954.60
#+END_SRC

+ Current Cost of AWS using VM = Rs.99680
+ Approxmiate cost with Docker = Rs.45,954.60

[[./current-vs-approximate-cost.png]]

* Conclusion
With Docker already explored, and cost benefits mentioned in the document are exciting, we should try to consider docker for production purposes and shift our current architecture to docker.






134 changes: 134 additions & 0 deletions docs/aws-cost-study/aws-cost-study.org
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,134 @@
#+Title: Billing study for AWS Architecture
#+Author: Ashay Maheshwari
#+Date: June 28, 2016

* Introduction
+ This document represents the broken structure of AWS cost for a month. Values obtained are estimated to be correct with error fraction of + or - Rs 10,000.
+ All instances mentioned here are On-Demand instances and are charged on per-hour basis.
+ $1 = Rs70 (Approximated)
+ 1 month = 30 days

* AWS Running Instances
|------------------+-----------+-----------------+--------------+-------------------------+-------------------+--------------+---------------+
| No. of Instances | Type | No. of Core/RAM | Rate($/hour) | Cost for 30 days(in Rs) | Total Cost(in Rs) | Region | Remarks |
|------------------+-----------+-----------------+--------------+-------------------------+-------------------+--------------+---------------+
| 86 | t2.micro | 1/1 | 0.013 | 655.2 | 56,347.200 | N.Virginia | - |
|------------------+-----------+-----------------+--------------+-------------------------+-------------------+--------------+---------------+
| 13 | t2.small | 1/2 | 0.026 | 1310.4 | 17,035.200 | N.Virginia | - |
|------------------+-----------+-----------------+--------------+-------------------------+-------------------+--------------+---------------+
| 1 | t2.medium | 2/4 | 0.052 | 2620.8 | 2,620.800 | N.Virginia | reverse proxy |
|------------------+-----------+-----------------+--------------+-------------------------+-------------------+--------------+---------------+
| 2 | t2.large | 2/8 | 0.136 | 6854.4 | 13708.8 | N.California | - |
|------------------+-----------+-----------------+--------------+-------------------------+-------------------+--------------+---------------+

** Inference

+ Total cost for running instances =Rs 89,710.2=
+ We have =101 running instance=
+ =62.81 %= cost is from t2.micro instances which is the cheapest of all VMs
+ We are paying for =132 GB RAM= ((86 * 1) + (13 * 2) + (1 * 4) + (2 * 8))

* Region wise Storage(EBS) Prices
+ Rate Snapshot data per GB-month (in $) = 0.1299999911
+ Rate SSD data per GB- month ( in $) = 0.1899999996
+ Magnetic Disk per GB-month (in $) = 0.0499999997

|-----------+---------------+--------------------+---------------+-----------+------------|
| Region | Snapshot data | General Purpse SSD | Magnetic Disk | Cost in $ | Cost in Rs |
|-----------+---------------+--------------------+---------------+-----------+------------|
| Sao Poulo | 54.040 | 417.639 | - | 86.38 | 6046.600 |
|-----------+---------------+--------------------+---------------+-----------+------------|
| US East | 41.429 | 494.253 | 1,604.189 | 133.65 | 9355.500 |
|-----------+---------------+--------------------+---------------+-----------+------------|
| US West | 42.490 | 91.042 | 226.389 | 34 | 2380 |
|-----------+---------------+--------------------+---------------+-----------+------------|

** Inference
+ Grand total for Storage disks =Rs 17,781=

* Instance type based division

|------------------+-------------------------------------------+-----------+------------|
| Type of Instance | Time for which instance were up(in hours) | Cost in $ | Cost in Rs |
|------------------+-------------------------------------------+-----------+------------|
| m3.large | 746 | 150.12 | 10,508.400 |
|------------------+-------------------------------------------+-----------+------------|
| t2.micro | 54,746 | 711.8 | 49,770.00 |
|------------------+-------------------------------------------+-----------+------------|
| t2.small | 7,341 | 191.03 | 13,372.100 |
|------------------+-------------------------------------------+-----------+------------|
| t2.medium | 704 | 37.33 | 2,613.1 |
|------------------+-------------------------------------------+-----------+------------|
| t2.large | 591 | 80.38 | 5,626.600 |
|------------------+-------------------------------------------+-----------+------------|

** Inference
+ Total Cost for Instances =Rs 81889=
+ Total Cost for Storage =Rs.17,781=
+ Total Bill paid for June =Rs.102596.200=



* Stopped instances

|------------------+-----------+--------------+---------|
| No. of Instances | Type | Region | Remarks |
|------------------+-----------+--------------+---------|
| 1 | m3.large | N.California | - |
|------------------+-----------+--------------+---------|
| 1 | t2.medium | N.California | - |
|------------------+-----------+--------------+---------|
| 2 | m3.large | Sao. Paulo | - |
|------------------+-----------+--------------+---------|


* Graphs
Following graphs gives a visible statistics for values shown in tables.
+ Cost share between Storage and EC2 instance ::
This graphs shows that our major cost is from EC2 instances.
[[./storage-vs-instances.png]]

+ Number of instances of each type ::
This graphs gives the number of each type of instances currently running in our AWS architecture. Major amount of those instances
are t2.micro, which is the cheapest of all.
[[./no-of-instances.png]]

+ Number of hours of each instance we pay for ::
#+BEGIN_SRC formula
Total No. of hours an instance type was up * rate of instance type in $/hour
#+END_SRC
+ Graph given below share the number of hours an instance type is running for a month.
[[./no-of-hours.png]]

+ Graph given below share the hour-cost relationship of each instance type
[[./hour-cost-graph.png]]

+ Cost spent on each instance type ::
This graph show the share of cost spent on each instance type.
[[./cost-spent-share.png]]

* Estimated price for Docker
+ We can containerize our VM based AWS architecture using Docker and OpenVz.
+ Assuming that we take 32 GB of AWS machine ::
|---------------+------------------+----------------+--------------------------+-------------------------|
| Instance type | CORE/RAM (in GB) | Rate in $/hour | Cost for 24 hours(in Rs) | Cost for 30 days(in Rs) |
|---------------+------------------+----------------+--------------------------+-------------------------|
| m4.2xlarge | 8/32 | $0.559 | 939.12 | 28,173.600 |
|---------------+------------------+----------------+--------------------------+-------------------------|
| m4.xlarge | 4/16 | $0.279 | 468.72 | 14,061.600 |
|---------------+------------------+----------------+--------------------------+-------------------------|

+ Expected cost for containerization ::
This can be the cost required for VM to run containers and Storage used. Probably Storage cost will remain
stable. For 32 GB VM used, monthly estimated cost -
#+BEGIN_SRC cost
$ Total Cost = Cost of VM for 30 days + Cost of Storage used
$ Total Cost = 28,173.600 + 17,781 = 45,954.60
#+END_SRC

+ Current Cost of AWS using VM = Rs.99680
+ Approxmiate cost with Docker = Rs.45,954.60

[[./current-vs-approximate-cost.png]]


Binary file added docs/aws-cost-study/cost-spent-share.png
View file
Open in desktop
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added docs/aws-cost-study/current-vs-approximate-cost.png
View file
Open in desktop
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added docs/aws-cost-study/hour-cost-graph.png
View file
Open in desktop
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added docs/aws-cost-study/no-of-hours.png
View file
Open in desktop
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added docs/aws-cost-study/no-of-instances.png
View file
Open in desktop
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added docs/aws-cost-study/storage-vs-instances.png
View file
Open in desktop
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading